Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The influence of red alder in adjacent conifer stands : nutrient cycling and light transmission Lavery, John Meredith 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2001-0055.pdf [ 8.74MB ]
Metadata
JSON: 831-1.0099531.json
JSON-LD: 831-1.0099531-ld.json
RDF/XML (Pretty): 831-1.0099531-rdf.xml
RDF/JSON: 831-1.0099531-rdf.json
Turtle: 831-1.0099531-turtle.txt
N-Triples: 831-1.0099531-rdf-ntriples.txt
Original Record: 831-1.0099531-source.json
Full Text
831-1.0099531-fulltext.txt
Citation
831-1.0099531.ris

Full Text

THE INFLUENCE OF RED A L D E R IN ADJACENT CONIFER STANDS; NUTRIENT C Y C L I N G A N D LIGHT TRANSMISSION by JOHN MEREDITH L A V E R Y B.Sc. Hon. Bishop's University, 1997 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE In The Faculty of Graduate Studies (. Faculty:,  of Forestry)  We accept this thesis as conforming to the required standard  University of British Columbia Vancouver, Canada o  2000  © John Meredith Lavery, 2000  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  Department of The University of British Columbia Vancouver, Canada  Abstract Red alder (Alnus rubra(Bong.)) is an important successional deciduous species in British Columbia with expanding commercial markets. Alder is a pioneer species which grows rapidly on disturbed sites and in riparian areas of coastal British Columbia. Alder's impressive early height growth and ability to increase the rate of cycling of nitrogen and other nutrients lead to a challenge in managing alder with conifers. Managing alder in mixedwood stands requires finding an appropriate balance between competition for light and the improvement in N status of a site, which would lead to improved conifer growth over time. Light transmission to the understory of adjacent alder and conifer stands was measured along a transect between the two stands using photodiode and quantum sensors, as well as hemispherical photography. Light varied with age and site, but alder stands generally had more light in the understory than the adjacent conifer stands. This difference was more marked in younger stands. Modeling using SLIM and LITE correlated well with the sensor information, and further modelling of the alder stands suggests that the amount of light available to conifers adjacent to an alder stand is of adequate quantity to sustain conifer growth within 5m of the alder/conifer stand boundary. The assessment of nutrient cycling in alder stands involved independent studies of litterfall, soil qualities, and a bioassay of the soils using Douglas-fir seedlings. Higher concentrations of N and several other nutrients were found in alder litter, but the conifer litter had very high levels of several micronutrients. The alder litterfall made significant contributions to nutrient inputs up to 15m into the conifer stands. The soil showed elevated N , and Total C, C:N and pH were influenced by alder presence in the oldest stands. This series of experiments also showed the influence of alder on the concentration of N in the shoots of the seedlings, as well as an influence on boron. The influence of alder litterfall, soil and bioassay N at points along the transects was correlated at the oldest and middle aged sites, suggesting that red alder produces a significant change in the nutrient cycling in adjacent conifer stands. Implications for managers and suggestions for management are discussed.  n  Table of Contents  Abstract  ii  Table of Contents  iii  List of Tables  vi  List of Figures  xi  Acknowledgments  xiii  CHAPTER I: Introduction, Objectives, Hypotheses, Sites  1  Introduction  1  Principal Objectives  11  Site Characterization  12  CHAPTER II: Light Transmission Through Mixed Canopies  19  Introduction  19  Methods  22  Results  26  Discussion  36  CHAPTER III: The Roles of Leaf Litterfall, Macrofaunal and Microbial Communities in Propagating the Alder Influence  39  Introduction  39  Methods  42  Results  44  Discussion  68  CHAPTER IV: The Influence of Red Alder on Soils, Part 1: Soil Chemical Analysis  73  Introduction  73  Methods  75  Results  78  Discussion  114  iii  CHAPTER V : The Influence of Red Alder on Soils, Part 2: Ex situ Douglas-fir pot trial  123  INTRODUCTION  123  METHODS  124  RESULTS  125  DISCUSSION  154  CHAPTER VI: Discussion of the Management of Light and Nutrient Cycling in Alder/Conifer Stands, Conclusions and Recommendations for Further Research  158  Literature Cited  162  Appendix 1  171  Appendix II  173  Appendix III  175  Appendix IV  178  Appendix V  181  iv  List of Figures  Chapter I  Fig. 1.1  Nested map of B C and the Lower Mainland/ Vancouver Island Region, Showing the approximate locations of all research sites.  13  Fig. 1.2  Elevation rise and fall along transects  14  Fig. 1.3  Diagram of the structure of the major transects used in this study.  15  Chapter II  Fig. 2.1  Relationships between SLIM interpretations and Photodiode measurements for P A C L in stands studied.  Fig. 2.2  28  Relationships between spatially explicit LITE simulations and Photodiode measurements for P A C L in stands studied.  Fig. 2.3  Percent above canopy light (PACL) determined using SLIM fisheye photo analysis for the month of July, 1998.  Fig. 2.4  29  31  Percent above canopy light (PACL) determined using LITE spatially explicit stand and fisheye simulation analysis for the months of June through September, 1998.  Fig. 2.5  32  Percent above canopy light (PACL) determined using LITE spatially explicit stand and fisheye simulation analysis with all conifers removed from each stand for the month of July, 1998.  Fig. 2.6  34  Percent above canopy light (PACL) determined using LITE spatially explicit stand and fisheye simulation analysis with all conifers removed from each stand for the month of June, 1998, rotating each transect through the four cardinal directions.  35  v  Chapter III  Fig. 3.1  Amount of leaf litterfall (kg ha" yr" ) in alder stands of various ages. Data 1  1  points A though N represent data collected in this study Fig. 3.2  The relationship between the age of stands, and alder litter migration from the alder stand.  Fig. 3.3  46  The relationships between alder litterfall input, distance from the alder stand, and the age of the stand for points 1 through 7.  Fig. 3.4  61  Millipede counts for all transects. Transects MK-1, MK-2,NM-1, and N M 2 show significant differences between points (Table 2.3).  Fig. 3.14  60  Mass of annual input of total Zn through leaf litterfall at points along the transects between adjacent red alder and conifer stands  Fig. 3.13  59  Mass of annual input of total Cu through leaf litterfall at points along the transects between adjacent red alder and conifer stands.  Fig. 3.12  57  Mass of annual input of total Fe through leaf litterfall at points along the transects between adjacent red alder and conifer stands.  Fig. 3.11  56  Mass of annual input of total B through leaf litterfall at points along the transects between adjacent red alder and conifer stands.  Fig. 3.10  55  Mass of annual input of total M g through leaf litterfall at points along the transects between adjacent red alder and conifer stands.  Fig. 3.9  53  Mass of annual input of total P through leaf litterfall at points along the transects between adjacent red alder and conifer stands.  Fig. 3.8  52  Mass of annual input of total N through leaf litterfall at points along the transects between adjacent red alder and conifer stands  Fig. 3.7  49  Mass of annual input of total leaf litterfall at points along the transects between adjacent red alder and conifer stands.  Fig. 3.6  47  The relationships between alder litterfall input, distance from the alder stand, and the age of the stand for points 8 and 9.  Fig. 3.5  45  65  Microbial respiration in forest floor at each point along the transects. There were no significant (a=.05) differences or trends at any of the transects  vi  67  Chapter IV  Fig. 4.1  Concentrations (%) of total N measured at 0-10 cm depth for each transect.  Fig. 4.2  Concentrations (ppm) of inorganic nitrogen (mg*kg"') measured at 0-10 cm depth for each transect.  Fig. 4.3  86  Concentrations (ppm) of soil nitrate (mg*kg ) measured at 0-10 cm depth _1  for each transect. Fig. 4.4  87  Concentrations (ppm) of soil ammonium (mg*kg"') measured at 0-10 cm depth for each transect.  Fig. 4.5  88  Concentrations (ppm) of mineralizable nitrogen (mg*kg ) measured at 0-10 _1  cm depth for each transect. Fig. 4.6  83  92  Ratios of N H - N : NO3-N from soils measured at 0-10 cm depth for each 4  transect.  94  Fig. 4.7  Carbon concentrations (%) measured at 0-10 cm depth for each transect.  97  Fig. 4.8  C:N ratio (dry weight equivalents for total C and total N) measured at 0-10 cm depth for each transect.  Fig. 4.9  98  Concentrations (ppm) of available P (mg*kg ) measured at 0-10 cm depth _1  for each transect.  101  Fig. 4.10  pH in water paste for soil from 0-10 cm depth for each transect.  104  Fig. 4.11  pH in CaCl for soil from 0-10 cm depth for each transect.  105  Fig. 4.12  The relationship between carbon concentration in the upper 10 cm of  2  mineral soil and annual input of alder litter at the Nanaimo Lakes site. Fig. 4.13  The relationship between nitrate concentrations in the upper 10 cm of mineral soil and annual input of alder litter at the Nanaimo Lakes site.  Fig. 4.14  Ill  The linear relationship between nitrate in the upper 10 cm of mineral soil and annual input of alder litter at transect MK2.  Fig. 4.15  110  112  The relationship between available P concentration in the upper 10 cm of mineral soil and annual input of alder litter at transect MK2.  vn  113  Chapter V  Fig. 5.1  Nitrogen concentration (%) of seedlings grown in soils harvested from 6 transects between alder and conifer stands.  Fig. 5.2  Phosphorus concentration (%) of seedlings grown in soils harvested from 6 transects between alder and conifer stands  Fig. 5.3  _1  142  Boron concentration (mg*kg ) in seedlings grown in soils harvested from 6 _1  transects between alder and conifer stands. Fig. 5.12  141  Zinc concentration (mg*kg ) in seedlings grown in soils harvested from 6 transects between alder and conifer stands.  Fig. 5.11  140  Copper concentration (mg*kg"') in seedlings grown in soils harvested from 6 transects between alder and conifer stands  Fig. 5.10  139  Manganese concentration (mg*kg~) in seedlings grown in soils harvested from 6 transects between alder and conifer stands.  Fig. 5.9  137  Iron concentration (mg*kg"') in seedlings grown in soils harvested from 6 transects between alder and conifer stands  Fig. 5.8  136  Sulphur concentration (%) in seedlings grown in soils harvested from 6 transects between alder and conifer stands  Fig. 5.7  134  Magnesium concentration (%) in seedlings grown in soils harvested from 6 transects between alder and conifer stands.  Fig. 5.6  133  Calcium concentration (%) in seedlings grown in soils harvested from 6 transects between alder and conifer stands.  Fig. 5.5  131  Potassium concentration (%) in seedlings grown in soils harvested from 6 transects between alder and conifer stands  Fig. 5.4  129  143  The linear relationships between N concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and Total N in the soils from transects NM1 and NM2. Lines represent significant linear relationships  Fig. 5.13  148  The linear relationship between N concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and Nitrate in the 0-  vm  149  10 depth of soil for NM1, NM2 and CH2. Lines represent significant linear relationships. Fig. 5.14  The linear relationship between B concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and Nitrate in the 010 cm depth of soil for NM1 and CH2. Lines represent significant linear relationships.  Fig. 5.15  150  The linear relationship between P concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and the available P (Bray PI) for transect NM1. The line represent significant linear relationships.  Fig. 5.16  151  The linear relationship between N concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and the annual input of alder litter for all transects. Lines represent significant linear relationships.  Fig. 5.17  152  The relationship between B concentration in shoots of Douglas-fir seedlings and annual input alder litter for all sites. Lines indicate significant relationships at individual transects  ix  153  L i s t of Tables  Chapter I Table 1.1  Summary of Climate data collected from representative Biogeoclimatic reference stations (adapted from Meidinger and Pojar 1991).  16  Chapter II Table 2.1  Regression statistics describing the relationships between SLIM and LITE simulations and longterm in situ photodiodes. All values are based on July 1998 data, except where data problems mandated using another month, in which cases June 1998 was used.  30  Chapter III Table 3.1  Analysis of Variance (ANOVA) for total litterfall at each point along each transect, and N , P, Mg, B, Fe, Cu, and Zn (kg ha" ) through leaf litterfall. 1  Table 3.2  50  Foliar litter nutrient concentrations for representative alder and conifer litter from each site.  54  Table 3.3  Pitfall trap count totals for each morphotype classified.  63  Table3.4  A N O V A results for examination of macro fauna proportion at all sites, at individual sites, and between sites.  64  Chapter IV Table 4.1  Summary of significant quadratic and linear regression R-squared and pvalues for all variables measured in the soil along all transects.  80  Chapter V Table 5.1  Regression R-squared and F-statistic for seedling nutrient concentrations found to show a significant relationship with position along the transect. Tukey's Honestly Significant Differences test shows relationships.  Table 5.2  126  Standard Ranges of adequacy for mineral nutrient concentrations in conifer shoots for bare root stock, container stock (adapted from Timmer 1991) and initial seedlings from nursery stock.  Table 5.3  127  Summary of univariate macronutrient deficiency diagnoses by nutrient and site. Nutrient deficiency tolerance scale was developed using Ballard and  x  145  Carter (1986) foliar nutrient values, and correcting for the inclusion of the stem nutrients. Table 5.4  Summary of univariate micronutrient deficiency diagnoses by nutrient and site. Nutrient deficiency tolerance scale was developed using Ballard and Carter (1986).  Table 5.5  145  Nutrient concentration means by site from seedlings grown in soils originating from the sample points.  xi  146  Acknowledgements  This thesis could not have been completed to the current standard without the efforts of many. I would like to thank my supervisors, Dr. Phil Comeau and Dr. Cindy Prescott, for excellent guidance, editing and encouragement. Timberwest's Eric Jeckland allowed me to use an ideal alder stand on their private land near Nanaimo. I would like to acknowledge all those who have helped me in the field. Those to whom I am very grateful for their time spent in the field or in the lab include Ben Andrew, Christina Cockle, David Gentleman, Sarah Harper, Jennifer Jones, Susan Leech, Angela Plautz, and Holger Wernsdorfer. In addition, I am indebted to Clive Dawson and David Dunn at the B.C. Ministry of Forests' Glyn Road Analytical Lab for their efforts with my samples, and to Peter Fielder, from the B.C. M.O.F. Research Branch, for his efforts in calibration of the photodiodes and quantum sensors. Students and faculty in Forest sciences at U B C and the Prescott lab group in particular have been instrumental in helping me to develop some of my ideas about nutrient cycling. I would like to thank Dr. Tim Ballard, Jennifer Bennett, David Blevins, Leandra Blevins, Kirsten Hannam, Dr. John McLean and Lisa Zabek for their discussion and ideas contributing to this thesis. A special thank-you is merited for Emily Pritchard, for her ideas, editing, and support.  xii  CHAPTER I Introduction, Principal Objective, Site characterization Introduction Red alder (Alnus rubra (Bong.)) is a prominent deciduous species on the Pacific coast of North America (Bormann and DeBell 1981). As a fast growing pioneer species it quickly occupies disturbed forest soils or newly formed river seres. On newly formed seres, alder is often the first invader, and its invasion is later followed by physiognomic changes and shifts in vegetation as the soil develops and becomes richer (Luken and Fonda 1983). The speed with which alder is able to grow on such sites allows it to overtop conifer species that might also take over disturbed lands, such as western redcedar (Thuja plicata Don. Ex D. Don), western hemlock (Tsuga heterophylla [Raf] Sarg.), Sitka spruce (Picea sitchensis [Bong.]Carr.), and Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco.). This rapid juvenile height growth threatens to impair the growth of nearby conifers (Cole and Newton 1987). Red alder has an actinorrhizal symbiont, Frankia, which allows it to fix atmospheric nitrogen (N). Much of this fixed N becomes available within the stand as it is incorporated in the large volume of autumnal alder litterfall. The excess N , coupled with macronutrient litterfall inputs on the order of two to five times greater than conifer litter inputs, make red alder litter a natural fertilizer. This is potentially ideal for improving the productivity of poor sites. A "mixedwood" stand is simply a stand with a mixture of different species. The broadleaf-conifer mixture (e.g. a red alder-Douglas-fir mixture) is a subset of this general type of forest. Mixedwoqds are a common natural occurrence and the species involved seem to avoid competition through differential shade tolerance, physical separation of canopies, phenological differences, separation of their successional roles, and key differences in soil resource utilization (Man and Lieffers 1999). There are several benefits to managing for mixedwoods in British Columbia and the Pacific Northwest. 1)  Broadleaves occur naturally in mixtures with conifers;  2)  They provide a valuable visual resource;  3)  Mixedwood stands support a high level of diversity; 1  4)  Mixedwood stands suffer reduced impacts from insect and disease problems, reducing the risk of loss from such pests;  5)  Broadleaves can act as a valuable nurse crop for conifer juveniles;  6)  Incorporation of a broadleaf component can improve nutrient availability in mixedwood stands;  7)  Mixedwood stands can provide greater biomass yield than pure stands;  8)  Mixedwood management has the potential to be more readily sustainable than single species management;  9)  Growing mixtures may give better economic returns than pure stands; and  10)  Management of mixtures is mandated where mixtures currently exist in British Columbia. (Comeau 1996)  Mixedwood management is more complex than single species management. The manager must have good knowledge of the site and of the ecology and silvics of the species involved. Knowledge of the species in the mixture is fundamental to the success of mixedwood management (Comeau 1996). There are a wide variety of mixture possibilities to consider in mixedwood management. Intimate mixtures are those in which two (or more) species are mixed in a heterogeneous, evenly spaced fashion, displaying no patchiness or clumping of either species. It is thought that the successful growth of an intimate mixture of red alder and Douglasfir requires keeping the red alder at very low densities (between 50-200 evenly spaced alder per ha.) (Comeau and Sachs 1992). Even intimate mixtures with a shade tolerant conifer, such as western redcedar, can have no more than about 400 evenly spaced alder per ha. One of the complexities of alder mixtures lies herein. At spacings of 50-400 alder stems per ha, alder spreads out laterally and develops a bushy morphology that results in poor wood quality. This poor growth structure degrades stand quality in two ways: the bushy morphology leads to knotty, unusable alder bolewood; and the long, wide branches have the potential to both limit height growth of the associated conifers and damage the conifer leaders through wind-generated whipping episodes. Alternatively, spacing of 3.0 - 3.5 m (800-1100 stems/ha) appears to maximize clear bole production (Hibbs and DeBell 1994). But planting at higher densities 2  requires thinning at an early stage to promote stem wood production (Peterson et al. 1996). At these densities the growth rates and competitive effect of red alder would quickly extinguish any conifers in the stand. This leads to severe growth inhibition and eventual extinction of the conifer component of the plot. Growing two species in intimate mixture, although predicted to be more sustainable than single species planting (Comeau and Sachs 1992) and a factor in the enhancement of non-timber resource values, does not always produce the economic benefits that make mixtures attractive. Growing alder at low densities required for reasonable growth of Douglas-fir will result in alder which are of poor wood quality (Peterson et al. 1996). Other methods of mixing the species in a stand, such as growing small, dense, localized patches of alder, strips of alder, and large mosaics across the landscape could give the benefits of diversity, coupled with the nutritional benefits and added biomass of the secondary species. For example, patchwork mixtures of adequate densities offer maximum commercial value and areas of alder influence extending into the conifer stands. This leads to stand borders with high structural and ecological diversity, and a beneficial influence of the alder nutrient regime with limited border competition areas. Patches and row mixtures have been investigated (Hibbs and DeBell 1994), but little data has been presented on the ability of these to successfully balance the conflicting variables. Availability of a threshold amount of light is critical for the survival of any plant (Messier 1996). Tree growth has been tightly linked with the quantity of light they receive, and knowledge about the effects of light variation on seedling growth is fundamental to the understanding of forest dynamics (Wright et al. 1998). Knowing the variables involved when light is transmitted from the top of the forest canopy to the ground can allow modification of the light regime in a forest to achieve maximal growth of crop species. The characterization of light availability under a plant canopy is complex. Complexity in modeling a forest canopy arises because of the spatial and temporal heterogeneity of light transmission (Parent and Messier 1996). Vegetation cover and density reduce the quantity of light that reaches a point in the understory. In addition, seasonal changes in the tree canopy, particularly in flushing and senescing deciduous stands, have profound influences on understory light. Light levels at any particular time also depend on solar positioning and sky conditions. Overcast conditions differ from clear conditions in their light dispersion characteristics. 3  Trees vary in their response to light. Shade intolerant species have morphological and ecophysiological traits that support high carbon gain under high light conditions. Shade tolerant and late-successional species are able to allow positive leaf carbon gain at much lower light levels, but are unable to maximize gains under high light conditions (Bazzaz 1979; Ogden and Schmidt 1997; Wright et al. 1998). Douglas-fir is categorized as a shade intolerant conifer, and mortality will occur at relative light intensities of 20% (Mailly and Kimmins 1997). Mixed stands of aspen and spruce in the boreal forest can intercept 87% of above canopy light (Constabel and Lieffers 1996). This type of light regime under the mixed stand is prohibitive to conifer growth in the understory. Even shade tolerant trees would have difficulty competing with understory vegetation at relative light intensities below 13%. Diurnal patterns of radiation within forests follow the variation of incident sunlight, but become increasingly variable with canopy depth. Variability of penetration through the canopy arises due to changes through space and time because the canopy is dynamic (Hutchison and Matt 1977). Mixed canopies exhibit higher light transmission capabilities in the spring, prior to leaf flushing, and in the autumn, during leaf-fall (Constabel and Lieffers 1996). These increases can be marginal though, due to the lower solar elevations and shorter light-days than the summer months. Solar radiation within a deciduous forest follows a predictable pattern relative to changes in solar zenith through the growing season, leaf flushing and abscission (Hutchison and Matt 1977). Radiation in the forest increases from a winter minimum to a maximum in early spring, prior to leaf flush. Transmission drops sharply following bud burst, and in the period following summer solstice, light transmission through the canopy slowly drops back to its annual minimum. This is caused by the presence of a canopy and the gradual decrease of light due to the lapse of the zenith angle. As leaves senesce in autumn, radiation increases, but then drops as the solar elevation reaches its annual minimum at winter solstice (Hutchison and Matt 1977). Apart from changes in the light environment due to seasonality, understory light varies depending on space, time, and the stage of forest development in temperate forests (Messier and Bellefleur 1988). Red alder has a unique effect on the light characteristics of a stand. In fifth year examinations of alder/Douglas-fir plots, growth of the conifer was depressed at all mixed sites, compared with pure Douglas-fir controls (Cole and Newton 1987). Under the alder canopy, 4  Douglas-fir trees were shorter than alder and less light was available for photosynthesis at any crown level. The Douglas-fir trees responded to the removal of alder by changing leaf morphology; increasing the leaf area: leaf weight ratio, which resulted in a greater proportion of shade needles (Cole and Newton 1986). These results indicate that it is important to understand the light regime and how it varies between alder and conifer stands. The extent to which alder will affect the light regime will indicate safe planting distances from its boundaries for conifer species. Alder compete with the Douglas-fir for both moisture and light (Shainsky and Radosovich 1992; Shainsky et al. 1994). The competitive effects exerted by alder on light and soil moisture are opposed by the increased availability of limiting nutrients in and near an alder stand. Alder is able to modify soil conditions more profoundly than non N -fixing species. 2  Alder litter provides essential nutrients required by conifers through soil amendment, even though alder retranslocates nutrients prior to leaf abscission (Radwan et al. 1984). In one study, the average nutrient returns in alder litter (in kg ha" yr') were: N , 82; Ca, 41; K , 19; Mg, 8; S, 7; 1  P, 4; Fe, 1; Mn, 1; A l , 1; Zn, 0.2; and Cu, <0.1 (trace) (Radwan et al. 1984). The nitrogen values correlate well with calculated N accretion values (Van Miegroet et al. 1989);(Binkley et al. 1992a; Beck and Elsenbeer 1999) indicating that most of the N in the litter is being cycled back into the soil. Because of the correlation of these values, it is supposed that nutrient return via leaf litterfall is one of the main mechanisms by which fixed N is transferred to the soils in an alder stand (Bormann et al. 1994). Large quantities of nutrient-rich litterfall are considered to be the primary mechanism by which alder influences the nutrition of adjacent conifers and conifers growing in mixture with alder (Zavitkovski and Newton 1971; Gessel and Turner 1974). Litterfall cycling of N is generally three to eight times greater in mixed red alder stands than in pure conifer stands (Binkley et al. 1992a). Concurrent cycling of P, S, Ca, M g and K is often just as significant, with two to ten times more nutrients present in mixedwood litter than in a pure conifer stand (Radwan et al. 1984). Alder litterfall aids in the acceleration of soil development for longer lived species (Bormann and Sidle 1990). Nitrogen may be the most common nutrient limiting agricultural and forest productivity, but of all nutrients, N limitation is the easiest to resolve through manipulation of N-fixing species (Crews 1993). Nitrogen-fixing trees provide a potentially useful silvicultural tool for managing 5  forest stands on N-limited sites, especially when concerns regarding biodiversity or the availability of inorganic-N fertilizer make fertilization less attractive. Past studies of red alder have focused on soil improvement through the ability of alder to fix atmospheric N . Red alder's N fixation system is not as well regulated as the fixation systems in leguminous plants, and alders continue expending energy on N fixation rather than making greater use of abundantly available soil N (Binkley 1992). Light availability, soil moisture, temperature and P availability control fixation of N. Surprisingly though, N availability and stand density do not appear to affect fixation rates (Binkley et al. 1994). Nitrogen fixation by alder has been intensively studied since it is the most obvious mechanism by which alder influences the growth of adjacent conifers in a stand. Nitrogen fixation rates vary between 20 and 85 kg ha" yr" (Binkley et al. 1994), but rates up to 320 kg ha" 1  1  1  yr"' have been reported (Van Miegroet et al. 1989). However, nitrogen fixation rates do not necessarily imply a benefit to site productivity. Nitrogen accretion on a site is favourable only if the site is nutrient-poor. On a nutrient deficient site in the Wind River Experimental Forest in southwestern Washington, consistently higher N values were observed under the alder stand (4340 kg ha" vs. 1960 kg ha" in the conifer stand). The mixedwood stand also had more biomass 1  1  (10.3 M g ha" in the mixed stand compared to 4.8 M g ha" in the pure conifer stand), and larger 1  1  conifers than in the adjacent conifer stand (Binkley et al. 1992a). But on a nutrient-rich site, alder seems to inhibit the full growth potential of the conifers (Binkley et al. 1992a). One study of very young red alder showed that it did not increase the soil N or foliar N in adjacent Douglasfir foliage (Cole and Newton 1986). Nitrogen is the source of positive interactions between alder and conifers, resulting in a reduction of competition between the two species through resource partitioning, since the alder is able to access a different N source. This leads to facilitation of Douglas-fir growth through N accretion and increased nutrient cycling in the mixedwood stand (Shainsky and Radosovich 1992). Nitrogen fixation does not seem to immediately affect the soil, and it is unclear if N fixed in the nodules is able to exude from the root system of the alder. There is no information on the contribution of root litter to soil N supply. The primary benefit from alder is the influence of the annual litterfall and its decomposition, resulting in a release of N and other nutrients. The juxtaposition of simultaneous facilitation and competition between the two species is an 6  excellent example of the need to consider positive and negative effects of plants concurrently (Holmgren et al. 2000). Life stages, physiology, direct and indirect interactions, and the physical environment all play significant roles in determining the balance of competition and facilitation in plant communities (Callaway and Walker 1997). This mandates a broad spectrum of study if competition is to be fully assessed. The influence of alder is not the same in every location. Red alder improves the fertility of poor sites, and can improve the growth of conifers when they are grown in stands mixed with alder (Miller and Murray 1978). This is attributed to N fertilization and accelerated nutrient cycling (Cole et al. 1978), leading to increased nutrient availability. Alder occurring on nutrient rich sites may not have the same beneficial effect on adjacent stands as alder on poor sites (Binkley et al. 1992a). This effect is site specific and must be assessed for a variety of conditions. Nitrogen fixation by red alder increases soil N by 20-50% over periods of 30 years or more, but annual N turnover rates in alder soils can reach eight times those of neighbouring conifer stands (Binkley 1992; Binkley et al. 1992a). Alder litter has been shown to have higher N and lower lignin concentrations than most conifer litter except Douglas-fir (Edmonds 1980), which can lead to soil organic matter of high substrate quality (Bormann et al. 1994). Red alder also increases soil organic matter (Bormann and DeBell 1981). Conifer seedlings planted on • soils previously occupied by alder show higher N status than conifer seedlings planted on adjacent sites previously occupied by conifers (Brozek 1990). Douglas-fir seedlings planted on soils where alder has been grown in the previous rotation show higher N/P and N/K ratios, as well as higher biomass than seedlings planted on similar soils where conifers had been harvested (Brozek 1990). Soils that are found under alder and in the alder litter shadow typically show elevated soil N content (Binkley et al. 1992a). The litter shadow is that area which accumulates litterfall on the ground. Red alder exerts an influence on soils beyond alder stand boundaries (Rhoades and Binkley 1992; Miller and Reukema 1993; Lieffers et al. 1999), but studies have differed as to how far the effects of alder on soils extend, and what mediates them. As early as 1978, it was noted that Douglas-fir trees within 10 m of an adjacent mixed alder-conifer stand were larger than those beyond that distance (Miller and Murray 1978). Douglas-fir trees had greater bole volume when interplanted with red alder or when within 15 m of a mixed stand (Miller and 7  Reukema 1993). At an infertile site at Wind River, Washington, the influence of alder on the soil chemistry and N-availability of adjacent conifer stands was insignificant upslope, but apparent between 8 -12 m downslope from the alder (Rhoades and Binkley 1992). Douglas-fir in pure stands that were downslope from a mixed alder-conifer stand were 17% taller than pure Douglasfir on the upslope side of the mixed stand (Miller and Reukema 1993). The elevated N H  + 4  levels  in the mixed stand declined sharply on the up-slope side and more gradually on the down-slope side (Rhoades and Binkley 1992). These studies reveal that patterns of alder influence on adjacent conifer stands are strongly dependent on slope. The relationship between slope and the extent of increased availability of N H  + 4  is consistent with the mechanisms of subsurface N flow  and litterfall (Rhoades and Binkley 1992). Increases in mineralizable N are common in alder stands, and when soils are well aerated leads to increased accumulation of N0 " (Hart et al. 3  1997). Waterlogged soils tend to display an increase in N H - N but not N 0 - N . 4  3  Phosphorus is also considered to be a limiting nutrient in coastal soils, and is usually surpassed only by N in paucity. The supply of P is thought to limit growth of mature red alder and lead to its decline and deterioration (Radwan et al. 1984). Alder cycles large amounts of available P, and maintains much of the P in its woody tissues (Compton and Cole 1998). Accelerated soil P depletion has been attributed to red alder (Brozek 1990). Phosphorus is required for N fixation, and so the productivity of red alder stands may reflect the availability of P. It has been hypothesized that understory plants competing for P with red alder can reduce the productivity of these stands (D. Binkley, pers. comm.). Alder does not become established in soils with low P. The P requirement in alder is very high (Binkley et al. 1994), and P fertilization improves the growth of red alder on both well drained and waterlogged sites (Radwan and DeBell 1994). The sustainability of a forest or agricultural system hinges on the extent to which it can rely on, but not deplete, endogenous nutrient supplies (Crews 1993). Questions about the impacts of forest management methods on P availability remain largely unanswered (Giardina et al. 1995), and the ability to rectify management-induced P deficiencies is poorly understood. Available P, measured using the Bray PI method has been observed to be very low under alder stands and was calculated to be 10% that of adjacent conifer stands (Compton and Cole 1998). But in the same alder stands, P cycling was up to 292% greater than the rate of cycling in 8  adjacent conifer stands (Compton and Cole 1998). Alder appears to increase available P under mixed stands (Giardina et al. 1995; Zou et al. 1995; Heneghan and Bolger 1998), but in pure stands, available P seems to decrease (Compton et al. 1997). In the long term, N -fixers cause a 2  reduction in most macronutrients, causing growth and foliar P concentrations to decline with age (Radwan and DeBell 1994; Giardina et al. 1995; Binkley 1997; Compton et al. 1997; Binkley et al. 2000). Depressed soil pH is another common trait of alder stands. The magnitude of the depression varies among stands (Binkley and Sollins 1990), ranging from almost 0 (Binkley 1983), to 1/50* of a pH unit per year (Bormann and DeBell 1981). Five characteristics of forested soils are thought to be the principal factors in determining soil pH (Binkley et al. 1992a): 1)  The ionic concentration of the soil;  2)  The degree of dissociation of the acids, referred to as base saturation;  3)  The quantity of acids present in a soil, referred to as cation exchange capacity (CEC);  4)  Acid strength, or the affinity of the acids for hydrogen ion (¥t)(pK );  5)  The redox potential of the soil (Beck and Elsenbeer 1999).  and  a  (Adapted from (Binkley and Sollins 1990; Binkley et al. 1992a)  Soil pH under a mixed stand was thought to be depressed by the greater ionic strength of the soil solution resulting from excessive concentrations of N0 " in the alder-conifer soils 3  (Binkley and Sollins 1990). The influence of pH in decreasing the concentrations of Ca, Mg, and available P in soil has also been shown in alder stands (Compton and Cole 1989). Of the five principal mechanisms that affect soil pH (Binkley et al. 1992a), one of the most important is the ionic concentration of the soil. Soil pH in deionized water has shown a difference in pH between mixed (pH 5.1) and pure stands (pH 5.4) while pH in CaCl of the same 2  soils showed no difference. The soluble anions in the mixed stand soil were considerably higher than in the pure conifer stand (330 umol/1 to 175 umol/1), a difference mainly due to the large amount of N0 " in the mixed stand soil. Thus the ionic concentration can be quite different, 3  though CaCl pH does not reflect this discrepancy (Binkley and Sollins 1990; Binkley 1992). 