Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An ecological study of the Sitka spruce forest on the west coast of Vancouver Island Cordes, Lawrence David

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1973_A1 C67.pdf [ 38.19MB ]
JSON: 831-1.0100979.json
JSON-LD: 831-1.0100979-ld.json
RDF/XML (Pretty): 831-1.0100979-rdf.xml
RDF/JSON: 831-1.0100979-rdf.json
Turtle: 831-1.0100979-turtle.txt
N-Triples: 831-1.0100979-rdf-ntriples.txt
Original Record: 831-1.0100979-source.json
Full Text

Full Text

AN ECOLOGICAL STUDY OF THE SITKA SPRUCE FOREST ON THE WEST COAST OF VANCOUVER ISLAND by LAWRENCE D. CORDES B . S c , University of Minnesota M . S c , University of North Dakota A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of Botany We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December, 1972 O In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the requi rements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r re ference and s tudy . I f u r t h e r agree t ha t permiss ion fo r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department o r by h i s r e p r e s e n t a t i v e s . I t i s understood that copy ing o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l ga in s h a l l not be a l lowed wi thout my w r i t t e n p e r m i s s i o n . Lawrence D. Cordes Department o f Botany  The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada D ate December 2, 1972 ABSTRACT A forest consisting of almost pure stands of Sitka spruce (Picea sitchensis) occurs in a narrow band varying in width from several feet to approximately 600 feet along the west coast of Vancouver Island. In this area spruce replaces western hemlock (Tsuga heterophylla) and western redcedar (Thuja plicata), the dominant species of the region. A comparison of the physical environment of the coastal Sitka spruce forest with that of the hemlock - redcedar forest revealed that ocean spray intensity was the major difference between, the two. Monitoring of ocean spray over a 17 month period indicated that incoming spray is highest during^ the f a l l and winter when major storms are frequent in the northern Pacific Ocean. However, the effects of ocean spray on vegetation were at a maximum during late spring and early summer when infrequent precipitation allowed salts to accumulate on the foliage. Young needles of coniferous species were killed by spray during this period. Sitka spruce was found to be much more tolerant than either western hemlock or western redcedar. During the spring and summer the amount of incoming ocean spray varied considerably from site to site depending on the configuration of the coastline, the type of shoreline and the degree of protection provided by vegetation in a seaward direction. The width of the Sitka spruce forest as well as canopy form were closely related to the amounts of incoming ocean spray. Measurements of salt concentrations in throughfall precipitation indicated that significant amounts of sodium, magnesium, calcium and i i potassium are added to the soil beneath the Sitka spruce forest. These additions improve the nutrient status of the soil but do not accumulate to high levels due to the high precipitation of the region. Retention of these cations in the soil varies considerably with calcium being the highest and sodium the lowest. Studies of the Sitka spruce forest have resulted in the delineation and characterization of seven ecosystematic units on the basis of veg-etation, topography, parent material and historical factors. Comparisons of the environmental components of these units strongly suggest ocean spray intensity and water relations to be the most influential factors. An investigation of successional dynamics revealed that Sitka spruce was the dominant climax tree species in most of the forest type, although western hemlock shares this role with spruce in one type. Successional rates and stability of the different forest types are dependent on the geomorphic processes responsible for beach plain expansion and the degree of protection from incoming ocean spray provided by vegetation on the seaward side. Combining the forest types of the alluvial Sitka spruce forest, which were also studied, with those of the coastal Sitka spruce forest, a hierarchic system of classification consisting of plant associations, alliances and orders is proposed for the Sitka spruce forest of the Coastal Western Hemlock Zone on Vancouver Island. i i i TABLE OF CONTENTS Page I INTRODUCTION 2 II REGIONAL DESCRIPTION 7 TOPOGRAPHY 7 PHYSIOGRAPHY 7 THE COASTAL WESTERN HEMLOCK ZONE 8 Major Tree Species 10 The Sitka Spruce Forest 10 III METHODS 12 ANALYSIS 12 Vegetation 14 Soil 16 Ocean Spray 18 SYNTHESIS OF BIOGEOCOENOTIC UNITS 20 IV OCEAN SPRAY: A LITERATURE REVIEW 24 GENERAL CHARACTERISTICS OF OCEAN SPRAY 24 Formation of Ocean Spray 27 Dispersal of Ocean Spray 30 Deposition of Ocean Spray 34 Effects of Ocean Spray on Plants 39 Effects of Ocean Spray on Soils 50 Ocean Spray and the distribution of vegetation 53 V THE COASTAL SITKA SPRUCE FOREST ° 58 PHYSIOGRAPHY AND PLEISTOCENE EVENTS 60 iv TABLE OF CONTENTS (Continued) VEGETATION 70 SOILS 72 CLIMATE 74 SOME ASPECTS OF OCEAN SPRAY 82 Seasonal Pattern of Ocean Spray and its Relationship to Wind and Surf 83 Foliar Damage by Ocean Spray 96 Distribution of Ocean Spray During the Growing Season 101 Ocean Spray Dynamics and the Coastal Sitka Spruce Forest 106 FOREST TYPES OF THE COASTAL SITKA SPRUCE FOREST 123 The Rocky Coast Land Type 123 1. Picea sitchensis - Maianthemum dilatatum forest type 125 2. Picea sitchensis - Gaultheria shalion forest type 135 The Beach Plain and Marine Terrace Land Type 143 3. Picea sitchensis - Eurhynchium oreganum forest type 144 4. Picea sitchensis - Rubus spectabilis forest type 155 5. Tsuga heterophylla - Picea sitchensis forest type 165 6. Picea sitchensis - Carex obnupta forest type 176 7. Picea sitchensis - Polystichum muniturn forest type 187 Vegetation Summary 200