2  9  The second mechanism affecting pH is the degree of dissociation of acids within the soil. Since organic matter and clay particles in the soil act as stabilized weak acids, base saturation leads to the dissociation of acidic H and A l . Nitrogen-fixing trees such as alder produce N0 " +  3 +  3  that causes the removal of base cations from the soil through leaching. A lower base saturation means a lower degree of dissociation of acids, and lower soil pH (Binkley 1983). The cation exchange capacity of the soil is another factor that influences soil pH. The inclusion of N-fixing trees in an ecosystem often stimulates production and may lead to an increase in soil organic matter (Binkley 1983). Soil organic matter containing higher levels of organic acids could lower pH by dissociation leading to subsequent leaching of cations. Van Miegroet and Cole (Van Miegroet et al. 1989) found that undissociated weak acids stored in the soil were 40% greater under an alder stand compared to an adjacent conifer stand. Base saturation was lower under the alder stand, not due to leaching as in the second mechanism, but to the increase in size of the exchange complex (Binkley 1992). The relative strengths of the acids in the soil also affect pH. Acids which are stronger cause lower pH values in the soil solution than weaker acids when present in the same quantity. Alder stands have been shown to produce stronger acids, lowering pH with the same amount of acid as a neighbouring conifer stand (Binkley and Sollins 1990; Binkley et al. 1992a). The fifth mechanism is the redox potential (pe) of the soil. Redox, or reduction and oxidation reactions, entail the flows of electrons from low-energy and high-energy states, respectively (Fisher and Binkley 1999). Because the equation pe + pH is generally constant for a given soil, changes in pe will have a resultant effect on pH (Fisher and Binkley 1999). This mechanism could be important on sites that have sufficiently different aeration values for their soils, which can occur with the higher amount of organic matter mixing that takes place under alder stands. Differing transpiration rates between mixed and pure stands can also lead to changes in redox potential. This in-depth understanding of the mechanisms of pH change help to elucidate the differences in the impact on the soil of Douglas-fir and alder. Any accumulation of carbon due to alder presence is important, because of the influence of soil C on physical and chemical properties of soils. Soil C accumulation occurs primarily through litterfall (above- and below-ground) and mortality, both of which make the soil a more hospitable environment for crop species. On poor sites, total C in mixed alder-conifer exceeds 10  total soil C in adjacent pure conifer stands (Binkley and Sollins 1990). The mixed stands also seem to have more C down deeper than the pure conifer stands, producing greater total soil C than the pure stands. Alder can also be used as a tool for pest management. It can be used to control Phellinus weirii centres on sites where the disease occurs endemically, because alder is immune to the root rot. Interspersion of alder can also reduce other pest outbreaks in coniferous plantations, as the breakup of monoculture with areas of mixed species can deter some insect pests and reduce their populations. Shading of Sitka spruce leaders by surrounding alder nurse crops has been shown to reduce the occurrence of spruce leader weevil outbreaks (McLean 2000).  Principal Objectives The prospects for alder management spread further than merely acting as a nurse crop for commercially important conifer species. Strong, consistent markets have developed worldwide for alder lumber and increased use will bring higher prices and better information about the species (Peterson et al. 1996). The potential benefits of having an alder component in coastal conifer stands has led to several investigations of the influence of alder on productivity and growth of conifers, intermixed or adjacent. This dissertation follows up on earlier studies of the influence of alder on 1) the transmission of light, 2) litterfall and active nutrient cycling mechanisms, and 3) soils and nutrient cycling in deciduous and mixedwood stands to address the question:  "What is the effect of an alder stand on an adjacent Douglas-fir or conifer stand?"  The objectives of the study are to use sites of varied age and structure to measure the presence of any perceived influences of alder on the light and nutrient regimes, and the distance to which these influences can be observed. The underlying objective of this thesis is to answer the question of spatial influence.  11  Site characterization In order to document gradients in soil nutrient levels and light with distance to red alder, two transects were established at each of three sites in the Vancouver Forest Region of southwestern B. C. The three sites are located at 1) Maguire creek in the Chilliwack Forest District (CH); 2) the Malcolm Knapp Research Forest near Haney, B.C. (MK); and 3) Nanaimo Lakes Road, southwest of Nanaimo, B.C. (NM) (Fig. 1.1). The study sites are separated by relatively large distances and are marked by slight differences in vegetation and soil characteristics. This was intentionally done to permit evaluation of alder influence across a range of site conditions. Temperature, precipitation and climate data relating to these sites are summarized in Table 1.1. The transects each had unique topography and slope characteristics, which are graphed in Fig. 1.2. The method used to study these effects was a linear transect between an alder stand and a Douglas-fir stand. In order to model the transmission of light, the transect was expanded to encompass a plot for spatial measurements. The transect layout is depicted in Fig. 1.3.  Spatial measurements of the stand were completed in a plot that extended  10 metres on each side, and 10m beyond each end of the transect (55m x 20m). The Maguire Creek site (CH) is located along a spur off of the Tamihi Forest Service Road, overlooking the Chilliwack River valley. The transects are set up in a group of forest assessment plots established by Brian D'Anjou of the B.C. Ministry of Forests, Vancouver Forest Region (FRDA project 2.35). The Maguire II study site was logged and windrowed in 1982 then planted with grand fir (2+1 BR stock) in 1983. The disturbed mineral soil proved ideal for red alder regeneration, and by 1987, a dense alder stand (13,900 sph), averaging 4.3 metres in height and 2.3 cm diameter at breast height had established itself. Grand fir averaged less than 1.5 metres in height at this time. Seven treatments were established in 1988. These included four different manual cutting treatments, each completed at a different time of year, a glyphosate "hack and squirt" treatment, a total vegetation removal (complete) and an untreated control. The U T M Zone 10 (NAD83) coordinates for the Maguire Creek site are 5435000 m N , 586000 m E, and elevation is 380 m. Transect CH-1 has a bearing of 62°. The transect ran from an untreated control plot to a young manually planted grand fir (Abies grandis) block.  12  Fig. 1.1. Map of the lower mainland/ Vancouver Island region of British Columbia and, Showing the approximate locations of all research sites. M K represents Malcolm Knapp Research Forest sites. N M - Nanaimo Lakes site. C H - Tamihi Creek site near Chilliwack. SC litterfall transects in the Sunshine Coast Forest District. Map shown is not to scale.  13  14  Alder Stand  Conifer Stand  T3  c  9 O  CQ Additional litterfall traps 55 O — 7m —*03m(P  o  o  o  o  o  o 5m  Principal measurement points  o  Om  o  7m  <p  i  Additional litterfall traps O O O O O - 5m - O  10m 13m 16m  19m 22m 25m  30m  o  35m  Fig. 1.3. Diagram of the structure of the major transects used. All dimensions remain the same at each transect. The boundary is defined as a straight line along the average dripline between the two stands. The stand was mapped to 10 m each side of the main transect.  15  Table 1.1 Summary of climate data collected from representative biogeoclimatic reference stations (adapted from (Meidinger and Pojar 1991)). Biogeoclimatic unit  CDFmm  CWHxml  CWHvml  Number of stations Name of reference station Elevation of reference station (m) Mean annual precipitation* (mm) May to September precipitation* (mm) Total mean annual snowfall range * (cm) Mean annual temperature * (°C) Mean temperature of the coldest month * (°C) Extreme minimum temperature * (°C) Mean temperature of the warmest month * (°C) Extreme maximum temperature * (°C) Growing degree-days >5 * (°C) Frost-free period * (days)  52 Victoria Airport 17 636 to 1263 105 to 272 17 to 92 8.8 to 10.5 1.6 to 5.0 -7.8 to-21.7 15.1 to 18.0 27.8 to 40.6 1728 to 2163 155 to 304  76 Cumberland 159 1100 to 2721 160 to 565 26 to 234 7.8 to 10.7 -0.5 to 3.9 -13.5 to-25.6 14.2 to 18.7 29.4 to 43.9 1498 to 2330 137 to 244  32 Haney (Loon Lk.) 354 1555 to 4387 364 to 1162 20 to 548 7.0 to 10.1 -4.5 to 3.7 -8.9 to-22.8 13.8 to 18.8 27.8 to 41.1 1313 to 2011 165 to 252  * Denotes average over several years  The slope at the site was negligible (Fig 1.2), but it occupied a northern aspect of the mountainside. Transect CH-2 was generated the same way, with different alder / grand fir stands which occupied a similar trial area. The soils were Humo-Ferric Podzols derived from glacial till that had been disturbed by site preparation (windrowing). The soils had a high sand content (sandy loam to loamy sand) and between 10 and 30% coarse fragment content. The soil moisture regime was fresh (4-5) and the nutrient regime was classified as medium to rich (C-D). The site is in the Dry Maritime Coastal Western Hemlock (CWHdm) zone (Green and Klinka 1994), along the southern edge of the Fraser Valley, southeast of Chilliwack, B.C. The Malcolm Knapp Research Forest (MKRF) site (MK) is located approximately three quarters the way up the G40 shunt-spur from G road within the M K R F . The U T M Zone 10 (NAD83) coordinates are approximately 5458400 m N , 530400 m E at an elevation of roughly 175 m. The site is in the Submontane Very Wet Maritime Coastal Western Hemlock (CWHvml) biogeoclimatic subzone (Green and Klinka 1994) at the foot of the mountains on the northern side of the Fraser Valley. The alder and conifer stands were approximately 25 years of age. The soil was a Humo-Ferric Podzol derived from colluvium and glacial till. The soils were a loamy sand with coarse fragment content of 20 to 50%, and the site had a westerly aspect with a moderate slope (Fig. 1.2). The transects MK-1 and MK-2 extend outward from an alder stand 16  20m in width into Douglas-fir stands on either side. MK-1 extends from north (center of alder stand) to south (25 m into the Douglas-fir stand). MK-2 extends from SW to N E , again from alder to Douglas-fir. The Nanaimo Lakes Road site (NM) is located on the south side of the Nanaimo Lakes road. The site is 2.8 km east of the Timberwest main gate on the road. The U T M Zone 10 (NAD27) coordinates are approximately 5437800 m N , 419300 m E at an elevation of roughly 220 m. The site is in the Very Dry Maritime Coastal Western Hemlock (CWHxm, formerly the CDFb) zone (Green and Klinka 1994). The alder stand is approximately 45 years old and the conifer component is slightly older. The soils at this site vary from Humo-Ferric Podzols to Ferro-Humic Podzols in the center of the alder stand, where an ephemeral spring is located. The soils varied from silty loam to loamy sand, with coarse fragment content ranging from 5 to 40%. Transect NM-1 extends on a bearing of 110° from the red alder stand into an adjacent mixedwood stand of Douglas-fir, western white pine, and western hemlock, with a small proportion of red alder. Transect NM-2 extends from the same stand on a bearing of 55° (NE) up a slight slope (10-12°) into a relatively pure Douglas-fir stand. The aspect of the site is southsouthwesterly, but the slope declines dramatically between the NM-2 (upper) and NM-1 (lower) transects (Fig. 1.2). Three additional sites were selected for a study to document litterfall distribution. A second site was used at M K R F in a 23-year-old alder provenance trial (MKRF project 76-08) which lies adjacent to a dense stand of planted Sitka spruce (MKRF project 80-02). This site is part of several other University of British Columbia research projects that are unrelated to the present study. The site is in the Submontane Very Wet Maritime Coastal Western Hemlock (CWHvml) biogeoclimatic subzone (Green and Klinka 1994). Two other sites are located on the Sunshine Coast (Fig. 1.1), along the East Wilson and Gough Creek forest service roads. These are the youngest stands in the study, with alder planted in 1992. The transects were set up in the same manner as the main sites, although they only extend 20 m into the adjacent conifer stand. More detailed information about these sites can be found in (Comeau et al. 1997). The Gough Creek site is located near Sechelt on the Sunshine Coast in the Coastal  17  Western Hemlock Dry Maritime biogeoclimatic subzone (CWHdm). A slightly dry moisture regime (3) and a mesotrophic (c) trophotope characterize the site. The predominantly second growth stand of Douglas-fir was highlead logged in 1991, spur road slash accumulations were burned in the spring of 1992 and the stand was planted in March of 1992. The stand was part of the B.C. Ministry of Forests Small Business Program (Sunshine Coast Forest District T S L A29063). The East Wilson Creek site is located near Sechelt on the Sunshine Coast of British Columbia and is found in the CWHdm. The site has a slightly dry moisture regime (3), and is described as a mesotrophic (c) trophotope. The site was logged as part of the B.C. Ministry of Forests Small Business Program (Sunshine Coast Forest District - T S L A31036) in similar fashion to Gough Creek. The Gough Creek and East Wilson Creek sites had two or more litter trap transects installed in different directions extending from the alder stands into the conifers using the same distances as found on the major transects. These data were collected to provide information on the litterfall distribution and litterfall shadows across a range of stand ages.  18  CHAPTER II Light interception and transmission through alder and conifer canopies  Introduction Light is the term given to electromagnetic radiation within the visible spectrum. Plants use a fraction of the visible spectrum known as photosynthetically active radiation (PAR) which encompasses light travelling at wavelengths between 400 and 700 nm. Light in relation to plants is commonly measured as photosynthetic photon flux density (PPFD), but can also be expressed as percent transmittance, which is a percentage of the total light available in open-sky (Comeau 1998). Typical light levels under broadleaves differ from conifer understory light regimes. Light levels in one study of a boreal mixedwood measured 5.9% of incident light (Constabel and Lieffers 1996), which is quite low. The percentage of above canopy light (PACL) in particular alder stands has been measured at 10-20 for 5-year-old alder stands at a 2 m spacing (2500 stems per hectare). Natural, dense stands of alder (4000 sph) have been measured to have only 2-5 P A C L (P. Comeau Pers. comm.). Light levels generally decline with increasing basal area of red alder (Comeau 1996). It is difficult to precisely explain light under mixed or deciduous canopies (Brown and Parker 1994). Vertical structural differences can account for some of the variation in light transmission through deciduous canopies. Vertical structure variability increases with age, lending a higher degree of complexity to light interception modeling for mixedwood canopies (Brown and Parker 1994). Diurnal patterns of radiation become more irregular with depth of the forest canopy because beam, or direct, radiation penetration varies through space and with time (Hutchison and Matt 1977). Hutchison and Matt (1977) produced a general pattern of annual radiation for deciduous senescent canopy structures. The model predicts a slow increase of light transmission in the spring, which is reversed with leaf expansion. Following summer solstice, the light regime in the forest assumes an almost static state. Abscission brings a slight elevation in light, but this is short lived as solar elevation falls to its annual minimum in autumn. The random and complex overstory of a mixedwood leads to an understory that is relatively inhospitable to all but the most shade-tolerant conifers. Boreal mixedwood 19  environments often lead to understory light regimes below the photosynthetic light compensation point for conifer seedlings (Constabel and Lieffers 1996). When incorporating a shade-intolerant species such as Douglas-fir in a mixedwood scenario, the tree requires at least 20% of the relative light intensity (RLI) to survive (Mailly and Kimmins 1997). This is substantially higher than the light shown to be available in a boreal forest understory, and light levels beneath red alder are typically much lower than those found under boreal stands (P. Comeau, unpubl. pers. comm.). Differences in canopy transmission of light have been studied for a variety of species (Thomas and Comeau 1998), using a variety of techniques (Gendron et al. 1998). To date, no study has examined the spatial influences of red alder, its relationship to adjacent conifer stands, and the influences of gap size on light in this type of mixedwood. Each of these factors plays a role in determining the ability of trees and shrubs to survive and thrive in understory conditions. The ability of alder to overtop its coniferous counterparts makes light a key component in the adjustment of management scenarios to enhance the growth of alder and conifers. A variety of methods can be used to quantify and characterize light transmission through a canopy (see (Lieffers et al. 1999);(Gendron et al. 1998; Machado and Reich 1999) (Comeau et al. 1998); (Zinke 1962); (Ingestad 1962) (Brunner and Nigh 2000)for reviews). Different methods of measurement may work better for a particular purpose than others, and techniques should be viewed as complementary rather than competing (Mailly and Kimmins 1997). For this project, hemispherical photography was used to measure light levels in the understory of alder and adjacent conifer stands. Long term PPFD measurement using photodiodes were used to calibrate light estimates obtained from analysis of fisheye photographs. Hemispherical (fisheye) photography is a method whereby measurement and estimation of light is performed using a photograph taken looking upwards from the ground with an equiangular 180° fisheye lens. The photos are digitized and analyzed using specialized computer software. The software packages use algorithms that calculate the position and movement of the solar disc throughout a day, season, or year. It calculates the positions of canopy gaps using the amount of diffuse light coming from all directions at all time intervals, interpreting the percent of open sky light that is transmitted to the point in the understory where the photo was taken (Messier 1996). Although, hemispherical photography can provide accurate estimates (although underestimated) of the daily light regime for any particular point in the understory, calibration of 20  the photograph with actual light measurements is desirable (Spies et al. 1990). Criticisms of the fisheye technique are the costs of the photographic equipment, the lens, the time required for photography, conversion of the picture to digital format, and interpretation(Messier and Puttonen 1995). Other methods of below canopy light analysis, such as Canham's GLI-C gap light index measurement have demonstrated reliability in estimating the gap fraction or light index at a given point (Canham 1988). Where it is not possible to obtain direct measurements, models that use stand data and other information can be useful for estimating understory light conditions (Comeau et al. 1998; Gendron et al. 1998; Lieffers et al. 1999). The LITE model was developed by (Comeau et al. 1998) to estimate PPFD and transmittance over time periods ranging from one hour to one year. LITE uses geospatial information (latitude and longitude) to model the sun's path across the sky over the period being studied. LITE generates a forest canopy using stand information and data from hemispherical photographs and open sky PPFD measurements, or LAI-2000 plant canopy analyzer (LICOR Inc., Lincoln, NE) measurements (Comeau et al. 1998; Comeau 1998).  Objectives and Hypotheses This chapter examines the effects of red alder on understory light levels of differing ages and also examines the distance over which red alder influences light at stand boundaries. Because alder grows more quickly than associated conifer crops, the differences in the amount of light reaching the understory were examined to give insight into the differences in light transmission between the alder and conifer stands. The following hypotheses and objectives were developed: 1)  The quantity of light that reaches the understory in alder stands varies with age, canopy cover and structure; thus older alder stands will show less available light in the understory than young stands.  2)  The quantity of light that reaches the understory differs significantly between alder and conifer stands, resulting in a change in light availability across the stand boundary, from alder to conifer.  3)  Light levels will increase rapidly as one moves from an alder stand into an adjacent opening, and the extent of the stand influence will change with changes in stand 21  orientation.  Methods This portion of the project focused on differences in understory light associated with age differences and associated differences in canopy cover and structure. Light was measured with hemispherical photographs at 10 points along the transect and light sensors were used to calibrate the photographs in each stand. The aim was to characterize the spatial influences of alder stands on light transmittance into an adjacent conifer stand by obtaining a representative sample of light values within a chronosequence of alder and conifer canopies.  Hemispherical photography Hemispherical photographs were taken in smmer 1998, when the trees had full leaf development. The pictures were taken on overcast days, or at dawn and dusk to avoid any penumbral effects of a solar disc in the photograph. The hemispherical canopy photographs were taken with the camera mounted on a tripod at 1.4 m height. A light emitting diode (LED) assembly was used to align the camera to mark the edges of the photograph for later identification. The camera was a Nikon F601 camera with a Nikorr 8mm f/2.8 fisheye lens and Kodak T M A X 100 black and white film. After development, the negatives were digitized using an Olympus ES-10 scanner. The files were saved in Windows Bitmap (*.BMP) format which is compatible with the SLIM (Comeau and Macdonald 1998) and LITE (Comeau et al. 1998) analysis software. SLIM overlays the solar path during the growing season on the photograph and calculates beam and diffuse light reaching the measurement point over the entire growing season. Calculation of understory light levels is based on dividing the sky into 480 segments of equal area and estimating the amount of beam and diffuse light reaching the measurement point from each sky segment at hourly steps.  Long Term PPFD Measurement A record of PPFD for open sky and below canopy light conditions is useful for the setup 22  of the SLIM/LITE modeling tools. They allow for calibration of the estimates of below canopy light, which are produced from the photos. Calibration was achieved using a quantum sensor to measure light in the open sky and 5 photodiode sensors placed beneath the canopy at selected locations along each of four transects, which measured the PPFD at selected points simultaneously with the open sky quantum sensor. For this study, two types of sensors were used. One was a quantum sensor built by LICOR (LI-COR Inc., Lincoln, Nebraska) and the second sensor was built and used by B.C. Ministry of Forests (Fielder and Comeau 1998). The photodiode at the heart of the sensor is a Hamamatsu gallium-arsenide-phosphide (GaAsP) diffusion type photodiode (Type No. G271101, Hamamatsu Corp. Bridgewater, NJ). The photodiode sensors were calibrated against two LI-COR Quantum Sensors (LI-190) in natural sunlight over a minimum of one full day (often 2 or 3 days) using 10 min averages to derive an average calibration coefficient for a complete day. Calibrating a sensor for solar radiation consisted of mounting it on a calibration plate next to at least one calibration standard (LI-COR Quantum Sensor (QS) LI 190SA or SB, LI-COR Inc., Lincoln, N E , USA) and collecting data under open skies for a full day (05:00 to 22:00 h). Testing and calibrations were made with a calibration array consisting of a maximum of three QSs arranged in a triangle around 8 test sensors such that all the sensors would be level with respect to each other. The array was made up of 2 squares (15X15 cm) of % inch aluminum plate. The top plate was drilled with three " / inch holes to accommodate the LI-COR sensors and eight holes / inch in 7  16  8  diameter with equidistant centers (34 mm) to accept the test sensors. The two plates were screwed together so that the bottom plate provided a base for each sensor. A hole was drilled in the bottom plate beneath each sensor to allow water to drain and to make sensor installation and removal easier. Three threaded holes were drilled, 10 mm from each corner of the plates in the form of an equilateral triangle. Bolts were threaded through the holes to provide a way of leveling the entire array. Leveling of each sensor was checked with respect to the rim of the sensors and not the calibration plate. Placing a bubble level on the rims of both the GaAsP and the LI-COR sensors leveled the plate. Data was collected and stored as 10 min. averages of Is scans. Data was analyzed using the regression procedure in the SAS Statistical package (SAS System for Windows, release 23  6.12). The procedure was run separately for each day, with the test sensor mV value as the independent variable and the intercept set to zero. The dependent variable was an average of the calibration (QS) sensors. The slope parameter from the regression was used to convert the mV signal to PPFD, effectively calibrating the sensor. The sensors were also tested for response against a light source of known output prior to being taken into the field for setup. Quantum Sensors were cross-checked using the LI-COR LI 1800-02 Optical Radiation Calibrator (ORC). GaAsP sensors were also cross-checked against the ORC, and for this purpose an adapter was constructed to hold the slightly smaller diameter GaAsP sensors in the same position as the QS in the ORC calibration port. The ORC uses a 200 Watt quartz tungsten halogen lamp calibrated via transfer calibration (traceable to U.S. National Institute of Standards and Technology) to a 1000 W working standard lamp at LI-COR's laboratories. The absolute calibration accuracy of the ORC is ± 4 % from 350 to 1000 nm. The calibration lamp output was approximately 200 umol m~ s". The spectral response of the 2  1  photodiode was obtained directly from Hamamatsu Corp. and the spectral output of the lamp was provided along with the calibration information. Gallium-arsenide-phosphide sensors could not be calibrated directly for the full range of solar radiation levels with the ORC. However, the machine calibration was useful for calibration of the QSs and to track calibration drift for the GaAsP sensors. Values from the ORC could be used to adjust for changes in calibration values. Measurements in uA were taken for each sensor to derive a calibration constant (Ca)(uA/1000umol s'm" ). The calibration constant and a calibration coefficient were 2  determined using a standard light source and were dependent on the resistance used in the field. This provides a more sensitive method for tracking changes in photodiode calibration between the beginning and end of the measurement period (Fielder and Comeau 1998). The calibration constants for the known light source (ORC) and sunlight were derived as follows: Known light source: Ca(uA/lOOOumol. s'm" ) =  uA*1000  2  Sunlight: Cv(mV/lOOOumol. s'm" )2  standard lamp output (umol. m V ) 1  Ca* resistance(ohms) 1000  24  (1)  (2)  Calibration Coefficient =  ^r^L Ca (or Cv)  ^  In the field, the sensors were mounted on specially designed mounting posts at a height of 1.4 m. The sensors were placed at points 2, 3, 4, 6 and 9 for transects MK1, Mk2, NM1, and NM2.  The sensors were adjusted using a spirit level and mounting screws. Cables were run  from each sensor to a central location where they were attached to a Campbell CR-10 datalogger. An open sky LI-COR sensor was also set up at the Malcolm Knapp site using coaxial cable to extend the distance at which it could be setup to just under 200 m. The Nanaimo transects were too far from a suitable opening to allow collection of open sky readings. Dataloggers were programmed to take readings every ten seconds and then record an hourly average PPFD based on the readings for each sensor, and then calculated a maximum and minimum daily value for each sensor. A sample program is shown in Appendix I. The data was downloaded every four weeks by connecting a portable computer to the datalogger. The sensors were cleaned, checked for damage and observed for proper function every time the system was downloaded, with malfunctions repaired on-site. The systems recorded PPFD data from May 1998 to February 1999.  Stand mapping for LITE Spatially explicit stand maps and tree data were collected from each transect for input and use with the LITE model. The bearing of the transect was calculated and a 20 m X 55 m area centered on the transects was intensively sampled. All trees in this area were tagged and numbered. Tree species, diameter at breast height (DBH), and spatial position were recorded. Position was initially recorded using distances and bearings from known points, then converted to an X - Y location using trigonometric functions. These values were recorded in MS Excel, and then converted to a stem list file (*.STM) which is compatible with LITE. A n example of a stem list can be found in (Comeau et al. 1998).  Analysis using SLIM / LITE Using digitized hemispherical photographs (see Appendix II for examples), SLIM generated an estimate of leaf area index (LAI) and gapfractionfor each point along each 25  transect. Using the information from SLIM, a stem map of each plot, and data from an open sky sensor to calibrate the images, LITE provided estimates for average monthly PPFD for each point. The values produced by SLIM and LITE are compared to the data provided by the photodiodes. See Appendix III for examples of SLIM and LITE interfaces. The simulations with LITE were individually generated for each transect. All transects used open sky data from the M K site, and the site specific stem file data. Tree heights, crown radii and crown depth were generated by LITE, based on Ministry of Forests data. The canopies were all generated using the tree data as a basis for height to crown base (HTCB) and height to top of crown (HTTC). LITE generates a canopy consisting of 1 m cells, with sky areas based on 3  hemispherical photo data interpreted using SLIM. Once the crown had been validated, and leaf area generated for the canopy, longterm PPFD was calculated based on the open sky readings from the quantum sensor data for M K . The PPFD for each point at each transect was calculated for the months of June, July, August and September. SLIM and LITE produced values representing monthly averages for percent abovecanopy light (PACL) in the 400-700 nm wavelengths (PACL). SLIM averages are constructed through analysis of hemispherical photographs, while LITE values are generated through a combination of the SLIM photo interpretation and a spatially explicit model of each forest stand. By removing the conifer portion of the stem map for the stands the effects of light transmission through alder on adjacent gaps were modeled. The stand orientations were altered to the four cardinal bearings while keeping other characteristics constant to estimate the change in light transmission dependent on aspect.  Results Validation of LITE and SLIM estimations for long-term PPFD calculations using long term photodiodes Comparisons between the photodiodes and the estimations observed from hemispherical photography (SLIM) and LITE were difficult to make due to malfunctioning photodiodes and quantum sensors. By the end of the field experiment, four understory photodiodes and the sole open sky sensor had ceased to give any detectable signal. Several of the photodiodes experienced 26  fluctuations in calibration of well over 5% over the period in the field. Given the very low light levels that were being measured, 5% is too large a drift in sensor calibration to provide useful data. Validation of the SLIM and LITE estimations was achieved using the sensor data for the month of July 1998. The sensors seem to have performed reasonably well during these months, but it is unknown if the drift in calibration occurred suddenly or gradually over time. It is assumed for the purposes of this validation that sensor calibration remained constant during the first three months of operation. There was a significant relationship between the data from the photodiodes and the values calculated by SLIM (Fig. 2.1), and between the photodiodes and LITE output values as well (Fig. 2.2). Regression statistics for the relationships are presented in Table 2.1. Correlation between model estimates and photodiode measurements are much poorer at the N M site than at the M K site. The likely reason for the weaker relationship at N M is due to the lack of an open sky sensor at that site. The correlation would be weakened by the differences between the sky conditions at the open sky sensor at M K R F and the photodiodes at N M . A more suitable substitute for open sky data could not be found. It is also possible that one or several photodiodes at this site did not perform properly, but I could not find the source of the faults. SLIM output was better correlated with the diodes than the LITE output. This is likely due to the added complexity of the stand model and its incorporation in the calculation of understory PPFD. Both SLIM and LITE appeared to systematically underestimate light levels that were measured by the photodiodes. Comeau (1996) and Gendron et al. (1998) also found that LITE tended to underestimate light levels.  27  30 1  25  20  15  10  5  10  15  20  Measured % total PPFD (at 1.5 m)  Fig. 2.1. Relationships between SLIM interpretations and photodiode measurements for P A C L in stands studied. P A C L was calculated based on measurement or modeling of light over the entire month of July 1998. All values are relative to open sky PPFD calibration with LI-COR 160-SA sensor.  28  Fig. 2.2. Relationships between spatially explicit LITE simulations and Photodiode measurements for P A C L at N M and M K sites. P A C L was calculated based on measurement or modeling for the entire month of July 1998. All values are relative to open sky PPFD calibration with LI-COR 160-SA sensor.  29  Table 2.1 Regression statistics describing the relationships between SLIM and LITE simulations and long-term in situ photodiodes. All values are based on July 1998 data. Site MK NM MK NM  Simulation SLIM SLIM LITE LITE  Month July July July July  N 7 9 7 9  Slope Intercept 0.945 7.0162 0.492 7.4039 2.067 -0.8122 1.126 -4.6337  R-squared F-statistic 0.9317 56.03 7.218 0.4743 0.9033 56.03 0.7364 22.35  />-value 0.0002 0.0276 0.0002 0.0015  Transmission through alder and conifer canopies with SLIM and LITE The values from SLIM for July 1998 display subtle differences between stands, and more obvious differences between the sites (Fig. 2.3). The youngest stands (CHI and CH2) have the greatest differences between the alder and conifer stands and display the highest light levels underneath the alder stands. At CHI and CH2, the alder have P A C L up to 20% higher than the adjacent conifer stands. In contrast, the MK1 and MK2 stands vary (partially due to their opposing bearings), but display only a 15% difference in light transmission between the two stands. The oldest stands (NM1, NM2) show slightly higher P A C L values in the conifer than in the adjacent alder. At NM1, this could be attributed to the presence of alder in the part of the transect designated as conifer, but a similar trend is seen in the NM2 transect where no alder is present in the conifer stand. There is a general trend of equilibration of light transmission between alder and conifer stands with age. The darkest understory measurements for both conifer and alder stands were observed at the youngest site. Likewise, the understory with the most light occurred at the oldest site. LITE produced similar results, with some minor exceptions (Fig. 2.4). The youngest site shows higher P A C L available in the alder compared to the adjacent conifers. There is a slight elevation of the P A C L at the end of the C H 1 transect which is due to nearby alder in the next planting block. LITE also indicated P A C L peaks in the conifer portions of the stands MK1 and MK2. These peaks in transmittance persisted through June, July and August simulations. LITE had substantially higher values for the oldest alder stands compared to the SLIM output. Disparity between the model and the recorded values could be indicative of a misinterpretation of the canopy. Error could be due to top damage of alder at N M , which could have been adjusted 30  Alder stand 10  Conifer stand 20  30  40  Position on transect (m).  Fig. 2.3. Percent above canopy light (PACL) along transects at 1.5 m above the ground determined using SLIM fisheye photo analysis for the month of July, 1998.  31  Fig. 2.4. Percent above canopy light (PACL) expressed as % total PPFD, determined using LITE spatially explicit stand and fisheye simulation analysis for July 1998.  32  using individually measured tree heights (which were not collected in this study). The oldest stands (NM) also showed lower P A C L values in the conifers nearest the alder stand, suggesting an increased leaf area in this part of the stand.  Gap light adjacent to alder stands modeled with LITE Removing all conifers from each stand map, light transmission was modeled into open gaps from each alder stand with LITE (Fig. 2.5). Removal of all trees at Y coordinates greater than 20m was achieved by saving the LITE file as a new entity and deleting each conifer individually. This was done to create a stand map with the characteristics of the natural alder stand, with a hypothetical open area beside it. The distance from the alder stand at which light availability reaches a threshold for growth is affected by the bearing of the transect, the density and height of the trees, and the location of the sun relative to the stand. Light proximate to an alder stand quickly attained a level that is acceptable for good Douglas-fir growth, i.e. beyond 60% P A C L (Carter and Klinka 1992). This threshold is attained between the stand boundary and 10 m beyond the stand boundary, with differences due to the slope and aspect of the site. In the case of a stand that has a boundary that runs east-west with the alder on the north side (as in MK1), 60% P A C L is achieved 3 m inside the alder stand boundary. The LITE model was used to investigate the effect of transect orientation on light gradients across the stand boundary. This was done by changing the orientation of the transect, while keeping the stand and canopy characteristics constant. The difference in the availability of light between MK1 and MK2 led to further investigations of the effect of bearing on the light regime of gaps adjacent to the alder stands (Fig. 2.6). A n examination of the open sky data showed an unequal partitioning between light accumulated before and after 12:00 noon. This is likely due to the positioning of the open sky sensor, and the nature of the terrain that was available at M K R F for the sensor. As a result, values for stand boundaries facing east and west showed unusual differences in the amount of light that they would receive. To remedy the skewed daily PPFD, a default pfd file was used assuming clear evenly-lit days for the simulation. The latitude of the study allowed for an improvement in light to occur on south-facing transect models. Model simulations predicted that 60% P A C L (required for good Douglas-fir growth) is attained at or very near the stand boundary for south facing sites, between 5 and 10 m from the 33  80  MK1 MK2 NM1 NM2 CHI CH2  Acceptable Douglas-fir growth (Carter & Klinka 1992)  60 Minimum for Douglas-fir growth  I  (Mailly & Kimmins 1997)1 40 Douglas-fir survival (Mailly & Kimmins 1997) 20  Opening 10  20  30  40  Position along transect (m)  Fig. 2.5. Percent above canopy light (PACL) determined using LITE spatially explicit stand and fisheye simulation analysis with all conifers removed from each stand. LITE estimate of % total PPFD is based on estimates of total PPFD for the entire month of July, 1998.  34  100  50  10  20  30  40  •  100  -  50  10  20  30  40  20  30  40  100  50  J  40 10 Position along transect (m)  Fig. 2.6. P A C L determined using LITE spatially explicit stand and fisheye simulation analysis with all conifers removed from each stand for the month of June, 1998, rotating each transect through the four cardinal directions. Legend name refers to the position of the opening relative to the alder stand, e.g. South = opening (no conifer area) to the south of alder stand, with eastwest boundary.  35  boundary for east- and west-facing sites, and between 8 and 15m from the stand boundary for north-facing sites. Sites with south-facing boundaries attained a light level acceptable for good conifer growth up to 10 m closer to the boundary than sites with north-facing boundaries. Eastand west-facing boundaries had only minor differences between them, likely due to the nonuniform edges of the alder stands.  Discussion When using any model, it is important to verify that it adequately reflects measured values. This was difficult to achieve in this study due to the small number of sensors used and several technical difficulties. The technical challenges of using long-term field sensors is a primary reason for the development of models that can accurately predict understory light regimes from very few calibration values (Comeau et al. 1998). The strong correlations between the photodiode measurement at M K R F and values estimated by the models indicate that the information provided by the models can be useful. SLIM and LITE underestimated the measured values in every instance that there was a significant correlation. This is a common trait of light transmittance models, due to their inability to adequately predict sunflecks, reflected light, and small canopy gaps. These problems have been addressed in previous examinations of LITE (Comeau 1998; Comeau et al. 1998). Consideration of light in alder - conifer mixedwoods is a requirement for effective management since red alder dramatically reduces the amount of light reaching the understory. Douglas-fir, a likely candidate for mixtures, requires an absolute minimum of 20% P A C L to survive (Mailly and Kimmins 1997). Optimum height growth of Douglas-fir can be attained with 40%> RLI (Carter and Klinka 1992), although 60% RLI is generally considered a management minimum (Mailly and Kimmins 1997). The alder stands studied are stocked naturally and predictable patterns of light usage between the two stands emerged. The conifers are initially slow to reach crown closure, but are more efficient at light interception. This leads to dark understory environments that gradually open up as the stands thin themselves and the needles begin a regular cycle of senescence. Alder quickly reaches crown closure, and maintains a stable crown leaf mass by 10-15 years of age (Zavitkovski and Newton 1971).  36  Simulations indicate that light levels are adequate for good growth of Douglas-fir in openings almost up to the edge of an alder stand. A patchwork system of alder and conifer stands would show a light gradient similar to the 'no conifer' variant as the alder overtopped the adjacent conifers. The model suggests that conifers could be planted right up to the alder interface with little likelihood that their minimum light requirements would be threatened. Thus, it may be possible to grow patches or strips of red alder within conifer plantations, with the negative influence of alder on light being restricted primarily to the area occupied by the alder patches. The effect of orientation of the stand boundary was found to influence the quantity of light near the boundary as well. It was demonstrated that on a north-facing boundary, a space of up to 15 m from the alder stand would not have enough light to support good Douglas-fir growth. Along a south-facing boundary, the space between the alder and the point where good Douglasfir growth could occur was much closer to the stand boundary (between 0 and 5 m). Examination of east- and west-facing boundaries showed that only minor differences in light transmittance occur between east- and west-facing boundaries. These results suggest that strips of red alder might have the least impact when they are oriented north to south. This exploration with the model merits further investigation of patterns that might provide adequate light regimes combined with the numerous benefits of planting mixedwoods (Comeau 1996). The simulations were conducted with all measurements being an estimated 1.5 m from the ground. The height of the conifer component relative to alder will also influence the amount of light it receives. Because light interception increases with depth of canopy, conifers which are taller than 1.5m will have access to more light. This would ensure that conifer growth would not be impaired by difficulties in obtaining adequate light, since they would be capturing more light with each year of growth. In summary, the results showed that although the understory light in the alder stands studied is inadequate for conifer regeneration, adjacent areas could be successfully regenerated almost up to the boundary of the alder stand. In the case of alder stands (as in MK1) with a south-facing alder boundary, conifers could be planted 3 m into the alder stand and receive adequate light. Further examinations with LITE and variations on alder and conifer patterns should be completed to find optimum planting systems.  37  The results showed that the quantity of light that reaches the understory in alder does very with age, canopy cover, and structure. The quantities of light reaching the alder and conifer understory areas were different, with higher light under the broadleaf canopy than under the conifer canopy. This difference in light availability was overshadowed by the knowledge that neither understory light regime had enough light transmission to promote successful Douglas-fir regeneration. The light levels in a hypothetical opening adjacent to an alder stand quickly reach levels that are adequate for conifer growth. Variation in the distance from the alder stand to the point at which an acceptable light regime is reached is dependent on aspect of the boundary, with southfacing boundaries showing acceptable light levels up to and into the alder stand.  38  CHAPTER III  The roles of leaf litterfall, macrofaunal and microbial communities in propagating the alder influence Introduction Red alder litterfall appears to reach a maximum mass of between 4000 and 9000 kg ha"  1  yr" by 10-15 years of age and declines only slightly with increasing age (Zavitkovski and 1  Newton 1971). Litter production and nutrient concentrations are stable over time (Gessel and Turner 1974), and litter production in red alder stands is not affected by tree height, stand density (sph) or site quality (Zavitkovski and Newton 1971). Mass of alder litterfall increases rapidly with age and eventually reaches a plateau near maximum height (Zavitkovski and Newton 1971; Gessel and Turner 1974; Swanston and Myrold 1997). The importance of alder litterfall is contingent upon where the leaves actually fall. The broad, light leaves are blown to the ground on the periphery of the origin tree. The distance to which they can travel is related to the height from which they are launched. They are subject to the effects of wind, resistance and obstacles encountered on the way down. All of these variables affect each leaf differently, but the quantity of the litter is such that the end result is a homogenous carpet of leaf litter, gradually fading with distance from the origin. The litter shadow of a tree delineates the immediate sphere of influence of that tree. Litter is not the only important influence a tree can have on its surroundings, but it is though to be an important mechanism of influence on soil an nutrient cycling. Studies have shown that a single tree can have a significant influence on its surroundings (Zinke 1962; Rygiewicz et al. 1984; Boettcher and Kalisz 1990); thus, mixed stands of trees can show a mosaic of influence. Zinke (1962) found that although the influence of a tree was most obvious directly under the crown, influences on soil nitrogen contents and pH levels extended beyond the edge of the canopy. He concluded that the influence of various forms of litter on soil properties was the major cause of trends in soil characteristics. Single-tree influences have been detected in mixed forest stands, leading researchers to consider the soil landscape as a mosaic reflecting vegetation influences (Boettcher and Kalisz 1990). These studies suggest that the litter shadow of a tree, or of a stand of trees, is 39  largely responsible for patterns in soil nutrient availability. If the consistent yearly deposition of alder litter into an adjacent conifer stand is observed to reach 10 m (Rhoades and Binkley 1992), then it would be expected that the imprint of alder influence on the soil would be observed to that point. In a spatially explicit model of leaf litterfall for hemlock/hardwood forests in Michigan, (Ferrari and Sugita 1996) found that 90% of litterfall is found within a 17.1 m radius of the origin. This litter chemistry: soil chemistry feedback appeared to influence the vegetation distribution in the forests (Ferrari 1998). The distance that litter travels from the source after senescence is thought to be species- and site-specific. The large amount of alder litter that falls in autumn can have a substantial influence on arthropod activity. Acceleration of nutrient cycling due to the presence of alder may result from more active arthropod and microbial decomposer communities. Litter of alder and other N fixing species is a preferred food of earthworms (Brown and Harrison 1982), and earthworms have demonstrated a preference for Alnus litter over Pinus and Larix litter (Satchell and Lowe 1967). Litter quality is more important than quantity in earthworm communities (Binkley et al. 1992b; Zou 1993). Earthworms are more abundant in the soils of alder stands than in soils of Douglas-fir stands, probably due to higher quality litter in alder stands. For this reason, they can be an important mechanism of decomposition and carbon movement in alder stands (Brown and Harrison 1982; Bormann et al. 1994). In a tropical study, using N -fixing Albizia and Eucalyptus 2  in Hawaii instead of alder and conifers, earthworm density under Albizia was five times that of the adjacent Eucalyptus (Zou 1993). Similarity between the scenarios of Albizia and Eucalyptus in the tropics and alder and conifer in coastal temperate forest are thought to be linked by the fixation of N and enhanced nutrient cycles. Similar examinations of macro fauna are lacking for alder-conifer ecosystems. Harpaphe millipedes show a slight preference for Douglas-fir foliage over alder in a microcosm study, but the millipedes were observed to aggregate in broadleaf stands (Carcamo et al. 2000). Very little work has been done examining the diversity of arthropod or mollusc communities under red alder, or comparing these communities to those found under conifer stands. Mixing of soil by soil fauna has been proposed as one of the reasons that higher levels of carbon are found deeper in soils under alder stands (Bormann et al. 1994). Potential differences in structure and size of the microbial pools between alder and 40  Douglas-fir stands would result in different nutrient cycling rates relating to decomposition rates. Microbial basal respiration rates in tropical plantations of Eucalyptus were not affected by the presence of Albizia (Zou 1993). Basal respiration rates may not be sensitive solely to differences in soil nutrient availability. Nitrogen transformation rates in laboratory incubations in the absence of macrofauna (soil only) did not show differences between Albizia and Eucalyptus samples (Zou 1993), although there were large differences seen in situ (Binkley et al. 1992b). This indicates that microbial pools may need a mutualist factor, such as increased earthworm populations, to increase respiration and speed up decomposition of qualitatively superior litter (Barois and Lavelle 1986). Microbial respiration has been shown to increase 800% in the anterior gut environment of Pontoscolex earthworms (Barois and Lavelle 1986). As well, basal respiration rates have been shown to be sensitive to some differences between tree species, stand age and soil type (Bauhus et al. 1998).  Objectives and hypotheses The objective of this study was to sample nutrient input mechanisms that may explain the effect of tree species on soils. The effect of litterfall in influencing the soil, as well as microbial populations and macrofaunal differences between the two stands have been studied. Because the soil quality, judged by the above variables, is thought to be a function of tree species, it would follow that the mechanisms of soil change will also vary between the two stands, leading to the following hypotheses: 1) There is a litter 'shadow' around alder, and it is related to increased nutrient cycling due to the influence of alder litter on the soils. 2) The spatial distribution of alder litterfall increases rapidly with age and then reaches a plateau at maximum height. 3) The concentrations of nutrients in alder will be higher than related nutrient concentrations in adjacent conifers, leading to a disparity between the nutrient inputs of the litter from each tree type. 4) The macrofauna and microbial populations are greater in alder than in adjacent conifer stands because of the quality of litter inputs in each stand. This difference will result in larger populations of arthropods and higher levels of microbial basal respiration in the forest floor. 41  Methods Litterfall Litter traps measuring 25 cm x 50 cm (0.125 m ) were used, with fine mesh placed in the 2  bottom to allow drainage. At each measurement point along each transect, three litter traps were placed in a row perpendicular to the transect. One was centered on the transect, and the other two were placed 5 m on either side of the original transect traps. The traps were installed in July 1998, sampled every two weeks from mid-August until mid-November, and then monthly through winter, spring and summer of 1999. Litter collected from each trap was immediately dried for 48 hours at 70°C, then separated into leaf and needle litter and composited into samples representing the total annual litterfall (leaf and needle) for that trap. Each of the 30 litter traps per transect has a needle and leaf weight, which has been converted to weight per hectare per year (kg ha" yr" ). Three composite samples for each site were generated for leaf and needle litter 1  1  by weighing out equal portions from each trap and mixing them thoroughly. These samples were ground and 1 g of plant tissue was individually analyzed for N , P, K, Ca, Mg, S, Cu, Fe, Mn, Zn and B concentrations through a modified microwave acid digestion process at B.C. Ministry of Forests Glyn Road analytical lab (Dawson and Dunn 1998). The process uses a very strong oxidizing acid mix heated by microwave energy to vigorously destroy plant tissue matrix and free up the elements of interest. A known amount of vanadium is added as an internal standard to every sample and control early in the procedure. The vanadium has the desired internal standard qualities of similar chemical and spectrographic properties to the elements being measured. The method is a variant of a well used procedure (Kalra and Maynard 1991), with the vanadium use being a modification of the procedure by the analytical lab (Dawson and Dunn 1998). The method has been verified and tested through participation in the University of Wageningen International Plant Exchange inter-laboratory correlation process. Nutrient concentrations were multiplied by annual litterfall mass to calculate nutrient inputs at each sample point.  42  Macrofaunal collection and classification A survey of macrofaunal communities along the 6 major transects was carried out over a 100-day period from June 15 to October 12 1999. At every second transect point, two th  th  replicate pitfall traps were embedded into the ground. The traps were filled to between 5 and 10 cm depth with a 50% glycol P solution. The traps were emptied at 30-day intervals (with the exception of the final removal, which occurred after 40 days) and samples were refrigerated until analyzed. The glycol-P solution was drained through a fine mesh sieve (0.5mm or #35 mesh), and the sample was rinsed with water to remove fine sediments and materials clouding the sample. The sample was then placed in a tin pan and irrigated with saline solution (10%) NaCl). The insects present were removed, sorted, and counted. Some insect orders were divided into families due to the large number of specimens trapped. The data were transformed to represent proportions per sample. The proportions were analyzed using random effects analysis of variance (ANOVA), looking for differences between samples from alder stands and samples from conifer stands. The A N O V A was performed on the entire data set and then separately by site. There were too few samples to obtain worthwhile tests by transect. The data were also compared among the three major sites.  Microbial Basal Respiration Rates Forest floor samples were collected from each site and kept refrigerated. A subsample from each sample tub was taken and weighed, then dried to determine moisture content. Equal dry weight subsamples were then placed in glass-jars and sealed. After 24 hours in the sealed jars, 1 ml samples of the headspace gas were extracted with a syringe and injected into a LI-COR Li-800 Gashound infrared gas analyzer (IRGA) (LI-COR, Lincoln NE.). A calibration curve was created and the data were analyzed in terms of equivalent volumetric units of a standard 1510 ppm C 0 gas. The method measured the amount of C 0 evolution over a given length of time, 2  2  which is an indicator of microbial basal respiration rates, indicating activity.  43  Results Litterfall The litterfall was sampled to address several questions relating to the influence of alder on nutrient cycling. The first hypothesis tested was that the age of the alder stand significantly influenced three factors: 1) the quantity of annual litterfall, 2) the quantity of individual nutrients in the litter, and 3) the distance over which the litter moves from the stand boundary. This first portion of the litter collection used stands of several ages comprising a chronosequence from 5 years old to approximately 35 to 40 years old, as well as previously published values. The litter collected in this and similar studies in the Pacific Northwest (Zavitkovski and Newton 1971; Gessel and Turner 1974) indicated that red alder reaches a maximum litter output of 4000 to 7000 kg ha"' yr"' between 10 and 15 years of age (Fig. 3.1). Litter production was low in the oldest stand (NM) relative to the other sites and was likely the result of top damage of alder trees at this site. There was little variability in the litter trap outputs for traps of similar distances within individual stands. There were large differences among sites in the distance to which alder litter traveled into adjacent stands. Alder litter shadows from the stand boundary varied from 0 m in the young stands to 20-25 m in the oldest stands. The maximum distance at which the litter shadow occurred showed a significant logarithmic correlation with age, which is displayed as a log-linear regression (y= -7.7425 + 18.59x, R =0.80,/?<0.001) (Fig. 3.2). Maximum distance was defined 2  as the measurement point nearest to the alder stand that accumulated less than 4 g of alder litter. The increase in the litter shadow distance with age is likely due to the increased height of alder with age, coupled with a maximum litterfall production by age 15. As the conifers reach pole stage, and the crowns lift, the area under the closed conifer canopy will become more spacious, allowing for more alder litter ingress from below, where the conifer crown has receded upwards, and the adjacent alder canopy is lower than the base of the conifer crown.  44  y y  6000  y  ^  y  y  y  F  yy y  y  y  y  y  A  V  X  J  'ca  y  y  y  X  x  x  Hff  4000  y  yy  X  E  y  x  c D  ~ 2000 CB o H  J  k  K  1 0  i  10  i  i  I  20  30  40  Age of stand (years)  Fig. 3.1. Amount of leaf litterfall (kg ha" yr" ) in alder stands of various ages. Data points "A" though "N" represent data collected in this study. Points marked "x" represent values from (Gessel and Turner 1974). Points marked "y" represent values from a previous study by (Zavitkovski and Newton 1971). 1  1  45  46  6000  * -»— -•— --*— •-  point 1 (alder) point 2 (-3m) point 3 (boundary) point 4 (+3m) point 5 (+6m) point 6 (+9m)  „ * .  CO  3  4000 1  c  < 2000  0  10  20 30 Age of stands (yr)  40  Fig. 3.3. The relationships between alder litterfall input, distance from the alder stand, and the age of the stand for points 1 through 6. Each curve represents the quantity of litter that is observed at a given distance for the variously aged stands. Points 1 through 6 are thought to exemplify the initial distance decay pattern in red alder litterfall emigration, limited by adjacent conifer crowns and height of the alder.  47  The unusual relationship between stand age and distance of alder litter deposition can be seen by analyzing the results by transect point and age. For the first 6 points (0 m - 19 m) the relationship between litterfall distance and age is punctuated by an increase of litterfall until 1015 m from the alder stand boundary (Fig. 3.3). At a point between 30 and 45 years, corresponding to the distance between the 6 and 8 points along the transect, the relationship th  th  between age and distance changes from a convex power curve to a concave exponential relationship (Fig. 3.4). This suggests that a second mechanism of under canopy thinning and opening takes over as the key factor determining the distance that alder litter travels. By measuring the change in litter distribution distance, and placing that value in the linear regression of age and distance, it can be shown that the relationship between the alder litterfall and the adjacent stand's physical resistance to the litterfall changes. The curve does not change its inflection until at least point 8 (25 m), which corresponds to a minimum 15 m from the alder stand. This point seems to be the maximum distance alder litter travels into adjacent conifer stands until the conifer stand begins to become pole-like, so that the alder litter can more easily travel under the conifer canopy, rather than through it. It becomes apparent using the linear regression of log, (age) and maximum distance that a 15 m maximum distance for leaf litter 0  movement occurs at approximately 16 years of age. This indicates that for the first 16 years, alder litter will increase its shadow to a maximum distance of about 15 m. Once the neighboring conifers reach pole stage, the alder litter shadow increases further into the neighbouring conifer stand. The inability for alder litter to penetrate deeper than 15 m into the adjacent conifer stand until the trees reach pole stage may be the principal reason that alder influence has been confined to 15 m from the stand boundary in previous studies.  48  point 7 (+12 m) point 8 (+15 m) point 9 (+20 m) -p3000  oo B-2000 c  T3  1000  10  20  30  40  Age of stands (years)  Fig. 3.4. The relationships between alder litterfall input, distance from the alder stand, and the age of the stand for points 7, 8 and 9. Each curve represents the quantity of litter that is observed at a given distance for the variously aged stands. The point of inflection at point 8 (15 m into the conifer stand) represents the point where the primary mechanism of alder ingress limitation is thought to change from height of the alder canopy to openness below the conifer canopy. The 45year-old stands are the only stands to demonstrate this transition.  49  The dry weight of alder leaf litterfall within the pure stand was greater than conifer leaf litterfall in the pure stand at all three major sites (Table 3.1). Concentrations of K, Ca, S, and Mn, N , P, Mg, Fe, Cu, B and Zn in the litter differed significantly between the two litter types. This led to separate tests for each nutrient, since the value for litterfall mass did not adequately describe the relationship between the individual components in the two litters. Multiple comparisons for these results are found in Table 3.1. Table 3.1 (Continued on next page). Analysis of Variance (ANOVA) for total litterfall at each point along each transect, and N , P, Mg, B, Fe, Cu, and Zn (kg ha") through leaf litterfall. Each significant A N O V A has Tukey's honestly significant differences test associated with it. 1  Transect  Attribute  MK1  Litterfall Litter N Litter P Litter Mg Litter B Litter Fe Litter Cu Litter Zn Litterfall Litter N Litter P Litter Mg Litter B Litter Fe Litter Cu Litter Zn Litterfall Litter N Litter P Litter Mg Litter B Litter Fe Litter Cu Litter Zn Litterfall* Litter N* Litter P* Litter Mg Litter B Litter Fe Litter Cu Litter Zn Litterfall  F-value  p-value  (kghayr)  MK2  NM1  NM2  CHI  6.21 10.71 9.86 8.76 9.04 161.36 35.09 5.92 4.74 8.82 8.38 20.75 22.24 70.31 32.12 6.67 5.04 3.17 3.23 3.02 2.69 81.77 31.39 10.12 2.33 1.42 1.26 8.093 11.77 86.46 38.54 7.58 3.54  0.0003 O.0001 O.0001 O.0001 <0.0001 <0.0001 <0.0001 0.0004 0.0010 O.0001 O.0001 <0.0001 O.0001 O.0001 O.0001 0.0002 0.0010 0.0150 0.0140 0.0190 0.0300 <0.0001 <0.0001 O.0001 0.0550 0.2440 0.3130 O.0001 O.0001 <0.0001 O.0001 O.0001 0.0090  Tukey's Honestly significant differences test results 1  2  3  4  5  6  7  8  a a a a a ab a a a a a a a a a a abc ab ab ab ab a a ac N/A N/A N/A a a a a acd ab  a ab ab ab ab a a ac a ab ab ab b a a ab b ab ab ab ab a a ac N/A N/A N/A a a a a acde ab  ab abc abc abc abc ab a abc b be be be c a a b c ab ab ab ab ab a a N/A N/A N/A ab ab a a ace a  be abed abed abed abed be a be b c c c c b ab b b ab ab ab ab b ab ac N/A N/A N/A ab abc b b be ab  c cd cd cd cd be a b b c c c c c be ab c a a a a c be acd N/A N/A N/A ab be c c b b  c d d d d c a b a c c c c cd cd ab abc ab ab ab ab c cd abed N/A N/A N/A b be c c b ab  abc cd cd cd cd d b abc a be be c c cd cd a abc ab ab ab ab c cd c N/A N/A N/A b c c c be ab  abc cd cd bed cd d b abc a be be c c d d a abc ab ab ab ab c cd bed N/A N/A N/A b be c c bd b  50  9  10  abc be cd cd cd cd cd cd cd cd d d b b abc abc a a be be be be c c c c d d d d a a ab a ab b ab b ab b ab b c c cd d Be be N/A N/A N/A N/A N/A N/A b b c be c c c c b be ab ab  Litter N Litter P Litter Mg Litter B Litter Fe Litter Cu Litter Zn Litterfall Litter N Litter P Litter Mg Litter B Litter Fe Litter Cu Litter Zn  4.74 4.39 5.07 3.22 3.23 4.61 2.85 10.4 13.54 12.65 14.31 12.81 4.87 5.62 9.08  0.0020 0.0030 0.0010 0.0100 0.0130 0.0020 0.0240 O.0001 O.0001 <0.0001 <0.0001 O.0001 0.0020 0.0006 <0.0001  ab ab ab ab a a a a a a a ab a a ac  a a a ab ab ab ab a a a a ab a a abc  a a a a ab ab ab a a a a a ab ab ace  ab ab ab ab ab ab ab a a a a a b c ad  b b b ab ab ab ab be be be be b a abc be  b b b ab ab ab ab b b b b c a abc be  ab ab ab ab b b b b b b b c a abc be  b b b b ab ab ab b b b b c a abc bd  ab ab ab ab ab b b b b b b c a ab bed  ab ab ab ab b b b ac ac ac ac ab ab be a  * A N O V A was not significant atp<0.05  The litterfall showed slightly different patterns for each transect. For transects MK1, M K 2 , NM1, and NM2, total litterfall mass for alder and conifer litter peaked in the respective stands, and the stand boundary area received less litterfall (Fig. 3.5). This initially leads to a situation whereby the boundary between the two stands is a point of relative depravity as compared to the enrichment on either side.  Transects NM1 and NM2 show low litter input from alder, likely  from recent (< 5 years) top damage to the alder. All other sites show higher alder litter mass than conifer litter mass. The tests for NM1 and NM2 did not show significant mass differences between alder and conifer inputs, but still highlighted the boundary zone as an area of rapid change. Points near the stand boundary, at distances of 10, 13, and 16 m, consistently show variation from means on either side of the stand boundary. There were several differences in nutrient concentrations between alder and Douglas-fir litter (Table 3.2). Nitrogen concentration was approximately 1% higher in alder litter than in conifer litter. This resulted in substantially higher amounts of N being deposited in the alder stands compared to the Douglas-fir stands (Table 3.1, Fig. 3.6). This effect was not as visible at N M , due to the top damage described above, but transects M K 1 , MK2, CHI and CH2 showed higher N input in the alder stand and extension of this influence 5-10 metres into the conifer stand.  51  0  5  10  15  20  25  0  5  10  15  20  25  30  30  35  35  0  5  10  15  20  0  5  10  15  20  25  25  30  30  35  35  Position along transect (m) Fig. 3.5. Mass of annual input of total leaf litterfall at points along transects from alder to adjacent conifer stands. Significant differences were found for all transects except NM2 (Table 2.1).  52  5  10  15  20  25  30  10  35  15  20  25  30  35  20  25  30  35  NM2  10  15  20  25  30  35  0  5  10  15  Position along transect (m)  Fig. 3.6. Annual N input through leaf litterfall at points along transects from alder to adjacent conifer stands. Significant differences were found for all transects except NM2 (Table 3.1).  53  Differences in litterfall P concentration between the two species were small (Table 3.2) but there was higher annual P input in the alder stands (Fig. 3.7). The difference is marginal for the oldest site (NM), but substantial for sites M K and C H . Magnesium concentrations were much greater in alder litter than conifer. Litterfall M g displayed a trend similar to N and P, with greater quantities coming from alder litter than from Douglas-fir (Fig. 3.8). Increased deposition was observed 5-10 metres into the conifer stand, when M g inputs were analyzed on a point-by-point basis (Table 3.1). Magnesium cycling showed significantly higher means for the alder stand at NM1, which were not seen for total litterfall, N or P at that site. Potassium, Ca, and S, concentrations were all higher in alder than conifer foliage, resulting in greater deposition in the alder stand (Table 3.2).  Table 3.2. Foliar litter nutrient concentrations for representative alder and conifer litter from each site. Values are means of at least two composite samples of each type of litter. Transect  Nutrient Means  Litter Type %N  MK1 MK2 NM1 NM2 CHI CH2  Alder Conifer Alder Conifer Alder Conifer Alder Conifer Alder Conifer Alder Conifer  2.21 1.57 2.18 1.56 2.15 1.32 2.24 1.32 2.56 1.48 2.41 1.48  %P 0.06 0.06 0.07 0.06 0.06 0.08 0.06 0.08 0.11 0.07 0.12 0.07  %K 0.28 0.06 0.24 0.09 0.18 0.13 0.21 0.13 0.22 0.09 0.30 0.09  %Ca 0.89 0.67 1.07 0.74 1.49 1.15 1.56 1.15 1.26 0.85 1.29 0.85  %Mg 0.09 0.09 0.11 0.06 0.15 0.09 0.17 0.09 0.14 0.08 0.16 0.08  %S 0.14 0.12 0.14 0.12 0.13 0.11 0.13 0.11 0.13 0.12 0.12 0.12  Cu (ppm) 9.67 8.50 9.07 63.50 8.87 18.10 8.70 18.10 9.37 30.03 . 9.03 30.03  Fe (ppm) 252.87 3427.40 414.50 921.10 152.90 1127.80 134.33 1127.80 566.07 1825.43 987.57 1825.43  Mn (ppm) 409.50 338.10 286.07 252.30 200.37 313.70 183.30 313.70 249.17 301.37 270.93 301.37  Zn B (ppm) (ppm) 37.67 12.70 49.80 6.70 38.73 14.40 49.30 8.80 30.20 25.93 26.60 14.20 32.40 29.43 26.60 14.20 9.37 32.63 9.90 41.90 31.67 11.10 41.90 9.90  Boron showed the same trend as N , P and M g (Fig. 3.9). Considering the proposed link between B and the Frankia activity in alder (Marschner 1995), this increase is understandable. The oldest stands (NM1 and NM2) had a much higher degree of B cycling than any other nutrient, while the youngest stands (CH-1 and CH-2) seem to have less B per amount of N (Figs 3.6 and 3.9). Boron cycling, as in M g cycling, was also significantly higher in the oldest alder stands that did not show trends in total litterfall.  54  0  5  10  15  20  25  30  35  25  30  35  MK2  15  20  25  30  35  0  5  0  5  10  10  15  20  25  30  35  Position along transect (m) Fig. 3.7. Annual P input through leaf litterfall at points along the transects from alder to adjacent conifer stands. Significant differences were found for all transects except NM2 (Table 3.1).  55  0  't-  'as  littei•fall  JS  64 2  _c  Toi  0  30  35  30  35  NM2 -  10  15  20  25  Position along transect (m) Fig. 3.8. Mass of annual input of total M g through leaf litterfall at points along the transects between adjacent alder and conifer stands. Significant differences were found for all transects except NM2 (Table 3.1).  56  0.12 i  0.12  0.08  0.04  0.00 0.12  0.08  0.04  0.00  10  15  20  25  30  35  10  15  20  25  30  35  0.12  0.08  0.04  0.00  0.00  0  5  10  15  20  25  30  35  0  5  Position along transect (m)  Fig. 3.9. Annual B input through leaf litterfall at points along the transects from alder to adjacent conifer stands. Significant differences were found for all transects (Table 3.1).  57  Not all nutrients increased due to alder litter deposition. Three micronutrients had higher concentrations in conifer litter than in neighbouring alder litter. Conifer litter concentrations of Fe, Cu and Zn were between 1.5 and 15 times greater than alder litter concentrations (Table, 3.1), resulting in very high returns for these elements in the conifer stands (Figs 3.10, 3.11, 3.12). Concentrations of Zn, Fe and Cu in conifer litter were related to the age of the stands. Similar quantities of these three micronutrients were cycled in alder compared with conifer litter in the youngest stands, while much higher levels of the micronutrients were cycled in conifer litter from older stands. Concentrations of the Fe, Cu and Zn were lower in older alder litter, exaggerating the difference between the two litter types.  Macrofauna Soil macrofauna were sampled to determine whether or not macrofaunal populations differed between alder and conifers. Population counts were transformed to proportions, which are found in Table 3.3. Due to the small sample sizes and the short sampling time, the samples were grouped by conifer or alder stand. Macrofaunal proportion results were variable. The A N O V A results are summarized in Table 3.4 where significance was defined byp=0.05. A n analysis by transect was attempted with little success due to lack of replication of samples at individual points along the transects. Millipedes, centipedes, banana slugs and Aschelminthes (roundworms) were significantly more abundant in alder stands at M K . At N M , abundances of Aschelminthes and Hymenoptera were significantly higher in alder stands. At site C H , counts of oligochaete and Aschelminthes were higher in alder stands. Significant difference in abundance between sites was observed for arachnids, millipedes, slugs, molluscs, oligochaetes, carabid beetles, dipterids and orthopterids. Most flying insects (Homoptera, Heteroptera, Neuroptera, Lepidoptera and Trichoptera) were found in low numbers. Insects with very low counts were categorized as "other" and viewed only in the context of their contribution to the total counts. Diptera and Hymenoptera had substantial numbers, but the counts were highly variable, possibly due to their ability to travel great distances, and no trends were detected.  58  CH2 ^ \  UJ^-l alder 1 k W ^ l conifer ' ' total  1111111  10  15  20  25  30  35  MK2  10  15  20  25  30  35  NM2  NM1  Jiiiii 10  15  20  25  30  35  0  5  10  15  20  25  30  35  Position along transect (m) Fig. 3.10. Annual Fe input through leaf litterfall at points along the transects from alder to adjacent conifer stands. Significant differences were found for all transects (Table 3.1).  59  r  0.12  0.08  1  0.04  Q  0  0  0.00  • r r i | i ^ . r . n . i . n i i i i i i i i i i i i T T i i i i i i f i i i i i i i 0  5  10  15  2 0  25  3 0  3 5  10  15  2 0  25  3 0  35  0  5  0  5  10  15  2 0  25  3 0  3 5  0  5  10  15  2 0  25  3 0  3 5  0.12  0.08  0.04  0.00 0  5  10  15  2 0  25  3 0  3 5 0.12  0.08  0.04  0.00 0  5  10  15  2 0  25  3 0  3 5  Position along transect (m)  Fig. 3.11. Annual Cu input through leaf litterfall at points along the transects from alder to adjacent conifer stands. Significant differences were found for all transects (Table 3.1).  60  0  5  10  15  20  25  30  35  0  5  10  15  20  25  30  Position along transect (m)  Fig. 3.12. Annual Zn input through leaf litterfall at points along the transects from alder to adjacent conifer stands. Significant differences were found for all transects (Table 3.1).  61  35  Overall, the number of insects caught in traps at the oldest site (NM) was greater than those caught at either of the other sites. The M K site had the lowest abundance of insects in the pitfall traps. Transect MK1 had the lowest insect counts by individual transect, followed by transect MK2. The highest millipede counts were consistently found in the alder stands at all sites. These trends are visible for transects MK1, MK2, and NM2 (Fig. 3.13). Transect NM1, the site which had rogue alder in the conifer portion of the stand, displays a different trend whereby millipede counts were also high in the conifer stand. The C H sites displayed greatly elevated millipede counts, but no pattern could be discerned between alder and conifer areas. The highest counts for millipedes at the older sites (MK, NM) were consistently found in the alder stands. The C H site was richer than M K of N M based on initial soil variables and herbaceous cover (see site description in Chapter 1), and perhaps the millipedes respond to a threshold level of litter nutrient quality. Millipedes are known to display a preference for alder stands, if not the litter (Carcamo et al. 2000). Their preference is highlighted at Transect M K 1 , where nutrient availability in the litter becomes quite poor in the conifer stand. Millipedes displayed lower count numbers in the alder areas of N M (point 1 at transects NM-1 and -2), possibly because these points were very wet. In contrast, centipedes had higher proportions in the pure alder at N M relative to the rest of the points at those stands, suggesting that they are adapted to areas of high moisture. The centipede counts are relatively low, with a maximum count of eight at one point, but the highest counts of centipedes at the oldest four sites were in the alder stands. At C H there were no centipedes in the pure alder stand, and centipede numbers peak at the fifth transect point at transects CHI and CH2, 6 metres into the conifer stand. Variation in conifer species litter quality (Grand Fir at C H versus Douglas-fir at M K and NM) could explain the difference, as could site richness. Aschelminthes (pseudocoelomate roundworms), were found only in stands of alder or alder mixture, although they only made up a small proportion of the total population. O f all fauna observed, roundworms were the only group to demonstrate presence/ absence between alder and conifer stands, being completely absent from any area devoid of red alder.  62  Table 3.3. Pitfall trap count totals for each morphotype classified. Insect groups defined are: Arac = Arachnida, M i l = Millipedes, Cen = centipedes, B-Sl = Banana Slugs, Asch = Aschelminthes, Moll = Mollusca, Olig = Oligochaeta, Iso = Isopoda, Staph = Staphylinidae, Carab = Carahidae, Cir = Circulionidae, Silph = Silphidae, Dipt = Diptera, Hyme = Hymenoptera, Ortho = Orthoptera. Insect Groups Defined* (see full names above) Point  Arac  MK1-1 MK1-3 MK1-5 MK1-7 MK1-9 MK2-1 MK2-3 MK2-5 MK2-7 MK2-9 NM1-1 NM1-3 NM1-5 NM1-7 NM1-9 NM2-1 NM2-3 NM2-5 NM2-7 NM2-9 CH1-1 CHI-3 CHI-5 CHI-7 CHI-9 CH2-1  4 15 3 11 4 4 4 3 7 14 25 46 25 18 33 20 19 17 24 7 8 10 3 4 18  CH2-3 CH2-5 CH2-7 CH2-9  5 13 7 11  Totals  Mil Cen B-Sl Asch Moll Olig 38  8  26  10 8 4 8  1  10 1 1  46 32 16 17 8 12 26 23 23 39 37 16 24 9 16 30 55 44 80 49 80 71 75 96 36  382 1028  1 4 4 3 2 3 7 2 5 2 2 7 5 4 3 2  11 7 6 3 9 10 2 15 6 2 3 3 1  3  12 9  4 7 1 2 4 6 2  50 33 16 19 21 48 14 4 14 4  1 2  91 339  55  4 8 5 1 1 4 2 1  2  Iso Staph Carab Cir Silph  10  10  1  15 6 2 1  8 1 2  1  15 17 20 20 25 25 31 6 13 15 30 8 11 3 3 5 10 9 9 8 18 13 8 3 5  18 12 12 16 1 20 3 2 1 9 10 2 3 8 6 0 6 5 8 9 3 6 3  15 12 10 2 5 5 7 4 3 1  15 13 14 5  2  364 184 115  16 1 21 26 20 8 11 25 22 11 25 7 36 52 23 82 53 116 20 10 10 20 18 15 12  3 8 2 11 17 15 13 11 7 48 100 68 45 94 86 44 30 16 26 62 19 32 34 40 49  8 12 9 10  28 91 37 27  1 1  2 4  8 3 1 3 5 1 2 4 2 2 3 4  1 4  683 1079 51  63  Dipt Hyme Ortho Other Total  19  32  2 5 3 2 34 16 9 14 26 15 54 19 36 49 27 30 15 66 10 36 20 38 2 26 15 13 34 14 5  36 54 20 67 57 7 87 37 66 67 224 36 109 123 125 76 63 111 41 20 25 33 58 30 38  6  31 107 114 32  9 6 14 25 19 13 4 8 25 13 2 9 2 4 5 8 7 6 8 4 5 3 48  10 4 16 4 14 7 2 7 6 2 1 6 13 5 8 2 23 6 5 14 3 2 1 4 3 5 1 8 4 3  654 1926  297  189  29 4 11  5  184  8 10 6 5 10 5 8 3 28 10 38 10 18 29 20 27 7 41 18 1 2 1 4 10 10 7 64 2 8  103 176 47 150 281 152 213 168 212 281 575 259 340 454 438 360 241 400 180 254 216 237 253 223 311 205 431 315 193  415 7852  Table 3.4. A N O V A results for examination of macro fauna proportion at all sites, at individual sites and between sites. Transects at each site were considered the replicates for the site A N O V A s . ANOVA RESULTS - INSECT PROPORTIONS PER TRAP All Sites SiteMK SiteNM Site CH 1 and 28 df 1 and8 df 1 and 8 df 1 and 8 df Morphotype  F-value p-value  Arachnida ns Millipedes ns Centipedes ns Banana Slugs 9.805 Aschelminthes 10.574 Mollusca ns Oligochaeta 10.273 Isopoda ns Staphylinidae ns Carabidae ns Circulionidae ns Silphidae ns Diptera ns Hymenoptera ns Orthoptera ns Others ns  0.004 0.003 0.003  F-value p-value ns 18.23 5.23 11.16 8.64 ns ns ns ns ns ns ns ns ns ns ns  F-value p -value ns ns ns ns • 6.4 ns ns ns ns ns ns ns ns 7.12 ns ns  0.002 0.05 0.01 0.01  64  0.03  0.02  F-value p -value ns ns ns 11.788 ns ns 11.81 ns ns ns ns ns ns ns ns ns  0.009  0.009  Between sites 2 and 27 df F-value p-value 5.506 21.73 ns 6.832 ns 5.29 5 ns ns 14.76 ns ns 4.12 ns 5.49 ns  0.009 2E-06 0.004 0.012 0.01  5E-05  0.0027 0.01  0  5  10 15 20 25 30 35  0  5  10 15 20 25 30 35  Points along transect (m) Fig. 3.13. Millipede counts for all transects. Transects MK1, MK2, NM1, and NM2 show significant differences between points (Table 3.3).  65  Microbial respiration Microbial respiration rates in forest floors of alder and conifer stands were measured to see if there was a consistent increase in microbial and fungal activity in the alder stands. There was no significant difference in respiration rates from the forest floors samples collected along the transects (Fig. 3.14). However, sampling was not replicated, and several transects showed potential differences between alder and conifer forest floor respiration. These results indicate that there was no discernible difference between the effect of alder and conifer on microbial respiration. There were no trends that were visible for any of the individual sites.  66  T  3  i  i  1  1  4  5  6  7  r  8  10  Point along transect, 3 = stand boundary  Fig. 3.14. Microbial respiration in forest floor at each point along the transects. There were no significant (a=.05) differences or trends at any of the transects.  67  Discussion Healthy alder produced more litterfall than conifer stands. Analysis of litter input by Zavitkovski and Newton (1971) presented similar findings. The ability of alder litter to fall further from the tree than conifer litter creates a disparity at the boundary of the two stands, demonstrated by low levels of litterfall near the boundary between the two stands. The positioning of this low point is stand-specific, and seems to move toward the stand boundary from about 10 m inside the conifer stand as the stands mature. This may relate to changes in the distribution of conifer crowns, litterfall and restriction of the alder litter spread. Using a chronosequence of alder stands it was possible to observe the shift in primary factors limiting the ingress of alder. From height and volume of litter, which limits litterfall to a distance of approximately 15 m from the stand, to the canopy density and resistance of the neighbouring stand, which restrains litterfall to 15 m until the understory of the forest opens up with advanced pole stage in the conifers. This is a significant finding for managers wishing to make use of red alder influence in adjacent conifer stands. The resistance of an adjacent juvenile conifer stand could be likened to that of a wall. The only way for leaf litter to penetrate the adjacent conifer stand is either by being blown in from underneath or over the tops of the trees. At a certain point (in this case, around 20 years), the transition is made from a wall to an open set of columns, among which alder litter travels greater distances into the adjacent conifer stand than can occur in younger stands. Alder leaf travel varies with the age of the adjacent conifer component more than with the age of the alder component. Previous modeling of leaf litterfall assumed that the amount of litter deposition at a point from the tree declines exponentially with distance (Ferrari and Sugita 1996). My study agrees with this assumption, provided that there are no obstacles, but not with the assumption that an allometric equation can describe the relationship of D B H to foliage biomass (Ferrari and Sugita 1996). Alder litter reaches a peak and remains at high levels regardless of D B H (Zavitkovski and Newton 1971). Because alder litter production remains constant, it is possible to observe an increase in litterfall distance using a chronosequence of alder - conifer stands. This comes over time with the removal of the conifer crown as the primary obstacle in the litter's path. The ability to distinguish between the effects of different vegetation, through the use of pure stands, has been found to be difficult in the past (Gower and Son 1992). Previous work has 68  focused on the relationship between leaf litterfall N and lignin through the use of a ratio which describes the ability of various types of litter to decompose (Melillo et al. 1982; McClaugherty et al. 1985; Gower and Son 1992). The strong influence of alder litter seems to stem from the rapid decomposition of a larger quantity of litter. Decomposition rate constants (k) have shown high correlations with the ratio of lignin concentration to N concentration in the litter (Melillo et al. 1982). Alder litter can have in excess of two times the N of Douglas-fir litter, and previous lignin values for alder have been similar to those of Douglas-fir (20.8± 1.5% vs. 18.3 ± 1.0%, respectively), giving alder a high corresponding k (Edmonds 1980). Litter in red alder ecosystems shows a much faster rate of decomposition than litter in adjacent Douglas-fir stands (Edmonds 1980). The rapid mass loss of alder is short-lived, lasting only through the first year of decomposition, after which it assumes a similar rate of mass loss to Douglas-fir (Prescott et al. 2000). It is not known if the ratio of the concentration of lignin: N changes during this time, but other studies have observed the C:N ratio decrease over the first 6 months then plateau, while lignin and N both decrease steadily in content over time (Edmonds 1980). Because rapid mass loss from alder litter is a yearly event, it is possible that the increased microbial and arthropod activity associated with the annual pulse of alder litter leads to increased decomposition of the remaining litter from previous years. This would allow an increase in the total rate of decomposition for alder litter. This possibility could explain the thinner forest floors observed under alder stands in this study. The arthropod and microbial communities would hypothetically act upon the autumnal surplus, and in doing so act upon the older foliage that is mixed in. Influences of one litter type on the decomposition rate of another have been observed in laboratory studies (Fyles and Fyles 1992). The decomposition rates of alder and Douglas-fir are similar over time, yet alder visibly improves the biomass of Douglas-fir on nutrient poor sites (Miller and Reukema 1993). This is likely due to the overall increase in beneficial nutrients that alder provides. Alder litter had over 1.5 times the N of Douglas-fir litter, but also showed higher levels of all macronutrients and cations concurring with previous work (DeBell and Radwan 1984; Radwan and DeBell 1994). Furthermore, the contrasts in the levels of nutrients between alder and conifer stands were significantly different across the transects of the sites. Pitfall trapping is a well-known method of sampling arthropods, and equally well-known are the drawbacks to this technique. The drawbacks of pitfall trapping were exacerbated in this 69  study due to the short time span of the trapping, and the one-sample-per-transect-point nature of the data. Arthropod counts of such short-term nature are difficult to interpret. Arthropod populations show high variability between seasons and among years, so inferences as to the actual nature of the differences in various populations attributable to alder cannot be made with a three month sample. The soil fauna and respiration measurements were added to gain some perspective on the mechanisms behind differences in nutrient availability. Arthropods and soil microbes are critical to the turnover and cycling of nutrient contributions through litterfall (Bormann et al. 1994), yet there has been little analysis of specialized insect niches in stands of red alder in B.C. Insects are great specialists, which makes analysis of their populations difficult if knowledge of their specialization is not completely understood. Substrate specificity occurred in Aschelminthes, the pseudocoelomate roundworms. Although there were low total counts, this morphotype was only found at points at which alder was present. Worms were present throughout the transect for the mixed stand at Site NM-1, but worms were trapped only in the first one or two points (which are found in the alder stand) of the other sites. Investigations of clitellate worms have shown increased microbial activity in their intestinal systems, leading to more rapid decomposition of complex organic matter (Barois and Lavelle 1986). Earthworms seem to prefer higher quality litter, defined as less cellulose and lignin with higher nutrient concentrations, regardless of quantity (Zou 1993). This would make alder seemingly ideal forage, compared to conifer substrates. Although millipedes showed no preference between Douglas-fir and alder foliage in a microcosm study, the same study showed higher abundances of millipedes in alder stands (Carcamo et al. 2000). Increased decomposer populations would lead to faster decomposition rates due to the number of organisms involved. Further investigation into the role of both clitellate and non-clitellate earthworms is merited, as is a thorough inspection of the millipede populations between the two communities. Rate of C 0 evolution from the forest floors showed no differences that were related to 2  stand type, litter or soil. This may indicate similar microbial activity , or it may have been an artifact of the procedure used. The microbial communities may have adjusted to the moisture content and the warmth of the lab quickly, or the period in which they were stored in the lab was too long prior to testing for respiration levels. The optimum time of year for testing forest floor microbial populations would be during the early fall, when alder input is maximal and the 70  microbial community would be growing to meet the size of the nutrient pool. Results from my samples are limited because these samples were collected in mid-summer, when microbial activity is potentially limited by moisture stress. Microbial biomass can be quickly reduced to one-third of its original size through drying and rewetting of forest floor material. Flushes in mineralization as a result of rewetting do not compensate for the large reduction in C 0 and 2  mineral N production found in continuously moist samples (Pulleman and Tietema 1999). It is therefore possible that microbial activity differs between alder and Douglas-fir stands, and that the difference was obscured in the seasonal timing of the sampling and the length of time it took to process the samples. Future attempts at assessing the microbial community in alder and Douglas-fir communities should be attempted in situ, and should be conducted on a year-round basis to capture the potential differences caused by the differences in litterfall seasonality. Observations of microbial communities in German forest soils have shown that it takes high levels of input of N , P and C to elicit a response from the microbial biomass, and that these levels of input are well above realistic inputs due to litterfall (Joergensen and Scheu 1999). It is also not known whether or not the differences seen are attributable to the bacterial community or the fungal community. Further investigation is required to determine whether or not microbial biomass differs between alder and Douglas-fir stands. Microbial contribution is not always visibly pervasive. Microbial respiration is thought to be responsible for less than half of C 0 evolution in Siberian conifer forests, and 2  environmental constraints on ecosystem processes, rather than microbial metabolism characteristics, are thought to be the cause of low soil organic matter content (Ross et al. 2000). It is thus questionable if basal microbial respiration alone can be sensitive to changes in organic matter and nutrient availability due to changes in tree species. Mutualism with macrofauna and worms may significantly increase the ability of the microbial pool to act on the complex organic matter in alder. Respiratory activity in earthworm gizzards and guts can be over seven times the soil rate (Barois and Lavelle 1986). The inability to see the differences in soil respiration is due to the fact that much of the respiration occurs in the gut. Earthworm casts show only 1.69 times the respiration of the soil fauna, whereas the anterior gut of Pontoscolex can show respiration rates 7 times greater than the soil fauna alone. It is thought that this increase soon fades once mixed with soil upon departure from the worm 71  (Barois and Lavelle 1986). These possibilities, coupled with the knowledge that similar stands oiAlbizia had higher earthworm populations than neighbouring Eucalyptus, suggest that further examination of earthworms in alder and Douglas-fir stands is needed.  Conclusion Mass of leaf litterfall and the nutrient concentrations of the litter differed dramatically between the two stands. Alder litterfall is an effective measure of the alder influence on adjacent stands, and can likely be used as a simple tool to gauge the distance to which alder may influence adjacent soils. It is likely that on average, the litterfall will be able to accumulate substantial deposits up to 15 m from the stand edge. Fine litterfall is possibly a primary source of alder influence on the soil and nutrient dynamics within the alder stand and in adjacent stands. The study validated the hypothesis that there is a litter shadow around alder that extends a significant distance into the adjacent conifer stands. This dispersal was greater in the oldest stand, and much lower in the youngest stands, validating the idea that spatial distribution of alder litterfall changes rapidly in the initial stages and reaches a plateau around 15 years of age. The unforseen secondary changes in conifer morphology lead to increased dispersal in the oldest stands, leading to a corollary to the hypothesis whereby the dispersal distance is also dependent on the morphology of the adjacent stand. The study also validated the hypothesis that there was greater abundance of macrofauna in the alder stands, compared to the conifers. The hypothesis that basal respiration would be higher in the alder stands, suggesting increased microbial biomass, was not validated, but there is published evidence to suggest that the cursory examination performed here is insufficient to determine if the microbial biomass is significantly different under alder stands.  72  Chapter IV The Influence of Red Alder on Soil Chemistry and The Relationship Between Soil Chemistry and Litterfall  INTRODUCTION The presence of red alder can have a significant influence on soil, tree growth and nutrition in forest stands. Most frequently observed has been enrichment of soil N as a result of red alder N fixation (Cole et al. 1978; Binkley et al. 1994). In addition to N enrichment, alder contributes to an increase in soil organic matter (Bormann and DeBell 1981; Bormann and Sidle 1990), higher soil acidity and cation leaching (Van Miegroet et al. 1989; Homann et al. 1992). Only nitrogen and pH have been observed as phenomena that can influence soils beyond alder stand boundaries (Rhoades and Binkley 1992). While Rhoades and Binkley (1992) found that elevated mineralizable N levels persisted up to 8-12 m downslope of alder presence in the stand, they found no effect upslope. This is in contrast with the findings of Miller and Murray (1978) and Miller and Reukema (1993), who found that the alder effect on adjacent Douglas-fir extended 15m into the stand, regardless of slope. The mechanisms by which alder improves nutrient availability are poorly understood. Careful analysis of alder foliar nutrition and litterfall nutrient concentrations have been undertaken (DeBell and Radwan 1984; Radwan et al. 1984), and we know that alder litter is nutrient-rich compared to conifer litter. The nutrient benefits of alder diazotrophy are not immediately apparent in the soil. It is unclear if this lag in visible influence is due to the magnitude of nutrient cycling required to change soil nutrient regimes or if ameliorative properties of alder are delayed until stand maturity. There is little known about root litter and turnover in alder, nor has research linked litterfall input to soil nutrient regime. This is a logical linkage, as litter shadows have been shown to be spheres of influence for single trees (Zinke 1962), as well as the basis for distinct patterns in forest soils (Crampton 1982; Boettcher and Kalisz 1990). Soil carbon is thought to increase and be mixed more deeply in alder stands (Bormann et al. 1994), but few studies have proven this. Theories about alder's role in the P cycle vary. Some studies have shown that alder 73  stands are efficient at cycling P, leading to P sequestration and a lack of available P in soil (Compton and Cole 1998). Contrasting reports suggest that the presence of alder increases soil available P in mixedwood scenarios (Giardina et al. 1995). It is therefore unclear if alder improves or hampers the P cycle. In addition to divided opinions about the P cycle in alder, researchers have been critical of the tests used to examine available P in forest soils. The Bray P1 test is best adapted for forest soils, yet may not be able to access some important forms of phosphorus, leading to inaccurate measurements of the actual soil P supply (Curran 1984). Inaccuracies in P measurement using the Bray method can be exacerbated by high pH, high cation exchange capacity, or the presence of iron sesquioxides that can bind phosphorus. Another major problem with extractable P measures is that they do not necessarily represent the P available via mycorrhizae as a result of phosphatase activity or solubilization.(Cade-Menun and Lavkulich 1997) Soil pH is a factor that is important to nutrient availability, but the effect of red alder on soil pH is not fully understood. Soil pH is an important component of soil study, because acid quantity (represented as the cation exchange capacity), strength of the acid complex, and base saturation are criteria required for the understanding of nutrient availability (Binkley and Sollins 1990; Valentine and Binkley 1992; Binkley et al. 1992a). Soil pH can give insights into the driving forces behind nutrient availability, and the reasons for nutrient availability in alder stands require more study.  Objectives and Hypotheses The objective of this study was to measure the spatial extent and magnitude of red alder influence on soil. Forest floor and mineral soil were sampled at several points along a series of linear transects runningfrompure alder stands to pure Douglas-fir stands (see chapter 1 for details of experimental design). The structure of the sites remained the same throughout, although aspect differed between transects. The transects used the crown dripline boundary between alder and conifer as a point of reference for placement of the transect line, which ran perpendicular to the stand boundary (see chapter 1 for site characterization). Total N concentration, inorganic N , N 0 , N H , mineralizable N , NH :N0 " ratio, total C, C:N ratio, +  3  +  4  4  3  available P, pH in H 0 , pH in CaCl , total S (for forest floor only) were measured in each 2  2  74  sample. Based on previous studies described above, the hypotheses were: 1)  Soil variables are influenced by tree species, and will differ between a conifer and an alder stand. The disparity between the two species will result in a trend from high values for nutrients in alder to lower soil nutrients in the conifer stand. In addition, pH will be lower in alder stands.  2)  The measured soil variables will demonstrate a traceable trend from alder into the conifer stand that will gradually bring the values from their elevated alder levels down to the conifer levels. This influence on adjacent soils will be affected by stand age, slope and microtopography.  Methods Soil Collection Two series of soil samples were collected along each transect in order to gain information about gradients in the nutrient availability of the soils with distance from red alder. The first collection occurred in the spring of 1998. Between June 10 and June 15 , soils were collected th  th  at four depths ( forest floor, 0-10 cm, 10-20 cm and 20-40 cm) from each point on each transect. Large samples were taken to ensure the appropriate minimum quantity would be available for assay (samples weighed between 300 and 600 grams fresh). The samples were kept on ice, and transported immediately to U B C . A l l samples were transferred to paper bags, and began drying in a forced air oven the day they were collected, for at least 48 hours at 70°C. After drying soils were then sieved through a #20 mesh (0.5 mm) grade sieve. The soils were transferred to plastic bags and kept in a cool dry place until shipped to the analytical laboratory on July 10 , 1998. th  Soils were sampled for a second time from transects M K 1 , MK2 and NM2 between June 25 and July 4 , 1999, after viewing the results of the first soil samples. These soil samples were th  th  taken solely from the 0-10 cm depth. They were sampled at 1.5 m intervals, starting 3 m into the alder stand and ending 15 m into the Douglas-fir stand. As with the first set of soil samples, these were kept on ice during travel, dried on the day of collection, and sieved through a 0.5 mm sieve and stored in a cool dry place until shipment. The soils were delivered to the lab on August 6 , 1999. th  75  Once delivered to the B.C. Ministry of Forests analytical laboratory at the Glyn Road research station, the soils were analyzed for total C, total N , available P, pH (water and calcium chloride based solutions), total S, mineralizable N , available N H  + 4  and available N0 ". This was 3  done according to B.C. Ministry of Forests protocols (Dawson and Dunn 1998).  i  Chemical Extractions Total Nitrogen, Carbon and Sulphur Total N , C, and S concentrations were measured by dry combustion, in which a small amount of finely ground material is combusted in a tin capsule at a very high temperature. During the combustion process, virtually all compounds containing the elements of interest are decomposed and the elements released as gases. The component gases were separated by chromatography or selective sequential absorption and converted to easily measurable forms. The N is quantitatively converted to N gas, the C to C 0 and the S to S0 . The gases were 2  2  measured by their infrared absorption or thermal conductivity, depending upon which instrument is used. The concentration of gases was determined by comparison against values obtained from certified calibration standards, and the values were converted back to element concentrations in the original sample (Dawson and Dunn 1998). For the analytical procedure, the sample was dried and finely ground. Soil samples were pulverized in a ring grinder. Tissue samples (forest floor) were milled with a Wiley mill for the use in Leco CNS instruments. For the Thermo NA-1500, with its smaller sample size however, a coffee grinder was used to ensure homogeneity and fineness (Thermo Instruments 1996). The methods of analysis were performed as described in the instrument operating instructions (Thermo Instruments 1996; Leco Intruments 1996a; Leco Intruments 1996b). The methods are validated by using NIST certified reference materials. For tissue samples, NIST 1515 (Apple), 1575 (Pine), 1547 (Peach) and 1567a (Wheat Flour) were used as calibration and verification standards. For mineral soils, NIST 2740 (Buffalo River Sediment) was used. The tissue analysis methods have been verified and well tested through participation in the University of Wageningen, International Plant Exchange, inter-laboratory correlation process.  76  Available Nitrogen (NH , N0 ) +  4  3  Dried soil samples were extracted with 2N potassium chloride to displace the readily available forms of N H  + 4  and N0 ". The N H 3  + 4  and N0 " in the extracts were measured 3  colorimetrically with a Technicon DP-1000 auto-analyzer. One blank and one Forest Service control sample was included for each 22 samples. Procedures were followed in accordance with methods by (Carter 1993) and (Bremner 1965). Modifications have been made to conform to the Analytical Lab's standard extraction procedures. For example, the samples are extracted in polyethylene bags, instead of glass apparatus. Procedures for autoanalysis were followed in accordance with Technicon Industrial Method No. 334-74W/B+, "Individual/Simultaneous Determination of Nitrogen and/or Phosphorus in B D Acid Digests", and Technicon Industrial Method No. 487-77A "Nitrate and Nitrite in Soil Extracts".  Mineralizable N in Soil by Anaerobic Incubation Soil samples were incubated under anaerobic, waterlogged conditions for 2 weeks at 30°C. Soil anaerobes convert easily mineralized forms of N, both organic and inorganic, to N H . The N H +  4  + 4  is subsequently displaced from the soil by IN potassium chloride extractant and  measured colorimetrically by an auto-analyzer. This method closely follows the method developed by Dr. Tim Ballard at the University of British Columbia's Department of Soil Science (Dawson and Dunn 1998). Some minor modifications have been made to accommodate the use of consumables over permanent equipment.  Soil pH Minerals and organic compounds in soil dissociate to liberate hydronium or hydroxyl ions in solution. Calcium chloride will cause more complete dissociation by competitively driving these ions from exchange sites. The pH meter is used to measure the concentration of free hydronium ions in the resulting supernatant solution to generate a measure of the soil pH. For some soils it may be difficult to properly immerse the electrode at the suggested soil/water ratio. If necessary, the minimum extra volume required for a proper reading should be added. Small changes in solution volume will not result in significant changes in the pH reading. 77  Soil pH was measured in 10 g samples of mineral soil and 5 g samples of forest floor following the procedure used by (Kalra and Maynard 1991). Samples blanks and controls were placed in disposable bags and 10 ml H 0 or CaCl solution was added as required. The soil was 2  2  allowed to become thoroughly wetted, then well stirred with a glass rod for ten seconds. The suspensions were further stirred four or five times over the next 30 minutes. The suspensions were allowed to settle for 30 minutes, during which time the pH meter was calibrated according to the manufacturer's directions using the standard pH buffer solutions. The combination electrode was immersed into the supernatant solution so that both the bulb and salt bridge were covered in order to record the value. Stabilization of the electrode took about 60 seconds (Kalra and Maynard 1991).  Available Phosphorus Phosphate in acid soluble forms was extracted from retention sites in soil by the action of Bray PI extractant which contains ammonium fluoride and hydrochloric acid. Aluminum phosphates are broken up by NFLF, forming complexes with the metal ion. The phosphate in the extracting solution was then complexed with ammonium molybdate and antimony potassium tartrate to form a stable antimony-phospho-molybdenum blue complex that absorbs strongly at 882 nm. The phosphorus extractant was measured against a calibration curve to determine ppm phosphorus in the solution (John 1970; Kalra and Maynard 1991; Dawson and Dunn 1998).  Results Soil N concentration A summary of the results of regression analysis of each of the transects is found in Table 4.1. The number of sample points used for the regressions was one per sampling point, or n=10 for each transect. There is no discernible pattern of total N differences between alder and conifer at the youngest site (CH). Total N concentration seems to be affected by stump proximity to certain points, which adds variability to the values and makes trends more difficult to discern. The total N pattern also appears more stochastic at C H than at either the M K or N M sites. This high 78  degree of variation could be related to recent disturbance, judging by the gradual damping of this variation as the sites age from C H to M K to N M (youngest to oldest). Transect MK1 (Table 4.1, Fig. 4.1) shows a significant increase in soil N away from the alder stand. This increase is significant, shown through a quadratic trend (R =0.61,/?=0.03) in 2  the top 10 cm of soil, and linear increases in the 10-20 cm layer (R =0.58,/>=0.01), and the 20-40 2  cm layer (R =0.48,/?=0.03). This trend is thought to be due, in part, to the slope of the site (refer 2  to Fig. 1.2 in Chapter 1). Unlike the other transects sampled, MK1 slopes down and away from the alder stand. Nitrogen transport through surface or subsurface flow is likely, and this site demonstrates that a beneficial nutritional effect of alder can extend well beyond the stand boundary if the alder are upslope from a target recipient stand. The phenomenon was not seen in the 1999 data to the same extent as in 1998. The likelihood of this phenomenon  79  Table 4.1  Summary of significant quadratic and linear regression R and p-values for all 2  variables measured in the soil along transects A through F. Four depths are recorded, and a second year's sample for 0-1 Ocm depth. NS= not significant. Q; .94(0.05) = significant regression relationship between transect and soil variable is quadratic, with a multiple R value of 2  0.94 and p-value of 0.05, L = regression relationship is linear. Soil Variable  Forest Floor  Mineral 0-10 cm Mineral 10-20 cm Mineral 20-40 cm  1999 Min., 0-10 cm  ns ns ns ns ns ns Q; .77(0.005) ns ns ns ns  Q; .60(0.04) Q; .61(0.04) ns Q; .58(0.05) ns L; .41(0.05) ns Q; .94(0.00004) L; .52(0.02) ns n/a  L; .58(0.01) Q; .64(0.03) ns Q; .62(0.03) Q; .64(0.03) L; .54(0.02) ns Q; .88(0.0006) L; .63(0.006) ns n/a  L; .48(0.03) ns ns ns ns ns L; .58(0.01) ns ns ns n/a  ns ns ns ns ns ns ns ns n/a ns n/a  ns ns ns Q; .85(0.001) ns ns ns ns L; .60(0.008) L; .83(0.0002) ns  ns ns Q; .65(0.02) Q; .71(0.01) ns ns Q; .71(0.01) Q; .61(0.04) Q; .85(0.001) Q; .83(0.002) n/a  L; .53(0.02) L; .41(0.05) Q; .64(0.03) ns Q; .69(0.02) Q; .78(0.005) NS NS NS L; .45(0.03) n/a  L; .51(0.02) ns Q; .67(0.02) ns Q; .76(0.007) L; .57(0.01) ns ns Q; .60(0.04) ns n/a  ns ns ns ns L; 0.44(0.04) ns ns L; 0.53(0.02) n/a Q; 0.60(0.04) n/a  ns Q; .69(0.02) Q; .63(0.03) Q; .71(0.01) ns ns ns ns ns Q; .63(0.03) ns  Q; .84(0.002) Q; .86(0.001) Q; .86(0.001) Q; .72(0.01) Q; .91(0.00002) Q; .82(0.003) ns  ns ns ns ns ns ns Q; .72(0.01) ns Q; .64(0.03) Q; .78(0.005) n/a  L; .47(0.03) ns ns ns ns ns ns ns ns Q; .70(0.01) n/a  n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a  TRANSECT MK-1  Total N Inorganic N Ammonium Nitrate Mineralizable N Total C C:N ratio Available P pH CaCl pHH 0 Total S 2  2  T R A N S E C T MK-2  Total N Inorganic N Ammonium Nitrate Mineralizable N Total C C:N ratio Available P pH CaC12 pHH20 TotalS T R A N S E C T NM-1  Total N Inorganic N Ammonium Nitrate Mineralizable N Total C C:N ratio Available P pH CaC12 pHH20 TotalS  Q; .75(0.008) Q; .65(0.03) Q; .59(0.04) n/a  T R A N S E C T NM-2  80  ns Q; .88(0.0005) Q; .95(0.0002) Q; .90(0.0003) Q; .78(0.005) Q; .81(0.002) ns L; .40(0.05) Q; .61(0.04) Q; .92(0.0001) ns Q; .87(0.0009) L; .41(0.05) L; .46(0.03) ns Q; .96(0) ns Q; .59(0.05) ns ns ns n/a  L; .47(0.03) L; .61(0.008) L; .50(0.02) ns L; .48(0.03) ns L; .62(0.007) Q; .90(0.0004) L; 0.45(0.03) ns n/a  Q; .61(0.04) Q; .66(0.02) Q; .61(0.04) ns ns ns Q; .81(0.003) Q; .71(0.01) ns Q; .65(0.03) n/a  L; 0.42(0.04) L; 0.41(0.05) L; 0.43(0.04) ns Q; 0.70(0.02) ns Q; 0.70(0.01) Q; 0.70(0.02) n/a ns n/a  ns ns ns ns ns ns ns ns L; .59(0.01) ns ns  ns ns ns ns ns ns ns ns Q; .73(0.01) ns n/a  ns ns ns ns ns ns ns Q; .69(0.02) Q; .90(0.0003) ns n/a  ns ns ns ns ns ns ns Q; .66(0.02) ns ns n/a  n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a  Total N Inorganic N Ammonium Nitrate Mineralizable N Total C C:N ratio Available P pHCaC12  ns ns ns ns ns ns ns ns ns  ns ns ns ns ns ns ns ns ns  pHH20 TotalS  ns ns Q; .82(0.002) n/a  ns ns ns Q; .96(0.00001) ns ns ns ns ns ns n/a  ns ns ns ns ns ns ns ns ns ns n/a  n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a  Total N Inorganic N Ammonium Nitrate Mineralizable N Total C C:N ratio Available P pH CaC12 pH H20 Total S TRANSECT CH-1  Total N Inorganic N Ammonium Nitrate Mineralizable N Total C C:N ratio Available P pHCaC12 pH H20 Total S T R A N S E C T CH-2  being caused by alder influence will be discussed in the context of inorganic nitrogen later in this section. The ability to tie the influence of alder to nutritional flow with certainty is hampered by the high variability of nutrient and C conditions at MK1 and MK2.  M K 2 shows a significant  increase in N concentration in the alder stand, and in general a high level of N concentration throughout the transect. Nitrogen concentration produced significant linear regressions with  81  position along the transect at depths of 10-20 cm (R =0.63,j9=0.03) and 20-40 cm (R =0.51, 2  2  p=0.02). The need for consideration of site topography when attempting to explain and manage the alder influence is demonstrated by contrasting MK1 with MK2. The two transects, which are less than 20 m apart, display different trends that are partially due to the site features and topography. Point 10 on MK2 is in a windthrow-generated hollow, and consistently displays inflated chemical characteristics. This may be a result of microsite influence of the small depression in which the point resides. MK1, the only transect to slope down from the alder stand, shows N concentration increases from alder to conifer, a reversal of the other sites. Benefits of red alder contribution to total N concentration are more clearly defined at the older transects (NM1 and NM2). The 0-10 cm soil layer under the red alder stand contained up to three times the total N of mineral soil under the adjacent conifer stand. This was observed through strong significant regressions over the transect points. Total N concentration at NM1 was significantly higher in the alder stand and showed a quadratic decay pattern in mineral soil at 0-10 cm (R =0.84, ^=0.002), and a linear decay in the 20-40 cm layer (R =0.47, p=0.03). The 2  2  j  alder stand at NM2 also had significantly elevated N concentrations. The variable showed a quadratic decay pattern in the 0-10 cm soil layer, a linear relationship for 10-20 cm (R =0.47, 2  /?=0.03) and a quadratic relationship (R =0.42, /?=0.04) for the 20-40 cm depth. Total N declined 2  quickly as the transect approached the stand boundary at the 10 m mark. Total N levels remained slightly elevated for 3 to 9 m beyond the stand boundary and then were depressed to what is thought to be the average level for the conifer stand. NM1 displays a slightly higher total N value around the final four points along the transect in comparison to the same distances along NM2. There were some alder trees within the conifer component at NM1, giving the latter portion of the stand a heterogeneous quality. At NM2, 1999 values were very similar to 1998, and showed a weak significant linear trend of decay into the conifer stand (R =0.42, /?=0.04). 2  82  CH2  • •  •  . • •  MK1  • •  1  •  _  — •  •  t I—fa—  o  io  —:to  '  1.0  • •  0.5  1  • •  0.0  • •  • i  1.5  • 30  i  •  NM2  \ \ ••  1Q 20 Position along transect (m)  •  MK2  • . f—  '  :  •  •  ,  — — ——-  1998 1999  ^K:10 20 Position along transect (m)  30  Fig. 4.1. Nitrogen concentration in soil from 0-10 cm depth for each transect. Graphs include 1998 sampling and 1999 sampling. Lines indicate a significant regression for that series of samples. Regression results are in table 4.1.  83  Inorganic Nitrogen, N H - N and N C y N 4  Inorganic N is essential for conifer nutrition, and it can be partitioned into two primary inorganic components, N H  + 4  and N 0 \ Inorganic N is the sum of the N H 3  + 4  and N0 ~ 3  components, but the components are examined individually as well, since increased N0 " 3  presence in the soil is thought to be a trait of alder influence (Hart et al. 1997). The youngest site (CH) displays erratic inorganic N values throughout alder and conifer stands (Fig. 4.2, Table 4.1). These values may originate from decaying vegetation from previous harvests, as there is still a visible quantity of organic debris on the site in the form of tree stumps and coarse woody debris. Nitrate seems to be marginally higher in the alder stand along C H I , and significantly higher in the alder at CH2 (Fig. 4.3). There is one unusually strong trend in NCy at 10-20 cm (R =0.96,;?=0.00001). These data indicate that the alder is beginning to 2  influence the soils under the young stands, and in time the imprint may be seen further into the adjacent conifer stands. The C H site is fairly rich and high initial inorganic N values could mask the effect of alder on the conifer soil. The Malcolm Knapp (MK) transects were very different from one another. MK1 displayed questionably high inorganic N in the 1998 sampling season (Table 4.1), which showed a significant quadratic increasing value (R =0.61, p=0.05) from alder to conifer at 0-10 cm. This 2  pattern continued at 10-20 cm (R =0.64, p=0.03), but was not observed in the deepest mineral 2  layer sampled. The 1999 samples still show the maintenance of a high level of inorganic N in the MK1 conifer stand, but not the extreme values seen the year before. The data did not have a slope significantly different from 0, which still serves to show that the inorganic N regime at this site remains at similar levels well into the conifer stand. The two samples were taken at different times of the year (late April 1998, early June 1999) but MK1 was the only transect to display such a large disparity between inorganic quantities. Inorganic N along MK1 appears to display movement downslope. This is likely subsurface flow, and the high volume is in part due to the structure of the soils at Malcolm Knapp Research Forest. The study site had a compact till layer in the upper C horizon, allowing little drainage into soil below, and forcing high volumes of water to remain trapped in the upper horizons (K. Klinka, M . Feller, Pers. Comm). This perched water table is a potential reason for the highly stochastic nutrient values taken for the 1998 84  season at Malcolm Knapp. With this in mind, 1999 inorganic N for MK1 displays a more reasonable scenario, with inorganic N values peaking in the alder stand, but high values still found deep within the conifer stand. High inorganic N in the conifer stand from 12 m to the end of the transect, indicative of a high rate of mass flow into those areas, seems to correlate well with other nutrient data and N concentrations. A small hollow microsite effect may be the reason for the elevated readings for all nutrients at point 9 (30 m).  85  Fig. 4.2. Concentrations (ppm) ofl inorganic nitrogen (mg kg" ) measured at 0-10 cm depth for each transect. Graph includes 1998 sampling and 1999 sampling where available. Regression results are found in Table 4.1. 1  86  CH2  40 30 20 10 0  • •  •  MK1 (note different scale)  ^200 E ei50  MK2 40  point 9: 450.7ppm  — — •  1998 1999  30  H00  20  50 0  •  10  I  M  • •  0  NM2 40  •  30 20 10 0 10 20 Position on transect (m)  30  •  •  •  •  —  10 20 Position on transect (m)  •  •  30  Fig. 4.3. Concentrations (ppm) of soil N0 " (mg kg" dry weight) measured at 0-10 cm depth for each transect. Graph includes 1998 sampling and 1999 samples. Regression result are found in table 4.1. 1  3  87  0  10 20 Position on transect (m)  30  0  10 20 30 Position on transect (m)  Fig. 4.4. Concentrations (ppm) of soil N H (mg kg" dry weight) measured at 0-10 cm depth for each transect. Graph includes 1998 sampling and 1999 sampling. Regression results are found in table 4.1. +  1  4  88  The fractions of N0 ~ and N H 3  + 4  in the inorganic N (Figs 4.3, 4.4) tell a more detailed  story at for the 25-year-old stand at Malcolm Knapp. MK1 displays erratic values for both nutrient types, furthering the notion that the impact of alder is being distributed randomly due to intensified flow and slope effects at this transect. Nitrate shows a quadratic increase into the conifer stand (R =0.58,p=0.05) at 0-10 cm (Fig. 4.3). The relationship continues at the 10-20 2  cm depth (R =0.62,^=0.03), but there is no relationship at the deepest soil layer. MK2, with no 2  upslope effect of alder, displays a N0 " peak in the alder stand, with N 0 3  3  sloping off quickly at  first, then gradually into the conifer component (R =0.71,/?=0.01 at 0-10 cm). The N H 2  levels  + 4  along the MK2 transect appear to be higher in the conifer stand. This is a significant, although weak, quadratic relationship for the 1998 0-10 cm mineral data (R =0.65,/>=0.02), and this 2  relationship continues for soils at the 10-20 cm (R =0.64, p=0.03) and 20-40 cm (R =0.67, 2  2  p=0.02) depths. The trend is absent in the 1999 data though, which could indicate that the 1998 data is variable. The 1999 data shows a difference in the ratio of N H : N0 " (see below), which +  4  3  could be a more sensitive index of alder influence. The fact that these trends in N H  + 4  become  marginally stronger with depth indicates that alder litter may not be the primary source of inorganic N at this site. The high flow on a hardpan soil layer could be bringing N from another source, or allowing faster chemical weathering to occur under the surface, resulting in high inorganic N values beginning from the deep soil and rising to the upper layers. If this is the case, then a better gauge of the alder influence would come from N0 " values in the upper soil layer. 3  The transects at the older, poorer Nanaimo Lakes site display more clear trends in inorganic N. NM1 shows substantially elevated N levels in the alder stands (Table 4.1). At this transect, these observations are substantiated with significant decay relationships throughout the soil layers (Figs 4.2, 4.3, 4.4). Within the forest floor, inorganic N (R =0.69,p=0.02), N H 2  (R =0.63,/?=0.03), and N 0 2  + 4  (R =0.71,^=0.01) all had significant quadratic relationships along 2  3  the NM1 transect. These relationships continue in the upper (0-10 cm) mineral layer (R =0.91, 2  ^=0.0002 for inorganic N , R =0.86,p=0.001 for N H 2  + 4  and R =0.72,/?=0.01 for NC- at 0-10 cm). 2  3  To strengthen the notion that this trend begins from above with litter input, inorganic N components at NM1 showed no significant relationships below the top mineral layer. NM2 also shows strong decay relationships of an inorganic N source in alder. In forest floor at NM2, inorganic N (R =0.95, p=0.00002) and N H 2  j  + 4  (R =0.78, p=0.005) were significantly higher in the 2  j  89  alder and declined in parabolic fashion towards the conifer stand. In the 0-10 cm mineral layer, inorganic N (R =0.90,/J=0.0003) and N H 2  + 4  (R =0.82, ;?=0.003) were again high in the alder and 2  decayed into the conifer stand. These trends persisted with soil depth in this transect as linear relationships. This was seen in the 10-20 cm layer (R =0.61,/»=0.08 for inorganic N and 2  R =0.50, p=0.02 for NH ) and the 20-40 cm layer (R =0.66, p=0.02 for inorganic N and 2  +  2  4  R =0.61,/>=0.04 for NH ). The weaker correlation with increasing soil depth again supports the 2  +  4  idea that the primary source of input is litterfall. Nitrate and N H  + 4  data showed that in the wet  areas of the alder stands, N H - N levels were elevated, which could be due to year-round moisture 4  and anaerobic microbial activity. A lack of N0 " in the alder stand at NM2 was probably due to 3  inhibited nitrification in the alder stand from the very wet conditions. Nitrate was found in higher abundance at the drier points within the alder stand, and both types of inorganic nitrogen quickly dropped down to low levels in the conifer stands. Nitrate influence from the alder stand appeared to be minimal by the 6 m point in the conifer stands of NM1 and NM2, and N H  + 4  influence was minimal by 10 m into the conifers of NM2, and between 6 m and 9 m in the NM1 conifer stand. Rogue alder influence in NM1 appears to have a small effect on the N levels at points 7 through 10, but the effect is very small, and could be confounded by the natural heterogeneity of N forms in the soil.  Mineralizable (Anaerobic) Nitrogen Mineralizable N indices have often been used as an index to describe soil available N . The advantages of the anaerobic incubation technique are that it is simple to use and it correlates well with other methods. It has previously been used by (Rhoades and Binkley 1992) to observe the spatial impact of red alder on adjacent Douglas-fir stands. They determined that the influence of red alder can extend into a pure conifer stand more than 10 m past a mixed stand boundary. The youngest site (CH) showed no distinct mineralizable N that could be related to alder influences. However, the mineralizable N levels did seem to be influenced by the position of stumps relative to the transects at CHI and CH2 (Fig. 4.5). Malcolm Knapp mineralizable N data showed little evidence of alder influence (Fig. 4.5). There is one relationship between mineralizable N and MK1 in the 10-20 cm soil layer 90  (R =0.64,/?=0.03) with mineralizable N increasing into the conifer stand. There is a quadratic 2  relationship for mineralizable N and position on the MK2 transect. Mineralizable N is high in the alder and decreases into the conifer stand at 10-20 cm (R =0.69,/?=0.02) and 20-40 cm depth 2  (R =0.76,p=0.007), including the 1999 sampling of the upper mineral layer (R =0.44,p=0.04). 2  2  At MK2, the values in the center of the alder stand (point 1) are higher than point 2, but the values further into the conifer component are variable and often higher than the values in the alder stand. Peak values for mineralizable N of 147 and 130 mg/kg were observed in the conifer stands for MK1. The mineralizable N values for the upper mineral soil level at the M K site indicate no influence of alder in the neighbouring conifer stands.  It may be that the site is too  rich to show a N influence, or the N indices along MK1 maybe buoyed by alder N inputs from migration through the perched water table. The consistency of the mineralizable N values from alder to conifer are strong evidence that it is not the alder influence directly affecting the system, but a source that comes from the lower soil levels. Along MK2, the largest value for mineralizable N is along the boundary, if the very high value at the conifer end of the transect is discarded as a microsite artifact. Mineralizable N data at Nanaimo Lakes (NM) displayed clear influences of red alder on mineralizable N in the adjacent conifer stands. NM1 had a strong trend in mineralizable N , from a high value in the alder stand to a low value in the conifers. In the 0-10 cm mineral layer, there was a significant quadratic relationship (R =0.91,/?=0.0002). Typical of the oldest site, there 2  were limited trends at greater soil depths. NM2 showed a similar quadratic relationship of mineralizable N in the 0-10 cm layer (R =0.92,/?=0.0001). A linear relationship for 2  mineralizable N was seen in the 10-20 cm layer (R =0.48,;?=0.03) again from elevated alder 2  values to lower Douglas-fir values. Within the alder, N reached a peak of 362 mg/kg in the upper layers. This is assumed to be an inflated value, due to the wet nature of the soil at the transect ends inhibiting utilization. Mineralizable N drops to extremely low levels in the conifer stand. Along NM2, the conifer value for mineralizable N drops to 4.8 mg/kg. NM1 has a considerably higher base for mineralizable N of 18 mg/kg. This disparity between the two transects may be attributed to the presence of rogue alder in the conifer portion of NM1. The conifer stand along NM2 was pure. Using mineralizable N as an indicator of availability, N availability was perceptibly increased by the presence of alder. This influence extends 91  Fig. 4.5. Concentrations (ppm) of mineralizable nitrogen (mg kg" ) measured at 0-10 cm depth for each transect. Graph includes 1998 sampling and 1999 sampling. Regression results are found in Table 4.1. 1  92  approximately 9 m into the Douglas-fir stand along NM2, and 3 m into the stand along NM1, but with additions at points 7 (22 m) and 8 (25 m) due to alder ingress into the conifer stand.  Ratio analysis of NH -N:N0 -N 4  3  Nitrification increases due to alder influence have been demonstrated in past work (Hart et al. 1997), leading to a high rate of N0 " turnover in conifer stands (Stark and Hart 1997). 3  There is reason to expect a difference in the ratio of the two components of inorganic nitrogen, N0 " and N H 3  + 4  because of this effect. I therefore explored the utility of the N H : N0 " ratio as a +  4  3  tool to determine the extent to which alder influences soils adjacent to the stand boundaries (Fig. 4.6). The youngest site (CH) is relatively rich in nutrients, but also experiences some complications with detrital masses due to stumps and coarse woody debris. Along CHI, the ratio value remains below 6 up to 6 m into the fir stand (16 m). Points beyond 16 m are not within the limits of the ratio, and thus not thought to be under the influence of alder. CH2 is confounded by a stump within the alder stand (points at 7 m and 10 m), which demonstrates that even with constant alder influence, the N0 " signature is easily lost when the sample is near a large stump 3  or piece of coarse woody debris. The ratio at the 0 m and 13 m points is less than 6, indicating that the alder influence extends at least 3 m into the stand. A stump again confounds the next two points, and the remaining points exceed the expected limit value of 6. The 1998 data at MK1 is predominantly discounted from being used in the ratio due to the effects of the site on inorganic nitrogen distribution. In 1999 however, alder influence judged by the ratio guidelines is still apparent 23 m into the conifer stand. MK2 also displays alder influence much further into the stand using the ratio than when viewed with other, presumably less sensitive, N indices. The ratio remains under 6 up to 12 m into the stand, then jumps substantially higher. If the ratio is a reasonable measure of more delicate alder influence, then it suggests that alder influence is driven further into the stand. Use of the ratio analysis at the oldest site (NM) has complications due to the large number of points that could not be tested because of unusual site characteristics. Along NM1, the 10 m point is too wet, points at 10 m and 13 m are within lm proximity to a large stump, and points at 16 m, 19 m, 20 m and 25 m have N H  + 4  values below 30 ppm. 93  CH2 (note different scale) •  CHI 15  alder  40  conifer  •  •  • <  10  •  •  Z 15  MK1  fl  •  5  •  5  •  MK.2  •  .  10  • •  §  • •  15  • •  •  a  •  •  O  10  • •  0  0  |  •  20  •  5  •  30  •  10  •  •  •  •  •  •  " •  •  • •  •  •  0  0 '  NM2  NM1  15  15 •  •  10 5 0  •  • •  • •  10  •  • •  •  •  5  • •  10 20 Position on transect (m)  • •  30  •  •  1 " .  .  10 20 Position on transect (m)  ,  •  30  Fig. 4.6. Ratios of N H - N : N 0 - N from soils measured at 0-10 cm depth for each transect. Graph includes 1998 sampling and 1999 sampling. Solid circles represent 1998 sampling, solid squares are the ratios for 1999 samples. 4  3  94  This leaves only points at 7 m, 22 m, and 25 m. The 7 m point is still within the alder stand, and displays a N H  + 4  to N 0  3  ratio lower than 6. Points at 22 m and 25 m are seemingly affected by  rogue alder distribution, and also maintain ratios lower than 6. NM2 had half of the points on the transect removed from study. The point at 0 m is too wet, and points from 22 m through 35 m are below the 30 mg/kg mineralizable N threshold, making them unusable by the standards set out above. Points at 7, 10, and 13 m are below the ratio critical value of 6, while points at 16 and 19 m are above the value, indicating no substantial alder influence. This indicates that alder influence seems to be Jess influential on the steeper slopes of NM2, possibly due to retrograde flow back down towards the alder stand. It may also be that on a site with such poor N status, the N0 " signature is quickly lost without a certain level of alder input through rapid 3  transformation of N0 " into N H 3  + 4  by the detrital community. On sites with extremely low N , it  appears that mineralizable N totals give a better indication of alder influence than does the N H  + 4  :  N0 " ratio. A combination of the two methods, accompanied by a site description, appears to be 3  the best method of examining the influence of alder on soil N . After observing the fluctuation of the ratio along the transects, it appears that a ratio of approximately 6 or less may reflect a quantity of N0 " that is unusual for a conifer stand, and thus 3  is indicative of an alder influence on the soil. This ratio definition is amended by four corollaries. First, alder areas must be well aerated to acknowledge the ratio relationship. Areas that are perennially flooded are subject to high rates of anaerobic activity and the N 0 - N 3  signature of alder is lost through lack of nitrification brought on by the anaerobic environment. Second, areas that are extremely poor in available N (mineralizable N (<30 ppm) as an index) cannot be used, as minute fluctuations in the quantity of either N H  + 4  or N0 " can dramatically 3  alter the ratio in such a case. A third exemption involved areas in long term decay processes, such as samples located in close proximity (<lm) to stumps or other deposits. Decay of stumps seems to induce atypical shifts in N mineralization, which masks the effect of alder. The fourth rule in the use of the NH :N0 " ratio is that any ratio lower than 0.5 must be examined in depth. +  4  3  Exceptionally high N0 " values potentially indicate other processes in progress, which are 3  distorting the soil N ratio beyond the effects due to plant presence.  95  Total Carbon and C : N ratio The transects at the youngest site (CH) displayed highly variable C values and increases in C that seemed to be related to stump proximity to the sample point (Fig. 4.7). The site was scraped and windrowed prior to planting. The soil does not show any influence of alder on increased C, nor are there any unusual trends to explain. The lack of trends attributable to alder influence may be due to the short time scale of alder presence on the site. Alder rapidly increased organic matter in five years in Alaskan ecosystems (Van Cleve and Viereck 1972), but such rapid accumulation was not found to penetrate the mineral soil in this study. At MK1 and MK2, alder did not seem to have an appreciable effect on total C (Fig. 4.7). The trends observed did not correspond to the effect of improving C with soil depth that has been observed in previous studies. The general richness of the Malcolm Knapp site, as well as its age could be responsible for the lack of effect. MK1 showed a weak increase of C from alder into conifer in the upper soil layer (R =0.41,/?=0.05), which continued in the 10-20 cm mineral layer 2  (R =0.54, /7=0.02)(Table 4.1). 2  MK1 shows trends for total C, soil N concentration, and  inorganic N forms opposite to those expected. This anomaly is not present in the 1999 samples to the extent to which they were found in the 1998 data. The unusual 1998 data may have been a sampling error, an unusual pulse of dead plant material, or a nutrient flush that was not apparent at the time of the 1999 sampling. The point at 30 m was in close proximity to an old stump, and this could have affected the measurement of some of the soil characteristics, notably C (Fig. 4.7). The MK1 transect had a quadratic relationship for the C:N ratio in the forest floor (R =0.77, 2  p=0.006), increasing from alder to conifer. MK1 also showed a linear relationship between C:N ratio and position along the transect in the 20-40 cm depth (R =0.58,/?=0.01). Again, these 2  values suggest unusual patterns of nutrient deposition or accretion that are opposite the alder influence. MK2 had little variability in total C values along the transect for the upper soil layer (Fig. 4.7). Data from MK2 revealed relationships for total C in the two lower mineral layers studied. In the 10-20 cm layer, total C increased quadratically from the alder to the conifer stand. This was again atypical for adjacent alder and conifer stands. The elevated values for most variables at point 10 (35 m) on MK2 are due to micro- topographical influences of a small hollow generated 96  Fig. 4.7. Carbon concentrations (%) measured at 0-10 cm depth for each transect. Graph includes 1998 sampling and 1999 sampling. Regression results are found in Table 4.1  97  NM2„  •  30 25 20  •  • m 1998  15  —  a  —  1999  10 10 20 Position on transect (m)  30  0  • 10 20 Position on transect (m)  30  Fig. 4.8. C:N ratio (dry weight equivalents for total C and total N) measured at 0-10 cm depth for each transect. Graph includes 1998 sampling and 1999 sampling. Lines indicate a significant regression for that series of samples. Regression results are found in Table 4.1.  98  by a windthrow episode. The center of the alder stand did not display a similar microsite, but is found on an evenly sloping side hill. The C values along the transect describe a site that is moderately rich, and the influence of alder in adding C and nutrients to the site carry diminished importance at Malcolm Knapp. The MK2 transect also had a linear relationship for total C (R =0.58,£>=0.03) at the deepest soil depth measured. Like the quadratic relationship, it 2  increases into the conifer stand. The C:N ratio showed a significant trend, increasing into the conifer stand (R 0.71,/J=0.01) (Fig. 4.8). Because there was little change in the total C at this 2  depth, the ratio is an indicator of the lack of N input into the conifer system through aboveground litter, as compared to the alder stand. The oldest site (NM) is nutrient poor and the influence of alder on soil C is evident (Fig. 4.7, Table 4.1). Both transects have significant trends (quadratic) in soil C in the 0-10 cm horizon (R =0.82,/j=0.003 for NM1 and R =0.87 p=0.0009 for NM2). The 10-20 cm and 20-40 2  2  5j  cm depths revealed no trends in soil C in deeper soils. The first points (0 m) of each transect were difficult to interpret, as they were close to an ephemeral stream, and were seasonally saturated with water. This led to the soil having a pronounced organic layer. The C from this deep organic layer permeated the mineral soil, creating very high total C values in the upper mineral layer. The second point (7 m) was a drier, elevated point from which to interpret the alder contribution to the total C. The transition points (points at the 7 m, 10 m and 13 m marks) along NM1 all showed elevated total C values in comparison to points in the conifer portion of the transect. In a situation similar to that in MK1, total C at 13 m in NM1 was thought to be elevated due to proximity to a stump. The trend that develops at NM2 is similar to NM1; the total C values decline rapidly into the conifer stand. NM2 has lower total C values for the conifer stand than does NM1, potentially from the rogue alders in the conifer portion of NM1. Although the early-stage decay rate of alder litter is similar to the decay rate of Douglas-fir (Edmonds 1980), soil C will depend on rates of late-stage decomposition and maximum decomposition limits, which have not been compared for the two species. The alder has higher litterfall mass, which could produce higher total C values over time, if the litterfall pattern is consistent year to year. The extreme differences in N across the transects, coupled with average amounts of C in the conifer stand result in a pattern of C:N ratio trends (Fig. 4.8). At NM1 there was only one significant trend, in the 10-20 cm soil layer, where the C:N ratio increased 99  quadratically from alder to conifer (R =0.72,/?=0.01). NM2 had significant differences at greater 2  depths, with linear relationships in the forest floor (R =0.41,/?=0.05), the 0-10 cm mineral layer 2  (R =0.45,p=0.03) and the 10-20 cm mineral layer (R =0.62,/?=0.007). The 20-40 cm mineral 2  2  soil layer had a quadratic relationship (R =0.81,^=0.003), relating C:N over distance as it rises in 2  the conifer stand. This is again thought to be attributable more to the large difference in N rather than the differences in C.  Phosphorus One of the concerns related to the management of alder is the unresolved issue of P depletion. Alder is thought to access and cycle P more efficiently than neighbouring conifers (Sprent and de Faria 1988); (Compton and Cole 1998). Previous studies have addressed the low amounts of P in alder soils, mainly because of the observation that P and cations drive alder metabolism in the absence of a N requirement of the soil (Bormann et al. 1994). Separate studies have found alder to both increase soil P availability (Giardina et al. 1995) and decrease its presence in the soil through efficient nutrient cycling (Compton and Cole 1998). At the youngest site (CH), CHI showed a low range of P values, ranging between 4.75 and 8 ppm per sample, which equates to 4.75 to 8 ug/g or mg/kg of available P (Fig. 4.9; Table 4.1). In the 10-20 cm layer, there was a significant quadratic relationship between the points on CHI and available P (R =0.69,p=0.02). This trend extended to the 20-40 cm layer as well 2  (R =0.66,p=0.02). In the 0-10 cm layer, available P fluctuated, showing no significant trends. 2  The upper mineral layer was chosen for observation because it will indicate P produced by the forest floor as well as the mineral soil. At the youngest site, another contributor to P supply seems to be relic stumps from the previous harvest. A stump along CH2 exhibited a large P pulse on either side. This pulse, which pushes the levels of available P up to 40 mg/kg, is only seen at this site. High P may not have been originating from the stump per se, but from the roots decaying beneath. The transport and energy processes occurring in the roots are a sink for P in plants (Marschner 1995). CH2. exhibited elevated levels of available P across the entire transect. It was approximately double the P values of the neighbouring transect. The stumps were thought to be the most intriguing and easily explainable cause of the differences, but they lacked  100  o. ^12.5  9  MK 1  m  JlO.O  10.0  m  o.  S 7.5  Cu  •  —•— —«—  1999 1998  •  7.5  j=  % 5.0  MK 2  12.5  5.0  •  2.5  JS 2.5  •  -  0.0  < 0.0  12.5  y  NM 2  10.0 7.5 5.0 2.5  •  •  0.0 10 20 Position on transect (m)  30  10 20 Position on transect (m)  30  Fig. 4.9. Concentration of available P (mg kg" ) measured at 0-10 cm depth for each transect. Graph includes 1998 (circles) and 1999 (squares) data. Lines indicate significant regressions. Regression results are found in Table 4.1. 1  101  presence over the entire transect which would have explained the high levels of available P throughout. The intermediate-aged site (MK) showed opposing trends that seemed to be based more on slope than on the influential characteristics of alder (Fig. 4.9, Table 4.1). This disparity as it relates to slope has been discussed previously. The trends for available P at MK1 were quadratic, showing an increase into the conifer stand in the 0-10 cm mineral layer (R =0.94, 2  ^=0.0001) and in the 10-20 cm layer (R =0.8 8,^=0.0006). Lowest values for available P for the 2  transect were found in the alder, and these values gradually increased from below 5 mg/kg to above 13 mg/kg in the conifer stand. This trend would agree with (Compton and Cole 1998), who found that alder stands cycled P much more efficiently than adjacent conifer stands. The other transect at the site (MK2) had marginally higher values for available P in the alder component. The quadratic relationship showed declining available P from alder into the conifer stand for the upper soil layer (R =0.61,/>=0.03). A linear relationship showing a similar decline 2  into the conifer stand was found in the 1999 sample at 0-10 cm depth (R =0.53,/?=0.02). The 2  1999 data, which demonstrated considerable disparity in other soil data, were similar to the 1998 values, and show the same general trends for MK2. At the oldest site (NM), several trends surfaced for available P (Fig. 4.9; Table 4.1). NM1 had a significant quadratic regression, with peaks within both alder and conifer pure stands in the upper soil layer (R =0.75,^=0.008). NM2 had a similar hyperbolic trend for available P in 2  the upper soil layer which was very strong (R =0.95,/?=0.00002). Quadratic trends were also 2  visible in the 10-20 cm soils (R =0.71,^=0.01), the 20-40 cm soils (R =0.71,/?=0.01), and the 2  2  1999 data (R =0.70,/?=0.00). These points indicate that P availability in alder could be 2  dependent on moisture regime. If the site is drier, there appears to be depletion of available P in the upper and possibly the lower rooting soils. This also seems to confirm Compton and Cole's (1998) conclusion, that alder reduces available P. Available P seems to be much higher in the wetter areas of the stands, but low along the transects, and increasing at the ends of both transects.  Soil pH The acidity of the soil is important to the cation balance. High acidity can result in cation 102  leaching, making ionic balance difficult for plants (Homann et al. 1992). The production of nitric acid can lead to high levels of chemical weathering and leaching of K , C a +  2+  and M g  2+  (Homann et al. 1992). The transects at the youngest site (CH) had a high pH and little indication that the alder is beginning to depress pH in that portion of the stand (Fig. 4.10; Table 4.1). Although CHI had a significant quadratic relationship for pH in calcium chloride (R =0.59,/?=0.02 in the forest floor, 2  R =0.73,/?=0.02 in the upper soil layer, and R =0.90, JO=0.0002 in the 10-20 cm layer), it is 2  2  unlikely that this is biologically meaningful to alder. The trend is strongest at the deepest soil depth, which was similar to the unusual processes at the M K site. Because of the richness in carbon and nutrients at this site, and since rich sites have been the type to show more severe pH depression (Binkley et al. 1992a), this site could have a lower pH in the future. For this study the pH was high and relatively similar between the two components of both CHI and CH2. Measured pH in water solution for the soils at the M K site (Fig. 4.10; Table 4.1) showed moderate depression of pH for the alder stands at both transects. MK1 showed linear decreases in pH in CaCl solution from alder to Douglas-fir in the 10-20 cm layer (R =0.52,p=0.0\9) and 2  2  the 20-40 cm layer (R =0.63,/>=0.005). The trend at MK1 became more evident with soil depth. 2  The MK2 transect showed a weak significant relationship between pH in water and CaCl with 2  distance along the transect. In the forest floor, pH(H 0) (R =0.83,/?=0.0002) and pH(CaCl ) 2  2  2  (R =0.60,/>=0.008) increased, from the alder stand to the conifer stand. At the 0-10 cm depth, 2  the trends for pH(H 0) (R =0.84, ^=0.002) and pH(CaCl ) (R =0.85,^=0.001), and the 1999 2  2  2  2  pH(H 0) (R =0.60,/?=0.04) showed stronger quadratic trends in pH. In the 10-20 cm layer, only 2  2  pH (H 0) was significant (R =0.45,/?=0.03). The 20-40 cm layer showed a linear relationship 2  2  for pH (CaCl ) (R =0.60, j?=0.04). Although pH was variable at the site, it appeared as if the 2  2  alder was beginning to depress pH within the stand boundaries and slightly beyond. MK1 showed much more variability in the pH, and some of the points in the conifer stand showed lower pH than the alder stand in the 1998 data. This unusually high N0 " in the 1998 samples 3  may have caused pH depression due to nitrification. The values for pH in CaCl solution at site 1 2  (Fig. 4.11) show similarly low pH in the alder stand for MK2, and a consistently low pH for MK1. MK1 consistently shows the same trend, regardless of the test used. A n influence of alder seems to stretch evenly over the entire transect. 103  5.5  -  •  5.0 4.5  6.0  CHI  6.0 • alder  •  •  •  •  •  conifer  •  •  5.5  4.5 4.0  4.0 MK.1  •  •  4.5  •  • •  • •  •  •  •  •  •  •  NM1 •  5.5  •  5.0  •  a  4.0 6.0  •  •  -  -  MK2 •  5.5  5.0  •  •  •  6.0  5.5  o  CH2  •  •  5.0  -  6.0  -  •  •  6.0  — • —  • •  5.5  5.0  -  5.0  4.5  -  4.5  4.0  •  4.0  ' •  .A r  4.5  • —•—'  L..  -  1  NM2 • •  •  1998 1999  •  .• : s  •  • •  •  4.0 0  20 10 Position on transect (m)  30  0  10 20 Position on transect (m)  30  Fig. 4.10 pH in water paste for soil from 0-10 cm depth for each transect. Graph includes 1998 (circles) and 1999 (squares) data. Lines indicate a significant regression for that series of samples. Regression results are found in table 4.1.  104  CHI •  '  •  •  ^  •  •  •  alder  conifer  MK1  •  •  '  •  •  *  » •  10 20 Position on transect (m)  30  10 20 Position on transect (m)  30  Fig. 4.11 pH in CaCl for soil from 0-10 cm depth for each transect. Graph includes 1998 (circles) and 1999 (squares) data. Lines indicate a significant regression for that series of samples. Regression results are found in table 4.1. 2  105  A second scenario is observed at the oldest site (NM). Hypothetically, the N M site would have the greatest pH depression in the alder stand. The data however suggest that pH is highest in the alder stand and becomes progressively lower in the conifer stands (Figs 4.10, 4.11; Table 4.1). NM1 had significant quadratic trends for forest floor pH(H 0) (R =0.63,p=0.03), 0-10 cm 2  2  pH(H 0) (R =0.59,/J=0.04) and pH(CaCl ) (R =0.65,/>=0.03), 10-20 cm pH(H O) (R =0.64, 2  2  2  2  2  z  /?=0.03) and pH(CaCl ) (R =0.78, /?=0.005), and 20-40 cm depth pH(H 0) (R =0.70,p=0.01). 2  2  2  2  This reveals the importance of the moisture regime and how it affects the alder influence through nitrification and subsequent acidification and cation leaching. Lacking nitrification in waterlogged soil, the samples demonstrated pH levels that resembled stream water more closely than forest soils. This in turn elevated the cation levels, as leaching does not occur as rapidly in a neutral environment. At NM2, pH in CaCl showed trends from high to low to high pH at the 2  end of the transect. The transect showed a quadratic relationship in the upper (0-10 cm) soil layer (R =0.59,^=0.04), and in the 10-20 cm soil layer with a linear relationship (R =0.45, 2  2  7^=0.03). The pH drops with distance from the ephemeral stream, and then begins to rise along the transect. This may be due to the influence of alder in the conifer stand, which is evident until the 19 m point, where pH begins to recover. The drop in pH for the 30 m and 35 m points in the stand are thought to be unrelated to alder, but it is not known what mechanism is depressing pH at these points. Soil pH at NM1 remains at low levels well into the conifer stand. Here the small amount of rogue alder in the conifer stand may have instigated minor pH changes during the 40 years of alder presence. Due to the riparian nature of the alder at the N M site, it was not possible to obtain a base level for alder influence on pH for the alder stand. It was only possible to speculate, based on previous research, as to the degree to which the conifer soils were being affected by the adjacent alder. It is likely that pH at N M was masked by the water. When 1998 and 1999 data for NM1 were compared, there was a large degree of variation between the wetter spring 1998 sample and the drier early summer 1999 sample for the first point on the transect. If pH depression is sought as evidence of alder influence, then the best time of year to sample the alder soils would be in late summer or early fall, after the dry season. This would allow any pH depression due to nitrification to have time to develop over the summer, as the soil dries out.  106  Variability between sites and soil depths Most of the variables tested in the soil changed with soil depth (Table 4.1). Total C, total soil N concentration, C/N ratio, available P, mineralizable and inorganic N , N H  + 4  and both  measures of pH were affected significantly by depth (pO.OOl in all cases). Only N0 " did not 3  vary significantly with depth of soil. It has been demonstrated that soil C can increase with alder influence, showing higher total C levels in the soil as well as elevated C levels at greater depth in the soil. It appeared that in the oldest alder stands (NM), there was more C at the lowest depth in alder soils relative to conifer soils. There was more C overall in alder soil, when all depths were considered. Earthworm incorporation and aggregation of organic matter into deeper soil layers has been suggested as a possible mechanism in previous assessments of alder influences on soils (Bormann et al. 1994). However, assessment of C at fixed depths in this examination suggested that alder had more C at greater depth only at the oldest sites (NM). Because C loading of deeper soil is only consistent at N M , it is questionable whether there was any strong mechanism, other than time and deposition, that is incorporating C at a lower level at C H , M K , and N M . A confounding problem with the assessment of C quantity at N M was the amount of soil moisture present. Because a portion of the decomposition and incorporation of organic litter is performed by detritivores, the soil moisture may inhibit decomposition of organic matter at this site. This buildup of O M was evident in the deeper forest floor, and the incomplete decomposition of alder leaves compared to drier areas nearby. In questioning the ability to compare between the very wet alder points and conifer points along the transect, total C was not the only problematic factor. All soil variables were thought to be similarly questionable, or at least confounded by the differences in annual soil moisture regime. A N O V A models showed that available P (p<0.001), C:N ratio (p<0.001), pH(both CaCl  2  and H 0,/K0.001), total S (p=0.003) all showed significant differences between sites, but N H - N , 2  4  NO3-N, total N and total C did not vary significantly between sites (Table 4.1). Nitrate was the only variable to vary significantly across all points, irrespective of site or transect (/?=0.001). This difference is thought to stem mainly from the large values found in the 1998 data for MK1. pH measured in CaCl was the only variable to express a significant site : point interaction in the 2  forest floor (p= 0.02). 107  In the top 10 cm of mineral soil, there were significant differences between sites for C:N ratio, mineralizable N and both water and pH(CaCl ) values (pO.OOl for all variables). Available 2  P, N0 " and N H 3  + 4  were all significant at p<0.05. The sum of the two inorganic N forms was not  significantly different by site, nor were soil N concentration or total C. Total N concentration, total C and mineralizable N showed significant point and site: point interaction terms (p<0.001). Ammonium had a significant site: point interaction term as well (p<0.003). No other variables had significant interaction differences for points or the site:point interaction. The soil data displayed similar A N O V A results for the 10-20 cm and 20-40 cm soil depths, with most of the data showing significant differences between sites at the p<0.0\ level. Only N0 " at 10-20 cm depth and N H 3  + 4  at 20-40 cm depth were not significantly different by site.  Significant differences between transect points were observed for pH(CaCl ) at 10-20 cm, both 2  pH measurements at 20-40 cm depths and mineralizable N at 20-40 cm (p<0.05 for all). Significant differences for site:point interactions were observed at 10-20 cm for mineralizable N , total C, and total N (pO.Ol for all), and at 20-40 cm depth for water-based pH (p<0.05), total C and total N (p<0.01 for both). Transect-by-transect regressions to analyze each soil variable for trends were justified by variables displaying significant differences between sites, or having a site: point interaction.  Evaluating soil through litterfall To determine if the influence of alder on soil characteristics is due to alder litter, I compared litter nutrient inputs with the soil nutrient data at each transect point. Previous studies have shown that alder has both higher litter quantities and higher concentrations of desirable nutrients within that litter (Radwan et al. 1984). Calculation of the difference between the two litter types (alder and conifer) was required to determine whether any change in soil attributes could be linked to differences in litter inputs. Measurement of litterfall patterns for N, P and C were compared with the soil availability indices for N, P and C (see Chapter 3 for litterfall data) at each point along the transects to see if litter inputs explained differences in soil nutrient availability. The soil data and litter inputs correlated well in some cases, and poorly in others. The relationships were consistently exponential, and became linear with the cubic transformation of alder litterfall mass. The 108  biological meaningfulness of this transformation is logical, as the soils are affected by alder in two ways. They are affected by the volume of alder litter falling, which exhibits an exponential decay into the adjacent conifer stand in most cases. The soil is also affected not by a single year's litter, which was measured, but by the accumulation of annual deposition. These two factors roughly approximate a cubic relationship. Because litter accumulation develops with time, relationships were not expected for the younger stands. It would be difficult to link anything relating to years of litterfall, due to the lack of strong trends in the soil data. There were no correlations between litterfall and soil indicators at M K 1 , CHI or CH2. Total C in the soil correlated well with alder litterfall input at the oldest transects (Fig. 4.12). Only the Nanaimo Lakes site demonstrated a trend in soil C related to alder. The relatively strong relationships support the hypothesis that the trends in soil C are primarily due to the annual input from alder litter. Nitrogen deposition from alder litter correlated with several indices of soil N availability at N M , the oldest site. Mineralizable N , inorganic N and N0 " in soil were all significantly 3  related to the N input in alder litter. Of these three indices, N0 " was the most sensitive and 3  provided the strongest correlation (Fig. 4.13). The relationship between N 0 - N and N input due 3  to litter was sensitive enough to show the beginnings of a relationship at MK2 as well (Fig. 4.14). The large number of points showing various levels of N0 " but no alder litter are areas in 3  the conifer stand that are thought to be affected by upslope vegetation and soil properties (see chapter 3). Phosphorus input due to alder litter did not relate well to the index of available P in the soil. This is understandable considering the lack of distinct trends seen in the soil variables initially. Only one site showed a correlation between available P in soil and P input in alder litter. MK2 showed a trend in available P in 0-10 cm soil, and this trend was correlated with P input from alder litter over time (Fig. 4.16). The lack of a relationship between soil and litter at the oldest site is due to the lack of trends in the soil.  109  O.OeO  5.0e9  l.OelO  1.5elO  Total alder litterfall (kg ha" y r ) 1  Fig. 4.12  _1  2.0el0  3  The relationship between carbon concentration in the upper 10 cm of mineral soil and annual input of alder litter at the Nanaimo Lakes Site.  110  30  — : — NM1 — A — NM2  25 1 E  a. a.  ^20  R = 0.90  •s  p-value = <0.00l  2  is  c u  o o u  I  4  /  15  10  R = 0.75 2  p-value = 0.001  0 1 — I —  O.OeO  1.5e5  1.0e5  5.0e4  2.0e5  Nitrogen in alder litterfall (kg ha" y r ) 1  _1  2.5e5  3  Fig. 4.13. The relationship between N0 " concentrations in the upper 10 cm of mineral soil and annual input of alder litter at the Nanaimo Lakes site. 3  Ill  0  I  I  I  I  l  l  I  O.OeO  1.0e5  2.0e5  3.0e5  4.0e5  5.0e5  6.0e5  (Nitrogen in alder litterfall (kg ha" yr "') ) 1  3  Fig. 4.14. The linear relationship between N0 " in the upper 10 cm of mineral soil and annual input of N in alder litter at transect MK2. 3  112  10  20  30  50  40  60  (Phosphorus in alder litterfall (kg ha" yr "') ) 1  3  Fig. 4.15. The relationship between available P concentration in the upper 10 cm of mineral soil and annual input of P in alder litter at transect MK2.  113  Discussion  The results show many differences in soil chemistry between the two stand types that lead to gradients at the boundary. The general trend expected was a gradient from high values in the alder to low values in the conifer stand. This trend was observed under several circumstances.  Nitrogen The youngest site at Chilliwack (CH) was used to test the age at which the alder effect is detectable in typical soil tests. Apart from the use of the N H : N 0 ratio, there is little evidence to 4  3  suggest that alder influence is prevalent at this site. Alder influence has been observed in as little as 5 years for Sitka alder and Thinleaf alder in Alaska chronosequences (Ugolini 1968; Van Cleve et al. 1971), but these studies represent colonization of river terraces, and are poor comparative examples for examination of alder influence in mature developed soils. In previous studies the minimum age where an alder influence on soils was observed was 23 years (Binkley 1983), but this was a very infertile site. Because the Chilliwack (CH) site was richer and younger than the stands studied previously, it was unreasonable to expect any effect of alder on soils or the adjacent canopy at the site. Many of the parallels between the trends seen here and those in previous research are obscured at the M K site. The M K site is younger than many of the sites reviewed by (Binkley et al. 1994), which comprise much of the information about the effect of red alder. Respectful of the disparity in ages, MK2 shows the beginnings of parallels with previous work done in mixed stands. Cascade Head and Wind River were 20 to 30 years older than the site at Malcolm Knapp when they were most recently studied, and the fertile Cascade Head site still does not show a significant increase in N . Wind River is an extremely poor site (D. Binkley, pers. comm.), and the patterns that are visible there may not be seen at all at either C H or M K , or only vaguely with age. The study sites used in Binkley's (1983) assessment were two 23 year old stands, similar in age to Malcolm Knapp. Using soil N concentration as an indicator, in comparison to the fertile Skykomish site (Binkley 1983), both the M K and C H sites are not very fertile. Nitrogen is only beginning to show increases in the alder stand along MK2.  114  The MK1 transect is atypical in the distance the influence of alder was observed. Previous researchers have found the influence to extend up to 15m from the stand (Miller and Reukema 1993), but at MK1, alder influence extended 25 m into the adjacent conifer stand. The difference in mechanisms here seems to have to do with increased mobility of nutrients in the high subsurface flow conditions. There were hints that such a problem was present at the site throughout the experiment. The soil pits that were dug in the spring of 1998 retained water into late June and early July of that year. The inability of the soil to cope with this water is due to a perched water table. It is unlikely that the water table was simply saturated at that late time of the spring. The difference in the types of relationships found at the uppermost depth compared to the deeper mineral soils is probably due to the nature of alder's N input to the soil system. Because most input is coming from the annual pulse of alder litter, and the N is predominantly being used in the upper soil layers by macrofauna, microbes and plants, the effect of the alder in the deeper soils is residual compared to the upper layers. The 1999 values for total N at the 25-year-old Malcolm Knapp stand were similar to those from 1998, with MK1 again showing higher total N near the conifer terminus of the sampling than MK2 or NM2. The MK1 transect did not achieve the same peaks in 1999 that it had previously, but the same trend pervades the sampling, although not significant. M K 2 showed similar variability and similar high points in the alder stand and at point 10, which continued to indicate a possible microtopographical deposition zone; The results from the oldest site (NM) typify results from similar stands of alder on infertile sites (Bormann and DeBell 1981; Binkley 1983). The effect of alder on all forms of N at N M agrees with previous studies of the influence of alder on poor sites (Binkley 1983; Binkley 1992; Binkley et al. 1992a). Both transects NM1 and NM2, with problems concerning waterlogging at the first point aside, also agree with previously published information regarding distance and the influence of alder on soil N forms (Miller and Murray 1978; Rhoades and Binkley 1992; Miller and Reukema 1993). There was remarkable consistency among the many measures of N availability. This was especially visible at the infertile N M site, but it was also noted at the M K site. Concentration of soil N concentration, inorganic N , N H , N0 ", and even mineralizable N in the 0-10 cm soil +  4  3  115  consistently showed the same trends. This may be related to the N deficiency at these sites. The effect that is being generated by alder in these soils can be observed even in soil total N . The values for N concentration at the three sites are all quite low in comparison to the fertile sites studied in the past (Binkley 1983). The mineralizable N trends for M K are similar to those values observed in the rich stand at Cascade Head (Rhoades and Binkley 1992). The values are difficult to compare due to the different seasons in which they were sampled (Rhoades and Binkley (1992) sampled in November), but the sites have much more mineralizable N than the poor stand studied at Wind River. It stands to reason that Malcolm Knapp is an intermediate site, which may or may not display a distinct alder influence over time. Nanaimo Lakes showed the levels of a poor stand in the conifer portion, but had very high mineralizable N in the alder stand. This is a combination of accumulated alder influence coupled with slow cycling due to a saturated moisture regime. The values in the conifer stand are comparable with the index values given by (Rhoades and Binkley 1992), but the alder stand shows incredibly high values that exceed even the fertile alder site at Cascade Head. The use of ratios in understanding biological relationships is a widespread, but questionable practice. Empirical researchers often use ratios, while statisticians caution against their use, due to the precision and large sample sizes required to gain any statistical reliability through the use of a ratio. Such avoidance of the verification of empirical results is said to come from a substantial widening of the sampling variation of ratios, compared to that of the original variables (Jasienski and Bazzaz 1999). However problematic ratios may be for the statistician, theey are often of critical significance biologically. The use of ratios here is an attempt to define the influence of alder on adjacent conifer stands through the use of a ratio of the inorganic components of soil N . Realizing that the two components generally come from differing biological sources, the system indirectly describes a biological phenomenon through analytical measurements of available nutrients. The ratio is not a direct biological measurement, and not a biological hypothesis. It is not the intention of this work to statistically define alder influence through ratio analysis of inorganic N components. This work uses the ratio to aid in defining a limit to the extent of alder influence, so that the influence of alder in other, more subtle areas may be examined. Although the analysis with this ratio was weak, and the ratio was not robust 116  enough to serve any meaningful purpose for this study, the use of the disparity between N H  + 4  and  N0 " to increase the understanding of the influence of alder on soils remains intriguing. Further 3  examinations would be better served by analyzing these values through regression, with more points to observe, if any consistency is to be found.  Phosphorus The availability of P in alder soils is an enduring topic of concern. In most coastal B.C. soils, it represents the most limiting nutrient after N. Alder is heavily dependent on endogenous P supplies, and P-deficient alder sites are subject to premature decay and poor tree formation (Radwan and DeBell 1994). Alder uses P efficiently, and alder stands are able to access more P, and cycle a higher percentage of the total nutrient through plant biomass than neighboring conifer stands. Phosphorus testing in the soil has been the subject of many reviews (Curran 1984; Crews et al. 1995; Cade-Menun and Lavkulich 1997), and several testing methods have been recommended. In the study of alder's influence on P availability, there are contradictions between studies. One study showed that alder increases soil P availability in mixed plantations (Giardina et al. 1995). This runs counter to the assertions of a study by (Compton and Cole 1998) at the Thompson River watershed, in the same alder stands that were analyzed 30 years ago in another study by (Turner et al. 1976). Through examinations of the understory biomass, surprising amounts of P allocation to Polystichum in the understory have been noted (Turner et al. 1976). Over the ten years between 1963 and their study in 1974 (publ. 1976), sword fern density had risen to occupy over 63% of the understory area. Compton (pers. comm.) confirmed that the site still had high level of Polystichum present in the understory, and that it indeed dominated the site. The drier areas of the Nanaimo Lakes alder stand are dominated by sword fern, which succumbs to salmonberry dominance in the wetter portion of the stand. The conifer area of MK2 is also high in sword fern, which is in close proximity to points 7, 8, 9, and surrounds 10 completely. In contrast, the only transect which agrees with the findings of Giardina et al. (Giardina et al. 1995), who found that alders increase soil P, was MK2, which has very little sword fern encroachment near the alder stand sample points. The conifer component of MK1 had little understory vegetation 117  through the conifer part of the transect, but there was sword fern throughout the alder portion of the stand. Sword fern and other ferns have high concentrations of P compared to other understory vegetation (Turner et al. 1976), and sword fern litter is not preferred by first order detritivores and other decomposers (Carcamo et al. 2000). It is possible therefore that swordfern is acting as a P sink in these areas. Phosphorus which is available to the plants, but in organic forms, may not be accessible using the Bray PI technique, which does not include whole leaves and humus in the mineral soils extraction. A few studies have critiqued the abilities of several of the contemporary tests for P with mixed results (Curran 1984; Cade-Menun and Lavkulich 1997). Phosphorus depletion due to sequestration could accelerate decay of alder stand health, similar to the P regulation exhibited by tropical forest plantations (Crews et al. 1995; Ewers et al. 1996; Garcia-Montiel and Binkley 1998) and agroecosystems (Crews 1993).  Soil p H Soil pH is discussed in more detail in the next chapter, since it pertains to the availability of cations which was measured through uptake by seedlings used in a pot trial. Soils under alder stands are prone to cation leaching due to increased acidity, which in turn is attributed to nitrification under the alder (Van Miegroet et al. 1989). Acidification has been observed in almost all of the studies concerning soil chemistry under alder (Bormann and DeBell 1981; Van Miegroet et al. 1989; Binkley and Sollins 1990; Binkley et al. 1992a). Acidification can take place at virtually no change to up to one unit of pH over 50 years (Bormann et al. 1994). The effect of alder on pH can easily be hidden by factors that counteract each other, such as moving water or cation influx, making it difficult to compare to other studies. The results showed no definitive influence of alder on pH depression.  Site and Topography The first comparative studies of red alder sites showed that the nutrient status of the site influences the nature of the alder effect. Alder has a beneficial effect on nutrient poor sites, such as the plantation at the Wind River experimental station (Cole et al. 1978) or the site near Mt. Benson, Nanaimo (Binkley 1983), but not on richer sites, such as the stand in Cascade Head experimental forest, or at Skykomish (Binkley 1983). The beneficial nutrient accretion due to 118  alder at Nanaimo Lakes (NM) suggests that this site is poor enough to serve as a site that stands to benefit nutritionally from alder. It would appear that the Malcolm Knapp site, is also nutrient limited to the point where alder may contribute over time. The young site in Chilliwack had no clear response at this time, so this study is unable to discern if it could be ameliorated by alder presence. The phenomenon of extended alder influence downslope has been observed in a previous studies (Rhoades and Binkley 1992; Miller and Reukema 1993). The extent to which it is seen in this study is remarkably high. This is especially notable at MK1 where the distance to which N is increased exceeds the length of the transect. Unfortunately, this serves more as a testament to the unusual soil characteristics at M K 1 , than a revision of the distance to which alder can exert an influence on soils. The improved dispersal of alder soil amelioration downslope has the potential to be an important tool for mixedwood managers, if the desired goal is soil improvement through alder management. In comparing Nanaimo Lakes (oldest) with Malcolm Knapp (intermediate age), the variability of the N levels within the conifer component of the stands changes considerably between the two. This is thought to arise in part from the aforementioned perched water table at M K (M. Feller, K. Klinka, Pers. Comm.), which prevents downward mobility of soil water, and results in high rates of flow within the upper layers of the soil. Subsurface channeling could provide much of the variability that is seen in the M K chemical data. Nanaimo, in contrast, has a wet ephemeral area in the alder stand, and several creeks nearby, which would arguably leave the conifer component free of upslope drainage. The site also has less slope than M K .  Soil Depth During soil development, stratification and layering causes variation in soil properties with depth (Pritchett and Fisher 1987). This is a continuous process in a maturing forest, as litterfall produces an input source for several mineral nutrients. The inputs from red alder have been thought to penetrate to deeper depths than conifer litter, because of accelerated development of the soil though primary succession (Ugolini 1968). A specific contribution to increased organic matter in the soil by litter high in polyphenols and tannin, and a preference for alder litter by earthworms (Bormann et al. 1994) could lead to cycling to deeper soil levels. Fixed depth 119  sampling has been used by several studies (Binkley 1983; Brozek 1990; Bormann et al. 1994). Similar patterns showing decreased variability and content for most nutrients with increasing depth are common in forest soils from the Pacific Northwest and coastal B.C. soils. Differences between studies are most obvious in the upper 10 cm of mineral soil, which display high variability from study to study.  Relationship between litterfall and soil nutrition There are many indications in previously published research that litter influences soils. Ferrari's work (Ferrari and Sugita 1996; Ferrari 1998) was strongly influenced by the notion that individual trees can strongly influence soils in which they grow (Zinke 1962). In eastern hardwood hemlock forests, the influence of individual trees in a forest translates into a spatially heterogeneous mosaic of multiple single tree influences of multiple species (Boettcher and Kalisz 1990; Ferrari 1998). This study has observed the effects of a pure stand of alder on pure stands of conifer, with an underlying assumption that the alder stand, to some extent, was able to function as a single unit in its effect on the adjacent conifers. Instead of a single tree influence, the notion of a single pure stand influence has been adapted as a method of simplification of the two distinct ecosystems. The large amount of leaf litter within the well stocked alder stand, coupled with the notion that litterfall is the primary source of input, supports a "sphere of influence" approach to litterfall in the alder conifer transition zone. Knowing the yearly accumulations and how these change with time permits estimation of the potential impact of alder litter on soil. This impact, approximated in the results as the cube of the value of yearly alder litter input, reflects multiple years of continuous litter input directly underneath the trees. It also reflects the decline in alder litter mass with distance from the alder stand. Successful correlation of litterfall inputs with soil trends validates the hypothesis that soils are dependent on litterfall inputs over time. This method could also be applied to the other nutrient constituents of the alder litter if there were corresponding soil measurements, potentially showing a degree of success with both nutrients that were higher in alder (N, P, Mg, B) and an opposite effect for micronutrients that were intensely cycled by the conifer stands (Fe, Cu, Zn). The ability to equate litterfall input to soil characteristics using a quantification has not been attempted for many species, for the reason that few species have the ability to so strongly 120  influence their environment. Alder influences on soil properties have been extensively studied in the past (Miller and Murray 1978; Binkley et al. 1992a), as have influences of alder on soils in adjacent stands (Rhoades and Binkley 1992) and on the size of the adjacent trees (Miller and Reukema 1993). The relationship between these factors and litterfall distribution in these stands has not been previously studied. Leaf litterfall influences have been demonstrated to be important to soil N patterns on a fine spatial scale (Ferrari 1998). Areas where broadleaf litter dominate show higher nitrogen mineralization rates and percent nitrification than do adjacent conifer litter dominated areas (Ferrari 1998). My results are consistent with these interpretations of the effect of litterfall on a fine spatial scale. The history of a stand needs to be considered when examining relationships between alder litter and the properties of the soil beneath its influence. A surrogate history was used in this study, comprised of a chronosequence of stands on multiple sites. Alder tends to show little variation in its ability to grow on a given site, provided that the site has not previously been and alder stand (Compton and Cole 1989), or that it is not unreasonably poor in key nutrients. Such consistency in growth compared to conifers has been shown using site index of alder (Zavitkovski and Newton 1971). Alder achieves its maximum height and litter production early, at about ten years. Litter production shows only small variation among natural alder stands (Gessel and Turner 1974), and the differences in distance from the stand achieved by litter is likely related more closely to the density of the adjacent stand than the height of the alder stand. These factors combine to allow the use of alder of different ages from different sites to be observed as a chronosequence, in order to further our understanding of the history of alder litterfall at a site, and how the alder develops an annual litter input relationship with an adjacent stand. This history defines the reasons for trends observed in the soil.  Conclusion The influence of alder on soil of adjacent conifer stands was strongest at the oldest and most nutrient poor site. A mild influence was detected at the M K site that was approximately 25 years old, and little or no differences were seen at the C H site, which was only 12-15 years old.  121  Topography had a profound influence on the distance to which the improved soil characteristics from alder stands were carried into neighboring stands. The ability to link alder litter deposition to changes in the availability of soil nutrients will allow future analysis of the effect to concentrate on areas where deposition of alder litter is noticeable. This would allow ease in finding the suspected distance to which alder influence is observed. In addition, other factors such as initial soil richness and moisture regime, that are important in determining the ability of adjacent alder to influence conifer stands. The hypothesis that soil variables were influenced by tree species appeared to be validated in this study. In some cases this is confounded by lack of site history knowledge, but there is sufficient published proof to suggest that the differences between the alder and conifer stands does not originate in the prior soil conditions. Differences in pH were variable, leading to the suggestion that pH is not simply depressed by alder, but the changes are much more complex than can be measured with a pH test. The additional hypothesis that alder influences on the soil would be traceable into the conifer stands was also validated, with many of the stands showing trends from high carbon and nitrogen in the alder to loweer values in the conifer stands. These variables were affected by age and site topography.  122  CHAPTER V E x Situ Douglas-fir Pot Trial And Comparisons With Litterfall and Soil Chemistry  Introduction The influence of red alder in nutrient cycling has been observed in the improved conifer growth in proximity to alder (Miller and Murray 1978). The observed improvement was later delimited as being within 10 to 15 m of a stand (Rhoades and Binkley 1992; Miller and Reukema 1993). Douglas-fir growing on a site that had sustained an alder plantation previously demonstrated a 65% biomass advantage, with 45% taller stems, than similar seedlings grown in soils which had formerly been a Douglas-fir site. The seedlings grown in the former alder plantation showed an increase in N but a decrease in P, Ca and M g (Brozek 1990). It is thought that the nutritional differences and subsequent biomass increases in the seedlings are a reflection of the contribution of alder previously grown on the site. The significance of soil analysis lies in the ability to measure the nutrients and chemical qualities that allow plants to grow, quantifying their growth capability through the nutrients available to them. In addition to the soil variables measured in the previous chapter, a bioassay was conducted to ascertain whether there was a difference in the growth and tissue nutrient concentrations in Douglas-fir seedlings grown in soils gathered from the different transect points. Seedlings were grown in soils excavated from the three sites. The seedlings were analyzed for their physical attributes as well as their chemical qualities.  Nutrient concentrations after one  growing season were compared with initial values to test the hypothesis that nutrient availability to the plants varied depending on the soil they were planted in. Therefore, seedling growth and nutrient concentration would reflect the quantities found in the soil, and display similar trends to those found in the soil.  123  Methods In early April, 1998, 12 gallons of mineral soil (0-30 cm depth) were collected from each transect point at the Chilliwack, Malcolm Knapp, and Nanaimo sites (60 samples total). The samples were coarse sieved (10 mm grade) and fine sieved (1.0 mm grade) to remove coarse particulate matter. The soil was then mixed with an equal volume of perlite and left in large open buckets. Four hundred Douglas-fir seedlings (Fdc 1+0 PSB 415B) from the same provenance (seedlot #32401) were obtained from a local nursery (Pelton Nursery, Haney B.C.). The stock was removed from cold storage and left in a shaded area for three days to thaw. Seedling root systems were gently but thoroughly rinsed to remove all residue, and then transplanted to 1 gallon plastic pots (Listo Products Inc., Surrey B.C.) containing equal volume of soil mixture. Each seedling was labeled to identify the soil it was planted in and its placement in the series of six plants per soil sample type. The transplant occurred on April 29 , and the seedlings were th  given seven days to acclimatize. Any trees that yellowed or lost needles within the seven-day period were replaced. After 7 days, the trees were measured for root collar diameter, total height, and longest lateral branch. Nine trees that were not used for the pots were dried and ground up for tissue analysis to determine initial concentrations prior to planting. From May 6 to November 15 , th  th  the seedlings were watered 4 times per week, and any weeds that were found were removed immediately. In June, a second set of 2 gal. pots lined with perlite was added. The bioassay pots were placed in these larger "buffer" pots so that the roots would be unable to access any N sources on the ground through the aeration holes in the bottoms of the pots. The pots were harvested on the 15 of November, 1998. Height, length of the longest th  lateral branch, and root collar diameter were measured, and then the seedlings were pulled, the roots washed thoroughly, separated into root and shoot, and dried in paper bags in a forced air oven for 48 hours at 70°C. Once dry, the samples were weighed, then ground through a 0.05 mm size using a Wiley mill (Thomas Scientific). The shoot portion of each seedling was analyzed for N , P, K, Ca, Mg, S, Cu, Fe, Mn, Zn and B through a modified microwave acid digestion process (see Chapter 2 for analytical methods) at B.C. Ministry of Forests Glyn Road Research  124  Station (Dawson and Dunn 1998). The root samples were composited by sample, then ground and analyzed using the same methods.  Results Seedling growth The incremental height gain, change in root collar diameter (RCD), root and shoot dry weights, and the root: shoot ratio were calculated for each tree. These values were averaged for the six trees grown in each soil from each point along the transects.  This data was tested for  differences between sites, transects, points and a site: point interaction term using a fixed effects A N O V A . There were no significant differences in any of the physical attribute variables among sites, transects or points (data found in Appendix V).  Seedling nitrogen Concentration and content of nutrients were determined because discrepancies between the two values can elucidate major differences in nutrition and growth status between plants. Content and concentration were highly correlated for all macronutrients (R>0.86) and most micronutrients. Statistical analyses gave identical results in terms of significant responses, thus only the results for concentrations are given (Table 5.1). Root concentrations were also studied, and showed similar relationships to the shoots. The results presented for shoot concentrations also generally hold for roots and contents. The seedlings harvested after 1 growing season had much lower shoot concentrations of N than initial seedlings (see Table 5.2 for initial seedling nutrient concentrations) and seedlings were N-deficient to varying degrees across all sites and points. Nitrogen effects due to alder were not found at the youngest site. Although N concentrations dropped between the initial nursery seedlings and the bioassay seedlings, total N content increased in all seedlings during the pot trial when compared to the initial seedlings. As indicated by low N concentrations and chlorotic foliage, the seedlings may have been limited by N in all of the soils that were provided. Space limitation and drainage characteristics imposed by the pots in which they were grown may have contributed to the overall seedling N-deficiency.  125  Relationships between transect points and N concentrations were examined for significant differences between the point means using Tukey's Honestly Significant Difference Test (table 5.1). Several of the transects displayed significant differences between points, from higher values in the alder-origin seedlings to lower values in the conifer-origin seedlings. Table 5.1 Regression R and F-statistic for seedling nutrient concentrations found to show a significant relationship with position along the transect. Tukey's Honestly Significant Differences test shows relationships between points 1 (alder stand) through 10 (conifer stand). Point 3 is located at the dripline boundary between the two stands. Similar letters in the Tukey test represent no significant differences between the seedling concentrations at points along that transect. 2  Transect  Nutrient  Regression results R-squared F value(/>)  1  Tukey test results 2 4 3  5  6  7  8  9  10  MK1  N  0.44  6.28(0.04)  a  a  ab  ab  a  a  ab  ab  ab  b  NM1  N  0.51  8.38(0.02)  a  ac  ad  ad  b  b  b  bd  bed  bd  NM1  B p*  0.5  7.98(0.02)  a  ab  be  be  c  c  c  c  c  c  N/A  3.80(0.02)  a  a  be  be  be  c  abc  be  ab  abc  0.61 0.5  12.48(0.008) 7.90(0.02)  a a  . b bed  b c  c d  c b  c b  c b  c b  c b  c b  NM2  N B p*  N/A  3.80(0.02)  a  ab  c  c  c  be  be  be  be  be  NM2  Ca  0.72  20.31(0.002)  a  ab  ab  ab  ab  ab  ab  b  b  b  NM2  Mg  0.63  13.5(0.006)  a  be  be  be  be  be  c  be  be  NM2  S  0.65  14.56(0.005)  a  b  ab b  b  b  b  b  b  b  b  NM1 NM2 NM2  * used A N O V A for site to determine significant difference of points  126  Table 5.2 Standard ranges of adequacy for mineral nutrient concentrations in conifer shoots for bare root stock, container stock (adapted from Timmer 1991) and initial seedlings from nursery stock used in this study. Nursery stock seedling values represent means and standard errors of the mean. Nutrient  Seedlings  Units Container  Bare-root  Initial (u)  SE(u)  12-35  12-20  13.56  .13  P  2-6  1-2  3.72  .04  K  7-25  3-8  9.62  .08  Ca  2-10  2-5  2.14  .02  Mg  1-3  1-1.5  1.30  .01  S  1-3  1.2  2  .09  60-200  50-100  93.73  3.68  Mn  100-250  100-5000  138.26  2.17  Zn  30-150  10-125  45.71  .62  Cu  4-20  4-12  8.37  .35  B  20-100  10-100  18.32  .31  N  Fe  gkg  1  mg*kg"'  By the end of the growing season, N concentrations had dropped to sub-optimal levels in seedling shoots, compared to the initial optimal concentrations (Fig. 5.1, Table 5.2). Other macronutrients showed similar fluctuations to the changes in N concentrations and at times were also deficient themselves. It is not understood whether the changes in other nutrients are due to the N limitation, or actual deficiencies of these nutrients as well. There were differences in N concentrations in the seedlings along a transect, and some correlated well with N availability in the soil (see relationships between pot trial and soil). The relationships between all nutrients in the seedlings remained constant across the points on a site, suggesting that at no point was another nutrient more limiting than N . Nitrogen concentration was significantly greater in the alder stand at the oldest site (NM), and showed significant N improvement in seedlings grown in soils from the boundary point (NM2) to 6 or 7 m into the stand (NM1) (Fig. 5.1). Significantly improved N concentrations in alder stands were not found at the other transects, but there was a  127  slight increase at the beginning of MK2. The N concentrations of seedlings grown in soils from MK1 seemed to improve with distance from the alder stand and peaked at the final point in the conifer stand. MK1 was unique among the transects studied, because the alder stand was situated above the conifer stand on the slope (see Fig. 1.3 in Chapter 1 for changes in elevation along transects). The results would indicate that any alder effect that is occurring is able extend further downslope than previously thought. Previous studies have not shown a difference between the upslope and downslope influence of alder on conifers (Miller and Murray 1978). This could be attributed to differences in soil types between this study and that of Miller and Murray.  128  -  CH 1 T  -  1.1  T  11  I  %N  ~r  1.1  I1  0.6  0.6 "  1  MK 1 Alder  1.1  z s?  MK  Conifer  1.1  2  J  Z  0.6 "  5  i  0.6  NM 1  NM  1.1  1.1  E•  z 0.6  1 1  Z ±  1  0.6 0  2  10 20 30 Position along transect (m)  J  *  S a:  *  •  •  10 20 30 Position along transect (m)  Fig. 5.1 Nitrogen concentration of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  129  Phosphorus One of the concerns related to the management of alder is the unresolved issue of P depletion. Alder is thought to access and cycle P at increased and accelerated levels in comparison to neighbouring conifers (Sprent and de Faria 1988). By comparing the N and P status of the initial seedlings with the seedlings in the bioassay, this project had the capability to observe any trade-off in N and P, as it would be evident in the bioassay concentrations. Phosphorus concentration in the seedlings was correlated with N concentration (R=0.70, p<0.05 for all seedlings combined). This seems to indicate that the seedlings were displaying a concentration of P that was commensurate with the concentration of N that was present. Although P in the seedlings did not display any linear trends, they did show significant differences between transect points at the oldest site. The results (Table 5.1) demonstrate that the P values within the alder stand were significantly higher than those of the adjacent conifers. Previous studies addressed the low amounts of P under alder soils, mainly because of the observation that P and cations drive alder metabolism in the absence of a N requirement from the soil (Bormann et al. 1994). The Douglas-fir that have been grown in the alder soils demonstrate higher P concentrations at Nanaimo Lakes than any other points on those transects (Fig. 5.2). Some seedlings had P concentrations below the threshold thought to indicate sufficiency in bare-root stock conifers (Timmer 1991)(Table 5.2). These seedlings were grown in soils from Malcolm Knapp conifer stands, and more frequently in conifer soils from Nanaimo Lakes (Fig. 5.2). No seedlings grown in soil from the alder stands were insufficient in P at any of the sites. This is an important observation, especially for the trees grown in N M .  130  Conifer T"  Alder  0.2 -  *  CHI  CH 2  0.2  J  I  r  I  * o.i -  *  0.1  MK 2  0.2  $ T  a,  NM  0.2  1  0.2  NM 2  1 PH  0.1  SSO.l  0  10 20 30 Position along transect (m)  30 10 20 Position along transect (m)  Fig. 5.2. Phosphorus concentration of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  131  soils, where there were very low concentrations of both N and P. Seedlings grown in alder soils displayed elevated P levels compared to their conifer counterparts, in spite of low available P as indicated by Bray PI tests in these soils. This trend was limited to within-stand samples for P, which suggests that the uptake of P was independent of N concentrations. This disputes previous suggestions that P deficiency would threaten conifers grown on alder soils, due to the accelerated consumption and sequestration of P in alder tissues within the P cycle (Compton and ,Cole 1998).  Cations (K", C a , Mg ) 2+  2+  Potassium concentrations at all sites were sufficient, although a few individual trees showed exceptions. None of the sites displayed consistent patterns of K deficiency. Of the 9 trees that were below the sufficiency level, 7 of them were at the oldest site (NM), and the other 2 were at the intermediate site (MK)(Table 5.1). These deficiencies likely stem from more consistent and acute deficiencies of other macronutrients, such as N , which acts as a counter-ion in K uptake (Marschner 1995). Potassium levels in seedlings were not well correlated with %N. Potassium concentrations at the oldest site (NM) showed low points within the alder stand or at the boundary of the two stands (Fig. 5.3). This could be the effect of accelerated cycling of the cation due to alder metabolic demand, resulting in less available K in the soil for the new conifers. There were no significant trends relating potassium to the transect points. Calcium showed similar trends to K (Fig. 5.4). Calcium did show a significant trend at transect NM2 (R 0.72,^=0.002). The evaluation for this trend using a Tukey test showed that 2  the point in the center of the alder stand was significantly higher than the rest of the transect, although the mid-section of the transect was related to both ends. This trend might be related to acidification causing cation leaching. The acidity of the soil is important to the cation balance. High acidity can result in cation leaching, making ionic balance difficult for the plants to achieve (Homann et al. 1992). The production of nitric acid can lead to high levels of chemical weathering and leaching of K , C a +  2+  and M g  2 +  (Homann et al. 1992). The seedlings were  sufficient in Ca, with less than 5 exceptions in 360 seedlings. Calcium was also weakly correlated to N. A common thread between the results for Ca and the results for K is that the values were contained within a narrow range (±0.4%) and, this range was at the low end of  132  Alder  0.5  Ul  1  0.3  MK 0.5  0.5  Conife/^ '  0 s  0.3  1  MK 2 0.5 '  -  T  5  Ul  ^0.3  • ^ i  NM  Ul  1  ^0.3  "  1  NM 2  0.5  Ul  I  ^ 0.3 '  - T  I  II  0.5 -  L  Ul  I  ^0.3 "  ML  0  10 20 30 Position along transect (m)  ,  •  10 20 30 Position along transect (m)  Fig. 5.3. Potassium concentration of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  133  CH  CH 1  Alder 0.34  Conifer  _.  0.34 -  T  1  £^  u  0.22  • 1 -  0.22 "  £  MK 1  MK  0.34 -  0.34  0.22  £  £  £  £  ^  "  ^ *  NM 1  o  0.22 -  1 0  "  1 £  -r-  . . i s  " I  ]  NM  0.34 '  C3  CJ  £  £  T  T  x  0.22 -  5  I  0.34 "  u  £  2  X  <_>  [ L  2  I  5  2  I £  0.22 "  10 20 30 Position along transect (m)  £  0  '  10 20 30 Position along transect (m)  Fig. 5.4. Calcium concentration of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  134  their sufficiency range. Magnesium concentrations of the seedlings were often below the level considered sufficient (Fig. 5.5). Transect NM2 appeared to show some improvement in M g levels in the areas affected by alder and alder litter, as indicated by a significant trend at NM2 (R =0.63, 2  />=0.006) and the Tukey test for NM2 (Table 5.1). The Tukey Honestly Significant Differences test showed point 1 to be similar to point 3, and the rest of the site was different from them. The youngest site (CH) was more variable than either of the two more mature sites, and displayed no trend in M g deficiency. The reasons for M g deficiency are difficult to diagnose, because M g deficiencies can be induced by cation competition and low pH, apart from lack of the nutrient. The rate of M g  2+  uptake can be strongly depressed by M n  whether or not high levels of M n  2 +  2+  (Marschner 1995). It is unknown  in this study affected the M g content of the shoots, or if it was  due to other factors. Because of high values for other cation macronutrients, coupled with high Mn values (see micronutrient section below), M g concentrations could be insufficient to compete with Mn in cation uptake within the conifers.  Sulphur Sulphur concentrations in the seedlings were different among sites. At NM2, S was significantly higher in seedlings grown in soils from the center of the alder patch, compared to the other points along the transect (Fig. 5.6, Table 5.1). Sulphur is a component of several proteins, and important in many structural and metabolic functions within the plant. Sulphur displayed no trends and was generally deficient (under 0.1%) for soils from the sites, with the exception of the oldest (Nanaimo Lakes), where seedlings in the alder soils had sufficient concentrations of S. Seedlings grown in soils from NM2 (R =0.65,/»=0.005) had higher S when 2  grown in the alder soils. The Tukey test showed that only the center of the alder stand was significantly different from the rest of the stand.  135  J ±  T  1  T  Conifer  Alder  1 }  T  T -  T  i  T  T  -I'I  -  1  •  T  • • •  MK 2  0.15 -  ou . „  •-  T  i  so.io •  •  T  00  2 0.10  MK 1  0.15  CH 2  0.15  CH 1  T  •  t  -  I?  • 0.10 -  ;  L  i  i  0.15 -  i  NM 2 •  oo  2  10 20 30 Position along transect (m)  0.10 "  •  '•1*1. . 1  10 20 30 Position along transect (m)  Fig. 5.5. Magnesium concentration of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  136  T  CH 1  T •  0.12  1  CH 2  1  1 _L  T  1  ~  T"  T  _L  _L  _L  0.06  MK 1  MK 2  ~r  Conifer  Alder  iui  [  0.06  1  0.12  0.12 '  *  1  0.06  NM 1 T • _L  ~T • _L  0.12 T  '  1  I  1/3  x  ?  •  J  0.06 10 20 30 Position along transect (m)  10 20 30 Position along transect (m)  Fig. 5.6. Sulphur concentration of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  137  Micronutrient concentrations Iron was sufficient and in some cases abundant at all transects (Fig. 5.7). It is unlikely that iron had an inhibitory or toxic effect at any of the sites, based on values given by (Marschner 1995) and (Ballard and Carter 1986). There were no significant trends in iron at any sites, leading to the conclusion that alder does not exhibit a strong effect on the iron cycling at a site. Manganese was found in abundance in the seedlings at all sites (Fig. 5.8). Manganese is thought to compete with M g  2 +  and C a  2+  uptake and some metabolic reactions involving these  cations (Marschner 1995). This could play a role in Mg deficiencies that were evident at some of the sample points. There were no significant differences among sites or transect points in Mn concentration in the seedlings. Copper concentrations in seedlings suggest that supply was adequate (Fig. 5.9), and unlikely to be a problem for any of the seedlings or sites. Copper concentrations of the seedlings did not change significantly from site to site, and it is likely unrelated to the fluctuations exhibited by N and other macronutrients Zinc concentrations were sufficient in all seedlings across all sites (Fig. 5.10), and correlated with N concentration in a similar relationship to that of B (see B results). Zinc availability could cause depression of Mn and Mg in the plants, even at typical levels, but is not considered toxic until 300 mg kg (Marschner 1995). Zinc concentrations did not attain toxic -1  levels, as seen in the adequate supply of manganese found in most of the plants. Zinc showed no significant differences among sites or transect points. Boron was deficient, to some degree, at all sites (Fig. 5.11). Boron showed significant trends at Nanaimo Lakes. In these transects, B was significantly higher in seedlings grown in soils from alder stands than conifer stands (R =0.50,/J=0.02 for NM1 and R =0.50,p=0.02 for 2  2  NM2)(Table 5.1). The Tukey tests showed that there were progressions of grouped points along the transect that were related to each other, with the last six points being similar, and showed no effect of alder. Alder soils appeared to improve B status, in particular at the oldest site (NM), where B was primarily sufficient in alder soils. Boron is a metalloid nutrient whose functions remain largely unknown (Marschner 1995). Boron is known to be in high demand in diazotrophic communities, because it could play a part in the dinitrogen fixing process (Marschner 1995). 138  CHI Alder  j  CH2 T  Conifer ^  '1  T .  oo B  I  *.  ^240 • T  f  T  I  £120 -  1  1  "1 E,  T"  * 1'  120 "  fc  "  MK1 o  to  Fe (mg/kg)  I  o  i 5 5 5 5  * * *  MK2  — to  £ 120 -  1  5  S _«  a:  -SS240 "  oo 240  E, 120  PH  120 •  --  10 20 30 Position along transect (m)  10 20 30 Position along transect (m)  Fig. 5.7. Iron concentration (mg*kg~') of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. A l l transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  139  E300  I  T  I *  i  •  ; 1 o o  Mn (mg/kg)  I-  I  T CH 2  600 "  _CH 1  600  U  600  00  a 300  600 -  600  MK 1  MK 2  Conifer  Alder  00  '  i 300  i 5  *  *  :.:  *  • • . * *  «  *  NM 2  600  NM 1  * *  1  'oo  1  E 300  g 300 '  :•: - • . . • •  -J  1  10 20 30 Position along transect (m)  10 20 30 Position along transect (m)  Fig. 5.8. Manganese concentration (mg*kg') of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  140  6  CH2  CHI Alder  Conifer  i -T-  'So  r  B 4  I  i  1  T  1  - T  *4  J  i  T  5-  • ••  4  3  3  6  6  Cu (mg/kg)  MK1 'So.  T  1  V  3,  I  S ^4 "  u  6•  1  g ^4 "  1  -  S T4 -  I  O  0  •  6  NM1  "T • _1_  MK2  10 20 30 Position along transect (m)  [  5  1  NM2  5  0  10 20 30 Position along transect (m)  Fig. 5.9. Copper concentration (mg*kg"') of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  141  CHI  J  _  ' T«  34  i  :•:  ~r  1  >  I  22  N  Conifer  Alder  MK2  MK1  I  34  1 22  I  34 -  5  c  -  N  5  2  2  NM2  NM1 DO  C  N  34  22  '55 34 •  [ ^ , 1  CH2  I*  I  e N  22 •  . 4  j  I *  10 20 30 Position along transect (m)  10 20 30 Position along transect (m)  Fig. 5.10. Zinc concentration (mg*kg') of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  142  CH2  CHI  20  Conifer  Alder r  I  1  - I I  5  • «  J  I m 10  [  5  i  •  MK2  MK1 20 '  2  IE,  'DO  oa  03  0  "  E, 10 -  A.  10  :•:  NM1 B (mg/kg  I  I 10  1  o  IE  -  NM2  "T  20 -  I  20  £  « s s  .1  5 _  • r  :<  10 20 30 Position along transect (m)  10 20 30 Position along transect (m)  Fig. 5.11. Boron concentration (mg*kg"') of seedlings grown in soils harvested from 6 transects between alder and conifer stands. Points and bars represent means of six replicate seedlings and standard errors of those means. All transects are represented in the same manner, where 0 position represents the center point in the alder stand and the 10 m point is the dripline boundary between alder and conifer stands. Conifers extend from 10 m to 35 m and beyond. Points are placed at 0, 7, 10, 13, 16, 19, 22, 25, 30, and 35 m distances.  143  This gives a good reason for the elevated concentrations of B found in the alder soils. The high values for B which trail off into the conifer stands at the Nanaimo Lakes site could be a good indicator for the influence of increased diazotrophic activity in the alder stand. The use of elevated levels of B as an indicator shows that there is an influence, which elevates B up until 6 m into the conifer stand along transects NM1 and NM2, regardless of slope. Boron deficiency in the seedlings is well correlated (.70 for all seedlings sampled) with the degree to which the seedlings were N-deficient. Boron is easily leached from the soil, and loss of B from the pots is a possibility. With this in mind precautions should be taken in future bioassays to control, measure and recycle leachate, if possible.  Seedling nutrient deficiency diagnosis Nine initial seedlings were used as a standard for comparison against treated seedlings. The initial seedlings displayed healthy N concentrations in the aboveground tissue, and all other macro and micronutrients were well within suitable parameters (Table 5.2). The high levels of P may be due to luxury consumption of P in an optimal nursery setting (Timmer 1991). Because the seedlings were nursery container stock, the high P levels are perhaps applicable for the same reason. A nutrient diagnosis of the seedlings as a whole was conducted and compared to the initial nutrition of both macro and micronutrients (Tables 5.3 and 5.4). The diagnosis found that N is severely deficient in most seedlings in the bioassay. Phosphorus, M g and K were generally deficient, while Ca and most of the micronutrients were adequate. The lone exception to this statement is boron, which displayed point specific deficiencies at all of the transects. A fixed-effects A N O V A was also used to determine the effect of site on each nutrient that was measured in the seedlings. Measurement means differed uniformly at each site. This statistical technique was again used to test the similarity of points at a site and at each transect, respectively (Table 5.5). The difference among sites in nutrient status of all trees was significant in all cases (p < 0.01) except for Cu.  144  Table 5.3. Summary of univariate macronutrient deficiency diagnoses by nutrient and site. Nutrient deficiency tolerance scale was developed using Ballard and Carter (1986) foliar nutrient values, and correcting for the inclusion of the stem nutrients. Nutrient concentrations are found in chapter 4 Figs. Nutrient  Site  Diagnosis / Comments  N  CH, MK, N M  Dominantly severely deficient. Some scattered points deficient Range: 0.4% to 1.6%  P  MK  Slightly deficient with maximums adequate and minimums moderately deficient  NM  Slightly to moderately deficient with maximums adequate, minimum severely deficient  CH  Slightly deficient to adequate  K  CH, MK, N M  Moderately deficient with scattered severe deficiencies  Ca  CH, MK, N M  Little if any deficiency ranging to slight deficiency  Mg  MK  Moderate to moderately severe deficiency  NM  No deficiency to moderately severe deficiency  CH  No deficiency to moderately severe deficiency  CH, MK, N M  No deficiency to slight and moderate deficiencies  S  Table 5.4. Summary of univariate micronutrient deficiency diagnoses by nutrient and site. Nutrient deficiency tolerance scale was developed using Ballard and Carter (1986) foliar nutrient values adjusted to include stem nutrition. Nutrient concentrations by site are found in chapter 5 Figs. Nutrient  Site  Diagnosis / Comments  Fe  CH, MK, N M  Adequate  Cu  MK  Adequate to possible deficiency  NM  Adequate  CH  Adequate on 3E, Some deficient points on 3F  MK, N M  Adequate with scattered cases of possible deficiency  CH  Adequate  MK  Predominantly deficient  NM  Adequate to Deficient, point dependent along transect  CH  Adequate to deficient, no trend visible  CH, MK, N M  Adequate  Mn  B  Zn  145  Table 5.5. Nutrient concentration means by site from seedlings grown in soils originating from the sample points. Values are mean of site, either in nutrient percent per unit dry weight or nutrient parts per million units dry weight. Standard error of the mean is in parentheses. A N O V A results are all significant at a=0.05, except for copper. Superscript letters denote differences between means by the Tukey Honestly Significant Differences method. Sites NM  ANOVA /j-value F-statistic  Nutrient  Mac  N (%) P (%)  0.73(0.03) 0.12(0)  K(%) Ca (%) M g (%)  0.34(0.01) 0.21 (0.0 l) 0.08(0)  S (%)  0.08(0)  Fe (ppm)  105.88(12.76)  144.92(30.25)  0.1(0) 161.5(28.67)"  Cu (ppm)  4.1(0.18)  4.19(0.25)  4.11(0.22)  M n (ppm)  170.98(19.24)  108.8(12.02)  349.12(48.08)  B (ppm)  8.26(0.59)  Zn (ppm)  25.38(1.77)  10.92(0.78) 22.19(1.57)  a  CH  0.66(0.02)" 0.11(0) 0.33(0.02) 0.24(0.0 l )  0.9(0.05) 0.16(0.0 l ) 0.41(0.02) 0.27(0.01)  0.09(0)  0.1(0)  c  b  a  a  b  a  a  b  a  a  a  b  b  a  a  13.885 13.734  b  b  0.07(0)  a  b  b  a  a  12.810 42.016 29.182 17.889  b  a  a  b  15.392  <.001 <.001 <.001 <001 <.001 <.001 <.001  .19  .827* <.001  11.39(0.77)  58.879 7.454  30.24(2.73)  22.468  <.001  a  c  b  c  .001  Relationships between pot trial results and soil nutrient concentrations As suggested in the introduction, the ability of soil scientists to predict the growth of plants relies heavily on the ability of analytical soil tests to adequately predict the amount of available nutrient. Selected bioassay measurements were compared to several indices of nutrient availability in soils. This was done to see if there were any relationships that could be drawn between the abundance of a nutrient as predicted by a soil test and the amount of nutrient that was found in the potted seedlings after a year in the soils. The relationships occurred most frequently in the trees grown in soils from the oldest (NM) site. The N values for the alder soils were significantly higher than the conifer soils for transects MK1 (R =0.44,/?=0.04), NM1 2  (R^O.50,^0.02) and NM2 (R =0.61,^=0.008) (see Chapter 4, Table 4.1). Several soil indices 2  showed significant linear relationships with various bioassay concentrations. These relationships were not always very strong, but they indicate that some soil extracts can be correlated with seedling growth. There were significant relationships between N concentration in bioassay tissues and N concentration in soil (Fig. 5.12) for NM1 and NM2, and N0 " for NM1, NM2 and CH2 (Fig. 3  146  5.13). Shoot concentrations of B were also significantly related to N0 " for NM1 and CH2, but 3  were not significant for NM2 (Fig. 5.14). The significant relationships are based on a sample size of 10, using the average concentration of all the seedlings for each soil extract. Phosphorus concentration in the shoots and available P in soil determined using the Bray PI method only showed a significant relationship at one of the transects, NM1 (Fig. 5.15). There were no other significant linear relationships between concentrations of nutrients in the shoots and related soil tests.  Relationships between pot trial chemistry and litterfall Inputs Due to the strong relationships between soil N indices and litterfall (Chapter 4) and the previous relationships between soil N and N concentration in the shoots of Douglas-fir seedlings grown in pot trials (this chapter), a brief ancillary investigation of the possible linkages between bioassay nutrients and litterfall was performed. N , P and B in seedlings were modeled with their N , P and B concentrations in litterfall. Nitrogen (Fig. 5.16) and B (Fig. 5.17) in seedling shoots were both significantly correlated with the amount of each nutrient contributed to the ground by alder litter annually. These relationships were similar for the two oldest transects, NM1 and NM2, and they were not found at any other transects. The relationships are not as strong as some of the direct relationships between litter and soil, perhaps reflecting the degree of separation between the two independent, yet correlated, indices of nutrition at the site. These results suggest that the primary source of positive soil amendment in alder-conifer stands is nutrient input due to red alder litterfall.  147  1.0 -  — • — NM1 — • — NM2 R = 0.4058  •  2  p-value = < 0 . 0 4 7 ( ^ ^ ^ g0.9 -  •  , bo  )uglas-fii  -*-» O O to  Q •S 0.7 -  •  u  <u 00 a i—  ^ ^ *  •  <u < 0.6 -  -  - - - " " " R  • • _- _ •--•  2  = 0.4483  "  p-value = 0.0342 —  * *  0.5 " T  0.0  0.2  0.4  0.6  0.8  1.0  1.2  1.4  Total N in soil (%)  Fig. 5.12. The linear relationships between N concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and total N concentrations in the soils from transects NM1 and NM2. Lines represent significant linear relationships.  148  1.1  - NM1 - NM2 - CH2  R = 0.6032 2  p-value = 0.008  O O  R = 0.4272 2  « 0.9  p-value = 0.0404  "oo 3  O  Q  »0.7 03  > <  R = 0.6656 2  p-value = 0.004 0.5  10  15  20  25  30  Nitrate concentration (ppm)  Fig. 5.13. The linear relationship between N concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and NGy in the 0-10 depth of soil for NM1, NM2 and CH2. Lines represent significant linear relationships.  149  Fig. 5.14. The linear relationship between B concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and N0 " in the 0-10 cm depth of soil for N M 1 and CH2. Lines represent significant linear relationships. 3  150  7i  &  6  R = 0.73 p-value = 0.002  0  10  20  30  40  50  (Phosphorus in alder litterfall (kg ha" yr "')) 1  3  60 <  Fig. 5.15. The linear relationship between P concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and available P in litter for transect NM1. The line represent significant linear relationships.  151  1.2  NM1&2 MK2 CH1&2  — . — B  •  • •  A  o 1.0 o  •  •  CO  03  |  0.8  A •  A  0 s  • A  0.6  • •  '  X• y* •  A  y = 0.4823 + 0.0066x R = 0.74 2  p-value = <0.001 0.4 30  55  80  105  130  155  N deposition due to alder litter (kg ha yr" )  Fig. 5.16. The linear relationships between N concentration in shoots of Douglas-fir seedlings grown in soils excavated from transect points and the annual input N in alder litter for the three major sites (excluding MK1). Only the line representing NM1 & 2 is a significant linear relationship.  152  20  o o ^ 1  o OQ  O-  CQ  10  y = 7.2346 + 99.55x R = 0.56 2  p-value = <0.001  0.01  0.07  0.05  0.03  B deposition due to alder litter (kg ha" yr"') 1  Fig. 5.17. The relationship between B concentration in shoots of Douglas-fir seedlings and annual input alder litter for the oldest site (NM). Line indicates a significant relationships for the site.  153  Discussion The trends observed in the bioassay were similar but weaker than those examined in the soils. Nitrogen concentrations in the seedlings were closely related to the trend in the soil N data, particularly N0 ". The seedlings responded well to the N improvement due to alder, if soil 3  N0 " can be considered a surrogate for alder influence, as proposed by (Hart et al. 1997). The 3  bioassay trends paralleled the trends found in soil inorganic N forms. The seedling sensitivity to this soil N0 " is a validation of conifer response to the influence of red alder. This early response 3  would lead to later improvements in growth and yield of adjacent conifer trees (Miller and Murray 1978; Miller and Reukema 1993). The N inputs to the system, presumably from alder, are large enough that they are traceable from litterfall input to the soil, which is ameliorated by the N input over time, to the seedlings, which show some modest gains correlated to the initial input from the alder litter. This essentially takes tree influence on soils one step further, to influences on the next generation of trees. The patterns are remarkably similar to the ideas of (Zinke 1962) and (Boettcher and Kalisz 1990), that showed that single trees have a distinct influence on soils and that these trees eventually construct a mosaic of influence on the soil landscape. The influence is strong enough to be visible in the next generation of seedlings grown in those soils. Alder had a positive influence on the P concentrations within the seedlings. Transects NM1 and NM2 (and to a lesser extent, MK2) showed slightly elevated P in the alder stand. It is possible that the improvements in N concentration in the seedlings led to a parallel increase in P. MK1 showed no real change, although it is possible that the same hypothesis of flow that affected N could be applied to P. The youngest site, C H , did not show real trends. Phosphorus at M K and N M also raises the concern that P concentration is being moderated by the N deficiency in the seedling bioassay since the values are well correlated with each other. This dependence is only visible with large changes in N supply and has been demonstrated for pine by (Ingestad 1962). The trends obtained in seedling P concentration did not correlate very well with available P in soil measured using the Bray PI method. There is a possibility that available P tested with the Bray PI method is not a good measure of P availability in these soils. Although Nanaimo Lakes (NM) did show some improvement, there is no indication that P was limiting these 154  seedlings, or that alder influence improves P in the plants. Alder had either a beneficial or neutral effect on P in seedlings grown in soils from alder stands. This agrees with (Giardina et al. 1995), who showed that alder increases soil P availability. M y findings do not support the concerns raised by (Compton and Cole 1998), who found that alder stands sequestered much of the soil available P. Nor do the values in this study agree with (Brozek 1990) seedling samples, which showed declining foliar P concentrations. However, my seedlings were very young and could not account for P depletion that may occur at a later date, nor could they approximate the actual in situ planting found in Brozek's study. The question has arisen in the recent literature, as to whether or not the measures for available P truly reflect the available levels of the nutrient in the soil (Curran 1984; Cade-Menun and Lavkulich 1997). In this study, the measure of P availability used, Bray PI, did not adequately describe either the P that the seedlings were able to access, or the P inputs to the soil, in a consistent manner. This does not necessarily mean that the Bray PI test is faulty, but it does not seem to be suited to describing the P in these soils which is accessible. This result of disparity between soil, litterfall, and seedling concentrations, led to a set of soil P concentrations that did not reflect the availability of soil P to the seedlings. Soils under alder stands are prone to cation leaching due to increased acidity, which in turn is attributed to nitrification under the alder (Van Miegroet et al. 1989). Acidification has been observed in almost all studies of soil chemistry under alder (Bormann and DeBell 1981; Van Miegroet et al. 1989; Binkley and Sollins 1990; Binkley et al. 1992a). The effect of alder on pH can easily be hidden by factors that counteract each other, such as moving water or cation influx. The effect of the stream at N M on cation availability does not refute the contention that nitrification, which causes nitric acid acidification of the soils, could be a mechanism of cation leaching through increased chemical weathering. Cation availability to the seedlings through soils from this site was seemingly dependent on alder presence, but a closer look suggests that it is more likely dependent on dilution and prevention of soil acidification by water saturation. Because factors determining the degree of acidification of the soil are thought to be primarily acid strength and base saturation (Binkley and Sollins 1990), it is possible to assume that acid strength is diminished. When H is produced through nitrification, the ion presumably displaces +  the nutrient cations, which are leached from the soil. When the nitric acid is diluted by water in 155  the alder, the base saturation of the exchange complexes is lowered under alder, presumably due to the same dilution of the acid by waterlogging. Comparison of the exchangeable pools of cations found in alder soils at Wind River (Binkley and Sollins 1990; Binkley et al. 1994) with the cation uptake of the conifers in the seedlings grown at N M , illustrates that exchangeable cations in the alder stand at N M remained higher than the conifer stand. This may have been caused by dilution effect on H by waterlogging. In contrast, the richness of the site at Cascade +  Head resulted in lower exchangeable cation values in the upper soil layers of the alder-conifer stand. These values for the cations are confounded again by the uncertainty in nutrient dependence upon other nutrients for uptake in the seedlings. Cations are commonly associated with N uptake, as they are taken up with N H  + 4  to ionically balance the uptake. The lack of pH  depression does not seem to change the exchangeable ion availability for the seedlings, which agrees with previous studies (Van Miegroet et al. 1989; Binkley and Sollins 1990). Little comment can be made with regard to S, Fe, Cu, Mn and Zn because none of these variables displayed distinct trends. Most were generally adequately supplied, and were not a concern from a nutrition standpoint. There are no published studies with which to compare these data. The concentration of these micronutrients did not reflect litterfall nutrients as some of the macronutrients did, nor did they show acute deficiencies or any peaks that would suggest that their uptake is altered by the influence of red alder on the soils in which the seedlings grew. The one micronutrient about which comment should be made is boron. It appears that B is more rapidly cycled and accessible in the alder stands than in the conifer stands, resulting in higher amounts of B being found in the tissues of the Douglas-fir seedlings grown in the alder soil. Boron is thought to be of use in the actinorrhizae, in helping to fix N . This would explain 2  the high correlation between N and B in seedlings grown in alder soils. There is some indication that it could play a role in the nitrogen fixation process (Marschner 1995).  Conclusion The ability to see improvement in several foliar nutrients in seedlings grown in alder soils indicates that the increased cycling of nutrients from alder will improve the quality of the site for the next generation of conifers. Alder did appear to improve nitrogen availability to seedlings after 40-plus years of soil amendment at the Nanaimo site, and showed the beginnings of soil 156  amendment for the younger soils. It is unclear if alder has a negative effect on P availability based on this study. These improvements were seen in seedlings that were grown up to 9 m from the alder stand. The pot trial provided a valuable tool to aid in the assessment of the alder influence. The hypothesis that seedling growth varied with the soils was not validated. Seedlings did not show any significant differences in height or diameter increments. The concentrations of nutrients validated the hypothesis, showing most macronutrients and some micronutrients to have increased concentrations in seedlings grown in alder soils.  157  CHAPTER VI Discussion of the Management of Light and Nutrient Cycling in Alder/Conifer Stands, Conclusions and Recommendations for Further Research The influence of red alder could prove to be an important tool for managers interested in improving the nutrient status of sites for and during second and third rotations of trees along the B.C. coast. Red alder management alone is already a growing field for coastal foresters (Peterson et al. 1996). The next step for alder usage is management to improve conditions for existing and future conifer crops. This management scheme would reduce the costs associated with fertilization and return plantations through a more natural cycle of pioneer species (alder) to early or late successional species (Douglas-fir, hemlock, western redcedar). The intention of this project was to address several of the current problems associated with the growth of these species in mixture or in adjacency. The following is a synopsis of my findings and suggestions for further research and application of these findings. Light is an extremely important factor to consider since alder is easily able to extinguish conifers by overtopping them and outcompeting them for light. The decreased density at which it is necessary for managers to interplant with alder in order to ensure conifers will succeed in intimate mixture leads to bushy morphology in the alder (Peterson et al. 1996). This drastically reduces the commercial value of the alder in the stand. Additionally, there are complications with removal of intimately mixed alder, because the conifer component may not be ready for removal at the same time. Intimate mixture with two crop species in mind is not a practical solution. This is part of the reason why my project has focused on adjacent stands. On the southerly side of an alder patch, conifers may be planted right up to the border of the alder stand, and even a few metres into the patch, while still exceeding the light requirements needed to grow. On the north side of such alder stands, trees should be able to meet the requirement less than 15 m from the alder stand boundary. With light being addressed, the next goal of the manager is to maximize the benefit that the site receives from an alder stand. In addition to the commercial hardwood resource that can be extracted from a site, alder contributes substantial N to the site. This study has again  158  confirmed that N accretion occurs on older poorer sites, but it also demonstrates that the nature of the alder influence is both site and age specific. Careful assessment of the site is required before an alder prescription can be considered. The site needs to be deficient in N , yet have sufficient P, as this is crucial for the good growth of red alder. Nitrogen and P seem to be the most important limiting nutrients in the forest soils of coastal B.C., and care must be taken to preserve them as a resource for future growth. The management concern regarding red alder's sequestration of the P resource is not resolved here, but new directions regarding the reasons for discrepancy in study outcomes regarding P in alder stands (Giardina et al. 1995; Compton and Cole 1998) have been suggested. It seems unlikely that a tree that increases soil fauna activity would be able to sequester large amounts of P. The influence of red alder stems primarily from the litterfall that it generates. Once the nutrients are transferred to the soil, they become subject to the influences that dictate the availability and flow of soil nutrients. These include topography, moisture regime, soil texture, soil structure, and the parent material, to name only a few. Sequestration of organic matter in forest floor does not seem to occur to a great degree, since forest floors under alder are thin. Management of the beneficial influence of alder may be defined based on the influence of its litter shadow. Alder planted upslope of conifers will probably exert an influence that increases in distance downslope over time while it may have little or no upslope influence. Alder alters the chemical properties of a site significantly over time. This influence can cause an increase in growth of adjacent conifers (Miller and Murray 1978; Miller and Reukema 1993) up to 15 m into the adjacent conifer stand (Rhoades and Binkley 1992) and it is enhanced if the alder is planted upslope of a conifer stand (this study). Alder influence can also aid the next generation of conifer planted on the site, yielding a distinct advantage over conifers grown in conifer-origin soils (Brozek 1990; Compton et al. 1997). These are attractive options for a manager interested in improving the productivity of the land. Provided that the site is suitable for improvement through alder nitrogen accretion, it can improve both current and future conifer productivity and add a short rotation alder harvest.  159  Recommendations for future research in alder-conifer stand dynamics 1)  I recommend that light levels adjacent to dense alder stands be measured so that the  buffer distances around such a patch can be more exactly understood. This is an important next step leading to the resolution of perhaps the most serious concern regarding the mixedwood forestry of alder and conifers, that of light competition inhibiting free growing of the conifer. 2)  The other important competitive influence that was not covered by this thesis was the  ability of alder to control the water resource in a stand. While adjacent planting will probably limit this competitive influence, it needs to be understood in greater detail if a buffer zone between the two stands is to be precisely identified for planting purposes. 3)  Analysis of a series of stands of different ages in the same area would allow us to more  accurately pinpoint the time when alder influence begins to display a visible effect on the soils in which it resides. Although this study confirms visible alder influence at age 25, more precise characterization would help managers solidify a timeline for their operational goals of fertilization and alder harvest. 4)  M y results show that a number of factors affect the relationship between red alder and P  acquisition and sequestration. Discussions with other researchers led me to suggest that perhaps the understory plays a larger role in P sequestration than we have previously acknowledged. In order to truly understand the P cycle and how it changes in the alder stand, a review of the techniques for assessing P is required. It is important that we learn what forms of P are actually available to the conifers in a stand, and what forms of P are present in the soil and in the organic matter of a stand. It is possible that P is being sequestered in organic material that is not desirable by decomposers (such as Polystichum munitum). Because red alder seems to increase the microbial and decomposer community in the stand, it appears that red alder P sequestration is not as serious as we have previously thought, but the understory brought on by aging alder is the sink for P. Microbial communities did not show any differences in respiration in small samples, but the true test of the difference between the two stands lies in an in situ examination of the respiration between two adjacent stands. Likewise a year-long and more extensive survey of the macrofauna community may also lead to further understanding of the accelerated nutrient cycle in alder stands. 5)  The link between nitrogen and boron is intriguing. It would be interesting to examine 160  why boron and nitrogen seem to be so closely correlated in the alder nutritional cycle. These are only a small selection of the several questions that could be investigated in the future to improve the ability of managers and scientists alike to understand the life cycle of red alder and its influence. Management of red alder as a silvicultural tool as well as a simple commercial crop seems feasible from a stand dynamic standpoint. Taking necessary steps in research will allow the aesthetic, economic and nutritional benefits of alder to be added to the tools that a forest manager can use.  161  Literature Cited 1. Ballard, T. M . and Carter, R. E. 1986. Evaluating forest stand nutrient status. B C Ministry of Forests, Victoria. Land Management Report No. 20. 2. Barois, I. and Lavelle, P. 1986. Changes in respiration rate and some physiochemical properties of a tropical soil during transit through Pontoscolex corethrurus (Glossoscoleicidae, Oligochaetd). Soil Biol. Biochem. 18: 539-541. 3. Bauhus, J., Pare, D., and Cote, L. 1998. Effects of tree species, stand age and soil type on soil microbial biomass and its activity in a southern boreal forest. Soil Biol. Biochem. 30: 1077-1089. 4. Bazzaz, F.A. 1979. The physiological ecology of plant succession. Ann. Rev. Ecol. Syst. 10: 351-371. 5. Beck, M . A . and Elsenbeer, H. 1999. Biogeochemical cycles of soil phosphorus in southern alpine spodosols. Geoderma 91: 249-260. 6. Binkley, D. 1983. Ecosystem production in Douglas-fir plantations: interaction of red alder and site fertility. For. Ecol. Manage. 5: 215-227. 7. Binkley, D. 1992. Mixtures of N -fixing and non-N -fixing tree species. In The ecology of mixed species stands of trees. Edited by M.G.R. Cannell, D.C. Malcom, and P. Robertson. Blackwell Scientific Publishing, Boston pp. 99-123. 2  2  8. Binkley, D. 1997. Bioassays fo the influence of Eucalyptus saligna and Albizia falcataria on soil nutrient suppply and limitation. For. Ecol. Manage. 91: 229-234. 9. Binkley, D., Cromack Jr, K., and Baker, D.D. 1994. Nitrogen fixation by red alder: biology, rates, and controls. In The biology and management of red alder. Edited by D.E. Hibbs, D.S. DeBell, and R.F. Tarrant. OSU press, Corvallis pp. 57-72. 10. Binkley, D., Dunkin, K.A., DeBell, D.S., and Ryan, M . G . 1992b. Production and nutrient cycling in mixed plantation of Eucalyptus and Albizia in Hawaii. For. Sci. 38: 393-408. 11. Binkley, D., Giardina, C , and Bashkin, M . A . 2000. Soil phosphorus pools and suppply under the influence of Eucalyptus saligna and nitrogen-fixing Albizia falcataria. For. Ecol. Manage. 128: 241-247. 12. Binkley, D. and Sollins, P. 1990. Factors determining differences in soil pH in adjacent conifer and alder-conifer stands. Soil Sci. Soc. Am. J. 54: 1427-1433. 13. Binkley, D., Sollins, P., Bell, R., Sachs, D., and Myrold, D. 1992a. Biogeochemistry of adjacent conifer and alder-conifer stands. Ecology 73: 2022-2033. 14. Boettcher, S.E. and Kalisz, P J . 1990. Single tree influence on soil properties in the 162  mountains of eastern Kentucky. Ecology 71: 1365-1372. 15. Bormann, B.T., Cromack Jr, K., and Russell III, W.O. 1994. Influences of red alder on soils and long-term productivity. In The biology and mangement of red alder. Edited by D.E. Hibbs, D.S. DeBell, and R.F. Tarrant. OSU press, Corvallis pp. 47-56. 16. Bormann, B.T. and DeBell, D.S. 1981. Nitrogen content and other soil properties related to age of red alder stands. Soil Sci. Soc. Am. J. 45: 428-432. 17. Bormann, B.T. and Sidle, R.C. 1990. Changes in productivity and distribution of nutrients in a chronosequence at Glacier Bay National Park, Alaska. J Ecol. 78: 561-578. 18. Bremner, J. M . 1965 Inorganic forms of nitrogen, in: Agronomy 9: Methods of soil analysis 1(2), Edited by Black, C. A. Madison, WI. pp. 1179-1237. 19. Brown, A.H.F. and Harrison, A.F. 1982. Effects of tree mixtures on earthworm populations and nitrogen and phosphorus status in Norway spruce {Picea abies) stands. In New trends in soil biology: proceedings of the VII international colloquium of soil zoology. Edited by P. Lebrun, H . M . Andre, A. De Medts, C. Gregoire-Wibo, and G. Wauthy. Dieu-Brichart, Belgium pp. 101-108. 20.  Brown, M J . and Parker, G.G. 1994. Canopy light transmittance in a chronosequence of mixed species deciduous forests. Can. J. For. Res. 24: 1694-1703.  21. Brozek, S. 1990. Effect of soil changes caused by red alder (Alnus rubra) on biomass and nutrient status of Douglas-fir (Pseudotsuga menziesii) seedlings. Can. J. For. Res. 20: 1320-1325. 22.  Brunner, A. and Nigh, G. 2000. Light absorption and bole volume growth of individual Douglas-fir trees. Tree Phys. 20: 323-332.  23.  Cade-Menun, B.J. and Lavkulich, L . M . 1997. A comparison of methods to determine total, organic and available phosphorus in forest soils. Commun. Soil Sci. Plant Anal. 28: 651-663.  24.  Callaway, R . M . and Walker, L.R. 1997. Competition and facilitation: a synthetic approach to interactions in plant communities. Ecology 78: 1958-1965.  25.  Canham, C D . 1988. A n index for understory light levels in and sround canopy gaps. Ecology 69: 1634-1638.  26.  Carcamo, H.A., Abe, T.A., Prescott, C.E., Holl, F.B., and Chanway, C P . 2000. Influence of millipedes on litter decomposition, N mineralization and microbial communities in a coastal forest in British Columbia, Canada. Can. J. For. Res. 30: 817-826.  27.  Carter, M.R. 1993. Soil sampling and methods of analysis. Lewis Publishers, Boca 163  Raton. 37 pp. 28.  Carter, R.E. and Klinka, K. 1992. Variation in shade tolerance of Douglas-fir, western hemlock, and western red cedar in coastal British Columbia. For. Ecol. Manage. 55: 105.  29.  Cole, D. W., Gessel, S. P., and Turner, J. 1978. Comparative mineral cycling in red alder and Douglas-fir. Pacific Northwest Forest and Range Experiment Station, For. Serv., U S D A No. U S D A For. Serv. Gen. Tech. Rep. PNW-70.Portland  30.  Cole, E.C. and Newton, M . 1986. Nutrient, moisture and light relations in 5 year old Douglas-fir plantations under variable conditions. Can. J. For. Res. 16: 727-732.  31.  Cole, E . C . and Newton, M . 1987. Fifth-year responses of Douglas-fir to crowding and nonconiferous competition. Can. J. For. Res. 17: 181-186.  32.  Comeau, P.G. 1996. Why mixedwoods? In Silviculture of temperate and boreal broadleaf-conifer mixtures. Edited by P.G. Comeau and K.D. Thomas. B C ministry of forests, Victoria. Land Management Handbook no. 36. pp. 1-7.  33.  Comeau, P . G . 1998. LITE: a model for estimating light under broadleaf and conifer tree canopies. B C Ministry of Forests, Victoria. Extension note no. 23.  34.  Comeau, P. G., Gendron, F., and Letchford, T. A comparison of several methods of estimating light under a paper birch mixedwood stand. 1998.  35. Comeau, P. G., Harper, G., Biring, B., Fielder, P., and Reid, W. 1997 Effects of red alder on stand dynamics and nitrogen availability F R B C project HQ96400-RE (MOF EP 1121.01). B C Ministry of Forests, Victoria. Extension note no. 9, 6p. 36.  Comeau, P. G. and Macdonald, R. 1998. SLIM user's guide, unpubl.  37.  Comeau, P. G., Macdonald, R., Bryce, R., and Groves, B. 1998. LITE: a model for estimating light interception and transmission through forest canopies. B C Ministry of Forests Research Branch. 93pp. (unpubl.)  38.  Comeau, P. G. and Sachs, D. 1992. Simulation of the consequences of red alder management on the growth of Douglas-fir using FORCYTE-11. Forestry Canada/BC Ministry of Forests F R D A Report No. 187.  39.  Compton, J. E. and Cole, D. W. 1989. Growth and nutrition of second rotation red alder. Agronomy abstracts.  40.  Compton, J.E. and Cole, D.W. 1998. Phosphorus cycling and soil P fractions in Douglasfir and red alder stands. For. Ecol. Manage. 110: 101-112.  41.  Compton, J.E., Cole, D.W., and Homann, P.S. 1997. Leaf element concentrations and soil  164  properties in first- and second-rotation stands of red alder. Can. J. For. Res. 27: 662-666. 42.  Constabel, A.J. and Lieffers, V.J. 1996. Seasonal patterns of light transmission through boreal mixedwood canopies. Can. J. For. Res. 26: 1008-1014.  43.  Crampton, C B . 1982. Pozolization of soils under individual tree canopies in southwestern British Columbia, Canada. Geoderma 28: 57-61.  44.  Crews, T.E. 1993. Phosphorus regulation of nitrogen fixation in a traditional Mexican agroecosystem. Biogeochem. 21: 141-166.  45. Crews, T.E., Kitayama, K., Fownes, J.H., Riley, R.H., Herbert, D.A., Mueller-Dombois, D., and Vitousek, P.M. 1995. Changes in soil phosphorus fractions and ecosystem dynamics across a long chronosequence in Hawaii. Ecology 76: 1407-1424. 46.  Curran, M.P. 1984. Soil testing for phosphorus availability to some conifers in British Columbia. Ph.D thesis, University of British Columbia.  47. Dawson, C. and Dunn, D. 1998. Laboratory Protocols. B C Ministry of Forests Analytical Laboratory, Victoria. 48. DeBell, D.S. and Radwan, M.A. 1984. Foliar chemical concentrations in red alder stands of various ages. Plant and Soil 77: 391-394. 49. Edmonds, R.L. 1980. Litter decomposition and nutrient release in Douglas-fir, red alder, western hemlock, and Pacific silver fir ecosystems in western Washington. Can. J. For. Res. 10: 327-337. 50. Ewers, B., Binkley, D., and Bashkin, M . A . 1996. Influence of adjacent stand on spatial patterns of soil carbon and nitrogen in Eucalyptus and Albizia plantations. Can. J. For. Res. 26: 1501-1503. 51. Ferrari, J.B. 1998. Fine scale patterns of leaf litterfall and nitrogen cycling in an old growth forest. Can. J. For. Res. 29: 291-302. 52. Ferrari, J.B. and Sugita, S. 1996. A spatially explicit model of leaf litter fall in hemlockhardwood forests. Can. J. For. Res. 26: 1905-1913. 53. Fielder, P. and Comeau, P. G. 1998. Design, construction and testing of an inexpensive sensor for measurement of photosynthetic photon flux density in the field. B C Ministry of Forests research branch. Working Paper. 54. Fisher, R.F. and Binkley, D. 1999. Ecology and management of forest soils. John Wiley & Sons, Toronto. 489pp. 55. Fyles, J.W. and Fyles, L H . 1992. Interaction of Douglas-fir with red alder and salal  165  foliage litter during decomposition. Can. J. For. Res. 23: 358-361. 56.  Garcia-Montiel, D.C. and Binkley, D. 1998. Effect of Eucalyptus saligna and Albizia falcataria on soil processes and nitrogen supply in Hawaii. Oecologia 113: 547-556.  57.  Gendron, F., Messier, C , and Comeau, P.G. 1998. Comparison of various methods for estimating the mean growing season percent photosynthetic photon flux density in forests. Agric. For. Meteorol. 92: 55-70.  58.  Gessel, S.P. and Turner, J. 1974. Litter production by red alder in western Washington. For. Sci. 20: 325-330.  59.  Giardina, C , Huffman, S., Binkley, D., and Caldwell, B.A. 1995. Alders increase soil phosphorus availability in a Douglas-fir plantation. Can. J. For. Res. 25: 1652-1657.  60.  Gower, S.T. and Son, Y . 1992. Differences in soil and leaf litterfall nitrogen dynamics for five forest plantations. Soil Sci. Soc. Am. J. 56: 1959-1966.  61.  Green, R.N. and Klinka, K. 1994. A field guide for site identification and interpretation for the Vancouver forest region. B C ministry of forests, Vancouver.  62.  Hart, S.C., Binkley, D., and Perry, D.A. 1997. Influence of red alder on soil nitrogen transformation in two conifer forests of contrasting productivity. Soil Biol. Biochem. 29: 1111-1123.  63. Heneghan, L. and Bolger, T. 1998. Soil microarthropod contribution to forest ecosystem processes: the importance of observational scale. Plant and Soil 205: 113-124. 64.  Hibbs, D.E. and DeBell, D.S. 1994. The management of young red alder. In The biology and management of red alder. Edited by D.E. Hibbs, D.S. DeBell, and R.F. Tarrant. O S U press, Corvallis pp. 202-215.  65. Holmgren, M . , Scheffer, M . , and Huston, M . A . 2000. The interplay of facilitation and competition in plant communities. Ecology 78: 1966-1975. 66. Homann, P.S., Van Miegroet, H., Cole, D.W., and Wolfe, G.V. 1992. Cation distribution, cycling, and removal from mineral soil in Douglas-fir and red alder forests. Biogeochem. 16: 121-150. 67. Hutchison, B.A. and Matt, D.R. 1977. The distribution of solar radiation within a deciduous forest. Ecological Monographs 47: 185-207. 68.  Ingestad, T. 1962. Macro element nutrition of pine, spruce, and birch seedlings in nutrient solutions. Akademisk Avhandling, Stockholm  69.  Jasienski, M . and Bazzaz, F.A. 1999. The fallacy of ratios and the testability of models in  166  biology. Oikos 8 4 : 321-326. 70. Joergensen, R.G. and Scheu, S. 1999. Response of soil microorganisms to the addition of carbon, nitrogen and phosphorus in a forest Rendzina. Soil Biol. Biochem. 31: 859-866. 71. John, M.K. 1970. Colorimetric determination of phosphorus in soil and plant materials with ascorbic acid. Soil Sci. 109: 214-220. 72. Kalra, Y . P. and Maynard, D. G. 1991. Methods manual for forest soil and plant analysis. No. Information Report NOR-X-319. 73. Leco Intruments, Corp. 1996. Instruction and Reference Documents for Model CHN-600 Analyzer. Leco Corporation, St. Louis, Missouri. 74.  Leco Intruments, Corp. 1996. Instruction and Reference Documents for Model SC-132 Analyzer. Leco Corporation, St. Louis, Missouri.  75. Lieffers, V.J., Messier, C , Stadt, K.J., Gendron, F., and Comeau, P.G. 1999. Predicting and managing light in the understory of boreal forests. Can. J. For. Res. 29: 796-811. 76.  Luken, J.O. and Fonda, R.W. 1983. Nirgoen accumulation in a chronosequence of red alder communities along the Hoh River, Olympic National Park, Washington. Can. J. For. Res. 13: 1228-1237.  77. Machado, J.-L. and Reich, P.B. 1999. Evaluation of several measures of canopy openness as predictors of photosynthetic photon flux density in deeply shaded conifer dominated forest understory. Can. J. For. Res. 29: 1438-1444. 78. Mailly, D. and Kimmins, J.P. 1997. Growth of Pseudotsuga menziesii and Tsuga heterphylla seedlings along a light gradient: resource allocation and morphological acclimation. Can. J. Bot. 75: 1424-1435. 79. Man, R. and Lieffers, V.J. 1999. Are mixtures of aspen and white spruce more productive than single species stands. For. Chron. 75: 505-513. 80. Marschner, H. 1995. Mineral nutrition of higher plants. Harcourt Brace & Company, Toronto 81. McClaugherty, C.A., Pastor, J., Aber, J.D., and Melillo, J.M. 1985. Forest litter decomposition in relation to soil nitrogen dynamics and litter quality. Ecology 66: 266275. 82. McLean, J. A . 10 year evaluation of Project G89-2 - M K R : test of the use of east west oriented red alder rows to reduce the infestation of sitka spruce. 2000. 83. Meidinger, D. and Pojar, J. 1991. Ecosystems of British Columbia. B C ministry of  167  forests, Victoria. Special Report Series no. 6. 84. Melillo, J.M., Aber, J.D., and Muratore, J.F. 1982. Nitrogen and Lignin control of hardwood leaf litter decomposition dynamics. Ecology 63: 621-626. 85. Messier, C. 1996. Managing light and understory vegetation in boreal and temperate broadleaf-conifer forests. In Silviculture of temperate and boreal broadleaf-conifer mixtures. Edited by P.G. Comeau and K.D. Thomas. B C ministry of forests, Victoria Land Management Handbook no. 36. pp. 59-81. 86. Messier, C. and Bellefleur, P. 1988. Light quantity and quality on the forest floor of pioneer and climax stages in a birch-beech-sugar maple stand. Can. J. For. Res. 18: 615622. 87. Messier, C. and Puttonen, P. 1995. Spatial and temporal variation in the light environment of developing Scots pine stands: the basis for a quick and efficient method of characterizing light. Can. J. For. Res. 25: 343-354. 88. Miller, E. K. and Murray, M . D. 1978. The effects of red alder on growth of Douglas-fir. Pacific Northwest Forest and Range Experiment Station, For. Serv., U S D A No. U S D A For. Serv. Gen. Tech. Rep. PNW-70.Portland 89. Miller, R.E. and Reukema, D.L. 1993. Size of Douglas-fir trees in relation to distance from a mixed red alder - Douglas-fir stand. Can. J. For. Res. 23: 2413-2418. 90.  Ogden, A . E . and Schmidt, M . G . 1997. Litterfall and soil characteristics in canopy gaps occupied by vine maple in a coastal western hemlock forest. Can. J. Soil. Sci. 77: 703711.  91. Parent, S. and Messier, C. 1996. A simple and efficient method to estimate microsite availabilty under a forest canopy. Can. J. For. Res. 26: 151-154. 92.  Peterson, E. B., Ahrens, G. R., and Peterson, N. M . 1996. Red alder manager's handbook for British Columbia. B C ministry of forests F R D A Report No. 240. Victoria  93. Prescott, C.E., Zabek, L . M . , Staley, C.L., and Kabzems, R. 2000. Decomposition of broadleaf and needle litter in forests of British Columbia: influences of litter type, forest type and litter mixtures (In press). Can. J. For. Res. 94.  Pritchett, W.L. and Fisher, R.F. 1987. Properties and mangement of forest soils. John Wiley & Sons, New York. 494 pp.  95. Pulleman, M . and Tietema, A. 1999. Microbial C and N transformations during drying and rewetting of coniferous forest floor material. Soil Biol. Biochem. 31: 275-285. 96. Radwan, M . A . and DeBell, D.S. 1994. Fertilization and nutrition of red alder. In The biology and management of red alder. Edited by D.E. Hibbs, D.S. DeBell, and R.F. 168  Tarrant. OSU press, Corvallis pp. 216-228. 97. Radwan, M.A., Harrington, C.A., and Kraft, J.M. 1984. Litterfall and nutrient returns in red alder stands in western Washington. Plant and Soil 79: 343-351. 98. Rhoades, C C . and Binkley, D. 1992. Spatial extent of impact of red alder on soil chemistry of adjacent conifer stands. Can. J. For. Res. 22: 1434-1437. 99. Ross, D J . , Kelliher, F.M., and Tate, K.R. 2000. Microbial processes in relation to carbon, nitrogen and temperature regimes in litter and a sandy mineral soil from a central Siberian Pinus sylvestris L . forest. Soil Biol. Biochem. 31: 757-767. 100.  Rygiewicz, P.T., Bledsoe, C.S., and Zasoski, R J . 1984. Effect of ectomycorrhizae and solution pH on [ N]ammonium uptake by coniferous seedlings. Can. J. For. Res. 14: 885-892. !5  101.  Satchell, J. E . and Lowe, D. G. 1967. Selection of leaf litter by Lumbricus terrestris.. Progress in soil biology: Proceedings of the colloquium on dynamics of soil communities, pp 102-119.  102.  Shainsky, L J . and Radosovich, S.R. 1992. Mechanisms of competition between Douglasfir and red alder seedlings. Ecology 733: 30-45.  103.  Shainsky, L.J., Yoder, B J . , Harrington, T.B., and Chan, S.S.N. 1994. Physiological characteristics of red alder: water relations and photosynthesis. In The biology and management of red alder. Edited by D.E. Hibbs, D.S. DeBell, and R.F. Tarrant. O S U press, Corvallis pp. 73-91.  104.  Spies, T.A., Franklin, J.F., and Klopsch, M . 1990. Canopy gaps in Douglas-fir forests of the Cascade Mountains. Can. J. For. Res. 20: 649-658.  105.  Sprent, J.I. and de Faria, S.M. 1988. Mechanisms of infection of plants by nitrigen fixing organisms. Plant and Soil 110: 157-165.  106.  Stark, J.M. and Hart, S.C. 1997. High rates of nitrification and nitrate turnover in undisturbed coniferous forests. Nature 385: 61-64.  107.  Swanston, C W . and Myrold, D.D. 1997. Incorporation of nitrogen from decomposing red alder leaves into plants and soil of a recent clearcut in oregon. Can. J. For. Res. 27: 1496-1502.  108.  Thermo Instruments, Ltd. Instruction and Reference Documents for NA-1500 Analyzer. 1996. Thermo Instruments Ltd, Boston, Massachusetts.  109.  Thomas, K. D. and Comeau, P. G. 1998. Effects of Bigleaf maple on growth of understory conifers and the effects of coppice spacing on the growth of maple. B C 169  Ministry of Forests, Victoria. Extension Note no.24. 4p. 110.  Timmer, V.R. 1991. Interpretation of seedling analysis and visual symptoms. In Mineral nutrition of conifer seedlings. Edited by R. van den Driessche. C R C press, Boca Raton pp.113-134.  111.  Turner, J., Cole, D.W., and Gessel, S.P. 1976. Mineral nutrient accumulation and cycling in a stand of red alder (Alnus rubra Bong.). J. Ecol. 64: 965-974.  112.  Ugolini, F.C. 1968. Soil development and alder invasion in a recently deglaciated area of Glacier Bay, Alaska. In Utilization and management of alder. Edited by J.M. Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen. U S D A For.Serv., PNW For. Range Exp. Sta., Portland pp. 115-140.  113.  Valentine, D.W. and Binkley, D. 1992. Topography and soil acidity in an arctic landscape. Soil Sci. Soc. Am. J. 56: 1553-1559.  114.  Van Cleve, K. and Viereck, L.A. 1972. Distribution of selected chemical elements in even-aged alder (Alnus) ecosystems near Fairbanks, Alaska. Arctic and Alpine Res. 4: 239-255.  115.  Van Cleve, K., Viereck, L . A . , and Schlentner, R.L. 1971. Acccumulations in alder (Alnus) ecosystems near Fairbanks, Alaska. Arctic and Alpine Res. 3: 101-114.  116.  Van Miegroet, H., Cole, D.W., Binkley, D., and Sollins, P. 1989. The effect of nitrogen accumulation and nitrification on soil chemical properties in alder forests. In Effects of air pollution on western forests. Edited by R. Olson and A. Lefohn. Air and waste management association, Pittsburgh pp. 515-528.  117.  Wright, E.F., Coates, K.D., Canham, C D . , and Bartemucci, P. 1998. Species variability in growth response to light across climatic regions in northwestern British Columbia. Can. J. For. Res. 28: 871-886.  118.  Zavitkovski, J. and Newton, M . 1971. Litterfall and litter accumulation in red alder stands in western Oregon. Plant and Soil 35: 257-268.  119.  Zinke, P.J. 1962. The pattern of influence of individual forest trees on soil properties. Ecology 43: 130-133.  120.  Zou, X . 1993. Species effects on earthworm density in tropical tree plantations in Hawaii. Biol. Fertil. Soils 15: 35-38.  121.  Zou, X., Binkley, D., and Caldwell, B.A. 1995. Effects of dinitrogen fixing trees on phosphorus biogeochemical cycling in contrasting forests. Soil Sci. Soc. Am. J. 59: 14521458.  170  Appendix I Program used for Campbell CR-10 datalogger recording light at M K and N M sites  Program file for CR10 and explanation of programming Mode 01 scan rate: 10.000  system scans every ten seconds.  01:P10  Batterry Voltage reading 01:0012  Differential Voltage Reading (Li-Cor Quantum Sensor)  02:P02 01:01  one repetition  02:23  range 23 (60hz rejection +/- 25 mV)  03:01  channel 1  04:0001  location 1  05:+200.00 (279.05)  multiplier (calibration for individual licor)  06:0000  offset (not in use) Single Ended Voltage inputs (photodiodes)  03:P01 01:10  10 repetitions (one each for ten sensors)  02:23  range 23 (see above)  03:03  beginning channel 3 (diff voltage takes up 1 and 2)  04:0002  location beginning at 2  05:1.000  multiplier of 1 (will convert in excel later)  06:0000  offset (not in use) IF Time program control instruction  04:P92 01:0000  time into interval is 0 minutes (immediate)  02:0060  time interval of sixty minutes  03:10  set output flag to 0 real time function  05:P77 01:0220  day-hour/minute beginning at midnight  Average function  06:P71 01:11  11 repetitions  02:0001  beginning at location 1 IF Time program control instruction  07:P92  08:P73  Output to location 12  01:0000  time interval at 0 minutes (immediate)  02:1440  every 24 hours  03:10  set output flag to 0 Maximum function  171  01:11  11 repetitions (1 24h max for each sensor)  02:00  no time stamp  03:0001  output to 1  01:12  minimum fiinction  02:00  12 repetitions (will include min. voltage)  03:0001  output to area 1  09:P74  172  Appendix II Examples of hemispherical photographs taken in winter and summer for various sites  3. Transect N M 2 point 3 (10 m) winter  Appendix III Sample screen dumps from SLIM and LITE  File  Plot Atmosphere  D|G?|B|  View Window Help  tMMBl  • Plot Properties ^ World Location | Azimuth | § Horizon j ^ Sky Conditions 1938 S 1999  6  gp  i HI^Stakes1A1 -{x:10.0.y: 45.0. elevation: 7.2} S  B •  ft ft ft ft ft  1A10 • { x: 10.0. JI: 10.0. elevation: 1.8) 1A2-(x:10.0.y:38.0.elevation:5.1} 1A3-{x:10.0.y:35 0. elevation:4.7} 1A4 • {x:10.0, jr32.0. elevation: 4.3} 1 A 5 { x 1 0 0 . y 2 9 0 . elevation 4.0)  ft ft ft ft  • ' Photo 1A5: Angle ol View: 0.0. Width ol View 3S0.C Photo: 1A5; Angle of View 0 0 . Width ot View 360C 1 A 6 { x 1 0 0.jr 26 0. elevation 3 5} 1A7 - {x:10.0. jr23.0. elevation:3.3} 1A8-{x:10.0.jr20.0. elevation:2.8} 1A9.{x:10.0.jr15.0.elevation:2.1(  ft Cornerl (x:0.0,7.55.0. elevation: 8.0} ft corner2 (x:20.0.y:55.0. elevation: 6.5} ft corner3 {x:0.0. y:0.0. elevation:2.0} ft corner4 (x:20.0. y:0.0. elevation: 0.0} * Species £i AC: Cottonwood K AT: Trembling Aspen  For H elp. press F1  Times  Location: {0.00.0.00.0.00}..  June 1.1998 01:00:C  Recalculate Interval ) 30 minutes H Aug 31.1998 23:00:00  Diffuse Above Canopy Below Canopy Percentage  Direct  1 496813029 1 334 296 811 0 o: 0.0% " " 0 . 0 %  Total 2 831 109 840 1 0 ] 0.0%  ;  | Crown is Valid  1. Typical screen in LITE showing Plot Window, sample Below Canopy Reading window (BCR), and Longeterm PPFD calculation Window  175  LITE File  la.lit  Rot Atmosphere  For Help, press F l  2.  Trees Shrubs View  "  [  C  Window  r  o  Help  w  n  is Valid  j Display LA: Vertical. Up-Down @ 10.6 metres  Mouse Mode: I  Lite Window with Horizontal and Vertical crown view windows, and value window for a selected point on the vertical view window.  176  SLIM - Transect 1A.gfr Photo Attributes  Window Help lA-aTr rrtw««h... H I "  j&) Plot Properties ^ ) World Location • Azimuth 41 Horizon Sky Conditions + 1998 1999 Stakes fit 1A1 -{x:10.0,y: 10.0. elevation: 7.2} Photo: 1A1;Angleo( View: 0.0.Widtt Photo: 1A1; Angle of View: 0.0, Widtt fit 1A10 - { K 10.0, y:45.0, elevation: 1.8) ft 1A2 - { x:10.0.y:17.0. elevation:5.1}  rap™ m GFR: Transect lA.gfr  [Photo Wizaid|:2  Click "Change Color Threshold...' to select the color that separates sky from foiage in your photograph. When you are finished, click IMext >' to create a threshold view.  «j  Current Photo DAMy Change Circle Selection Type..  Change Color Jhreshold.  Next>  For Help, press F1  3. Typical window in SLIM, showing the photo-conversion wizard, and related output windows.  177  Appendix IV Raw data for litterfall mass inputs and nutrient concentrations.  Litter Type Alder Om 7m 10 m 13 m  Transect (all weights in kg/ha/yr) CHI  NM2  M K 2 NM1  MK1  MK3 MK4  CH2  SC2  SCI  SC3  SC4 | SC5 SC6  4931 3899 2815 2121 3860 5927 4734 4352 1285 4400 3111 2546 2282 6386 5561 4704 4922 1214 3318 1938 1739 1600 6551 5443 5940 4349 1014  467  142  688  776  654  351  149  443  751  223  330  52  229  230  318  1230 1999 5485 2861 3736 714 3997 2088 606 518  0  42  67  62  36  72  0  0  0  0  6  32  828  0  0  0  0  0  66  2269  702  1688  16m  1325  193  445  19 m  434  19  598  22 m  305  0  818  34  233  48  0  22  0  0  0  22  25 m  324  3 1391  147  168  34  195  51  0  119  0  0  0  0  18  0 1893  92  1503  191  237  44  0  82  0  0  0  0  138 2415 3957  156  0  0  0  0  0  0  0  30 m 35 m Conifer  0  0 1976 M K 2 NM1  MK1  374  542  98 3101  NM2  CHI  119 2379  CH2  Om  0  0  0  0  0  0  7m  0  0  33  14  32  32  285  356  359  332  280  472  919 893 1369 920 1177 2219 2032 2254  1282  1627  10 m 13 m 16m 19 m 22 m  1983 2050  1523 2785 2397 2341 2445 2312 2984 2940 2382 2308 2455 2304  30 m  3262 3299 2354 2312 2234 2272 3217 3278 2441 2465 2383 2140  35 m  3185 3263 2468 2431 2502 1631  25 m  ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^  1. Litterfall mass input totals for the year's sampling, 1998-1999.  178  Transect  MK2  MKl Dr  Nutrient  Con  Dr  Con  NM2  NMl Con  Dr  Dr  CH2  CHI  Con  Dr  Con  Dr  Con  N(%)  2.20  1.57  2.20  1.56  2.20  1.32  2.20  1.32  2.60  1.48  2.40  1.48  P(%)  0.10  0.06  0.10  0.06  0.10  0.08  0.10  0.08  0.10  0.07  0.10  0.07  K(%)  0.30  0.06  0.20  0.09  0.20  0.13  0.20  0.13  0.20  0.09  0.30  0.09  0.20  0.08 0.12  0.10  0.09  0.10  0.06  0.20  0.09  0.20  0.09  0.10  0.08  S(%)  0.10  0.12  0.10  0.12  0.10  0.11  0.10  0.11  0.10  0.12  0.10  Ca(%)  0.90  0.67  1.10  0.74  1.50  1.15  1.60  1.15  1.30  0.85  1.30  0.85  63.50  8.90  18.10  8.70  18.10  9.40  30.03  9.00  30.03  Mg(%)  9.70  Cu(PPM)  9.10  8.50  Fe(PPM) 252.90 3427.00 414.50 921.10 M n ( P P M ) 409.50 B(PPM)  338.10 286.10 252.30 6.70  12.70  Zn(PPM)  37.70  49.80  8.80  14.40 38.70  152.90 1127.80 134.30 1127.80 566.10 1825.43 987.60 1825.43 200.40 313.70 183.30 313.70 249.20 301.37 270.90 301.37  49.30  25.90 30.20  14.20 26.60  29.40 32.40  14.20 26.60  9.40  9.90  11.10  9.90  32.60  41.90  31.70  41.90  2. Nutrient concentrations in alder and conifer litterfall for composite samples measured.  Litter regression F-statistic and p-values Information  Transect (includes 3 replicate subsets a,b &c) MKl  Transect  Points used in 2-6 regression -331 Slope  MK2  NMl  NM2  CHI  CH2  MK3  MK4  SCI  SC2  SC3  SC4  SC5  2-6  2-6  2-6  2-6  2-6  2-7  2-7  2-4  2-4  2-4  2-4  2-4  -264  -172  -160  -590  -520  -303  -348  -202  -51  -14  -80  -40  0.81  0.78  0.85  0.73  0.97  0.87  0.80  0.82  0.83  0.87  0.89  0.83  0.76  F-statistic  99.88  62.58  42.13  53.94  44.79  72.64  10.94  148.2  6.68  4.03  8.82  9.58  20.85  p-value  <.001  <.001  <.001  <.001  <.001  <.001  0.03  <.001  0.24  0.29  0.09  0.09  0.02  Multiple R  z  3. Regression statistics for alder litterfall mass input at each site.  179  Insect Groups Defined* (see full names below) Proportion per trap Point Arac  Mill Cent B-Sl Asc Moll Olig Iso  Stap Carab Circu Silph Dipt  Hym  Orth Other Total  1A1  0.00 0.22 0.05 0.15 0.02 0.06 0.06 0.01 0.00  0.09  0.00  0.11 0.18  0.03  0.06  0.03  1.00  1A3  0.04 0.10 0.01 0.10 0.00 0.16 0.08 0.00 0.01  0.03  0.01  0.02 0.38  0.00  0.04  0.08  1.00  1A5  0.09 0.05 0.00 0.01 0.00 0.03 0.01 0.01 0.12  0.05  0.01  0.03 0.31  0.16  0.09  0.06  1.00  1A7  0.06 0.09 0.00 0.02 0.00 0.04 0.04 0.00 0.00  0.04  0.00  0.06 0.43  0.00  0.09  0.13  1.00  0.03  0.09  0.03  1.00  1A9  0.07 0.05 0.01 0.00 0.00 0.01 0.00 0.00 0.17  0.07  0.00  0.01 0.45  1B1  0.01 0.17 0.01 0.04 0.04 0.06 0.07 0.06 0.07  0.06  0.00  0.13 0.21  0.04  0.03  0.04  1.00  1B3  0.03 0.21 0.03 0.05 0.06 0.11 0.08 0.08 0.05  0.10  0.01  0.11 0.05  0.00  0.01  0.03  1.00  1B5  0.02 0.08 0.01 0.00 0.00 0.09 0.06 0.05 0.05  0.06  0.02  0.04 0.41  0.04  0.03  0.04  1.00  0.04  0.02  1.00  1B7  0.02 0.10 0.01 0.04 0.00 0.12 0.09 0.01 0.15  0.06  0.00  0.08 0.22  0.04  1B9  0.03 0.04 0.01 0.01 0.00 0.12 0.00 0.00 0.10  0.03  0.00  0.12 0.31  0.07  0.01  0.13  1.00  2C1  0.05 0.04 0.02 0.03 0.01 0.09 0.07 0.02 0.04  0.17  0.03  0.05 0.24  0.09  0.00  0.04  1.00  0.01  0.07  1.00  2C3  0.04 0.05 0.00 0.02 0.01 0.05 0.01 0.01 0.04  0.17  0.00  0.09 0.39  0.03  2C5  0.18 0.09 0.02 0.01 0.00 0.02 0.00 0.03 0.03  0.26  0.01  0.07 0.14  0.05  0.05  0.04  1.00  2C7  0.07 0.07 0.01 0.04 0.01 0.04 0.01 0.01 0.11  0.13  0.00  0.11 0.32  0.01  0.01  0.05  1.00  2C9  0.04 0.09 0.00 0.01 0.01 0.03 0.00 0.01 0.11  0.21  0.01  0.11 0.27  0.02  0.02  0.06  1.00  0.00  0.05  1.00  2D1  0.08 0.09 0.02 0.00 0.01 0.07 0.02 0.00 0.05  0.20  0.01  0.06 0.29  0.06  2D3  0.06 0.04 0.01 0.01 0.01 0.02 0.03 0.00 0.23  0.12  0.00  0.08 0.21  0.04  0.06  0.08  1.00  2D5  0.08 0.10 0.02 0.01 0.00 0.05 0.01 0.00 0.22  0.12  0.01  0.06 0.26  0.01  0.02  0.03  1.00  2D7  0.04 0.02 0.01 0.00 0.00 0.01 0.01 0.00 0.29  0.04  0.00  0.17 0.28  0.02  0.01  0.10  1.00  0.08  0.10  1.00  2D9  0.13 0.09 0.01 0.00 0.00 0.02 0.00 0.00 0.11  0.14  0.02  0.06 0.23  0.01  3E1  0.03 0.12 0.00 0.20 0.00 0.02 0.03 0.06 0.04  0.25  0.01  0.15 0.08  0.02  0.01  0.00  1.00  3E3  0.04 0.26 0.02 0.16 0.01 0.05 0.03 0.06 0.05  0.09  0.01  0.10 0.12  0.02  0.01  0.01  1.00  3E5  0.14  0.01  0.16 0.14  0.03  0.00  0.00  1.00  3E7  0.04 0.19 0.03 0.07 0.00 0.04 0.00 0.06 0.08 0.01 0.31 0.02 0.07 0.00 0.03 0.02 0.00 0.07  0.13  0.02  0.01 0.22  0.03  0.02  0.02  1.00  3E9  0.02 0.22 0.00 0.10 0.00 0.04 0.02 0.02 0.07  0.18  0.00  0.12 0.14  0.03  0.01  0.05  1.00  3F1  0.06 0.26 0.00 0.16 0.01 0.06 0.03 0.00 0.04  0.16  0.00  0.05 0.13  0.03  0.02  0.03  1.00  0.04  1.00  3F3  0.03 0.36 0.01 0.07 0.00 0.07 0.05 0.00 0.04  0.14  0.00  0.07 0.16  0.02  0.01  3F5  0.03 0.17 0.01 0.01 0.00 0.02 0.01 0.00 0.03  0.21  0.00  0.08 0.25  0.01  0.02  0.15  1.00  3F7  0.02 0.30 0.01 0.04 0.00 0.01 0.02 0.00 0.03  0.12  0.01  0.04 0.36  0.01  0.01  0.01  1.00  3F9  0.06 0.18 0.01 0.02 0.00 0.03 0.02 0.00 0.05  0.14  0.00  0.03 0.16  0.24  0.02  0.04  1.00  Total  0.05 0.13 0.01 0.04 0.01 0.05 0.02 0.01 0.09  0.14  0.01  0.08 0.25  0.04  0.02  0.05|  1.00  4. Pitfall trap count totals for each morphotype classified. Insect groups defined are: Arac = Arachnida, Mill = Millipedes, Cen = centipedes, B-Sl = Banana Slugs, Asc =  Aschelminthes, Moll = Mollusca, Olig = Oligochaeta, Iso = Isopoda, Staph = Staphylinidae, Cara = Carabidae, Cir = Circulionidae, Silp = Silphidae, Dipt = Diptera, Hym = Hymenoptera, Orth = Orthoptera, Other= all groups with very low counts.  180  Appendix V Physical attributes of pot trial seedlings.  Physical Attributes Measured Height Growth  Root Collar Diameter  Root Weight  mean  std dev  mean  std dev  mean  1A1  9.58  3.38  1.71  0.39  6.41  1A2  9.00  1.67  1.39  0.75  7.58  9.33  2.14  1.86  0.66  7.20  Sample Points  1A3  Shoot Weight  R:S Ratio  mean  std dev  1.76  9.61  3.01  0.67  2.17  10.83  2.00  0.70  2.37  10.02  3.10  0.72  std dev  1A4  8.50  1.70  1.82  0.81  6.46  1.51  10.43  1.49  0.62  IAS  9.50  2.24  2.21  0.65  6.78  2.70  10.08  2.91  0.67  1A6  8.67  1.08  1.89  1.58  6.54  1.76  8.76  2.49  0.75  7.67  1.74  0.85  1A7  9.08  1.53  2.31  0.75  6.51  1.66  1A8  9.75  3.11  1.17  0.38  9.23  2.71  10.59  2.10  0.87  1A9  8.17  1.03  1.72  0.34  8.39  3.48  11.38  3.20  0.74  1A10  11.50 8.67  3.58 3.11  1.65  0.72  8.38  1.46  10.30  0.90  0.81  2.07  0.78  7.49  1.31  9.61  2.61  0.78  1B-1 1B-2  9.50  2.61  2.03  0.33  6.50  1.25  9.25  3.19  0.70  1B-3  10.25  1.81  2.14  0.86  7.40  1.84  9.87  1.29  0.75  1B-4  11.67  1.66  0.89  1.78  0.71  8.29  1.76  9.32  1.74  1B-5  11.08  2.48  1.85  0.68  7.03  1.23  10.32  2.16  0.68  1B-6  8.75  3.75  2.07  0.71  5.96  1.41  8.93  3.41  0.67  1B-7  11.33  5.06  1.88  0.76  6.68  1.54  8.06  1.93  0.83  1B-8  6.67  2.82  1.67  0.61  6.80  2.17  9.48  1.78  0.72  1B-9  9.33  2.98  1.86  0.85  6.83  1.28  11.05  2.38  0.62  1B-10  9.33  1.97  1.71  0.96  8.42  1.71  10.23  2.22  0.82  2C-1  10.25  1.44  2.03  0.82  7.83  1.04  9.64  1.10  0.81  2C-2  9.08  2.75  1.64  0.49  6.15  1.83  8.79  2.25  0.70  2C-3  8.75  3.09  2.35  0.80  8.70  3.27  9.70  1.71  0.90  2C-4  8.75  1.04  1.59  1.22  7.01  3.15  7.80  2.13  0.90  2C-5  11.33  2.04  1.42  0.70  9.13  2.35  11.54  3.56  0.79  2C-6  8.75  1.97  1.88  0.65  5.76  1.57  9.48  2.82  0.61  2C-7  8.83  1.21  1.46  1.15  8.78  0.94  10.06  2.44  0.87  2C-8  12.50  2.12  1.41  0.64  10.76  0.85  10.77  1.30  1.00  2C-9  10.08  1.07  2.43  1.13  8.99  1.96  11.47  2.04  0.78  2C-10  10.50  2.26  1.69  0.60  7.12  2.73  9.38  3.28  0.76  2D-1  10.25  1.44  2.03  0.82  7.83 •  1.04  9.64  1.10  0.81  2D-2  7.92  1.59  1.84  0.75  6.24  1.55  9.92  1.45  0.63  2D-3  9.83  3.64  2.13  0.50  8.65  1.40  9.89  2.89  0.87  2D-4  8.17  2.32  1.93  0.62  8.39  1.89  10.64  1.86  0.79  2D-5  10.08  2.62  2.03  0.77  7.31 '  2.07  9.70  1.83  0.75  2D-6  9.33  2.71  1.12  0.66  6.61  2.36  8.77  2.48  0.75  2D-7  12.25  3.42  1.80  0.71  9.36  2.67  12.17  2.04  0.77  2D-8  9.67  2.07  1.61  0.42  6.05  0.94  9.57  1.97  0.63  2D-9  8.67  1.50  1.06  0.38  7.75  2.18  10.06  1.83  0.77  181  2D-10  9.92  3.53  1.36  0.55  6.88  . 1.44  11.27  3.77  0.61  3E-1 3E-2  9.67  1.89  2.09  0.64  7.83  3.15  9.80  2.42  0.80  9.75  3.34  1.57  0.92  7.68  1.78  9.55  2.01  0.80  1.52  1.07  7.24  1.41  9.16  1.25  0.79  3E-3  9.08  2.33  3E-4  9.75  1.75  1.90  0.53  7.44  1.49  10.26  2.12  0.73  3E-5  8.42  1.69  2.16  0.69  8.20  1.94  9.91  2.33  0.83  9.53  2.43  0.90  3E-6  10.58  2.40  1.81  0.60  8.56  2.60  3E-7  6.83  1.44  1.62  0.48  5.90  0.86  8.43  1.85  0.70  3E-8  10.67  2.11  2.09  0.54  7.33  1.92  9.90  1.68  0.74  7.79  2.67  7.90  1.93  0.99  9.46  1.13  0.79  3E-9 3E-10  6.25  2.86  1.08  0.45  11.08  2.60  2.49  0.74  7.51  0.89  3F-1  8.17  2.29  2.05  0.79  5.14  2.10  7.52  2.05  0.68  3F-2  9.50  2.77  1.73  0.38  7.71  2.98  11.20  1.98  0.69  1.67  10.33  1.41  0.93  3F-3  11.42  3.07  2.37  0.83  9.62  3F-4  7.75  3.13  1.47  0.71  8.10  1.38  10.14  1.28  0.80  3F-S  8.83  3.09  1.65  0.21  5.75  0.35  9.22  1.78  0.62  10.20  3.87  0.84  3F-6  11.00  2.92  1.51  0.95  8.60  2.72  3F-7  8.58  3.85  1.38  0.86  7.71  4.02  10.83  3.54  0.71  3F-8  9.00  2.98  1.77  0.52  7.27  1.48  10.85  1.17  0.67  3.13  1.82  0.54  7.46  0.64  9.76  1.63  0.76  2.16  10.40  2.48  0.94  3F-9 3F-10  8.67 9.67  2.86  1.22  0.74  182  9.74  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099531/manifest

Comment

Related Items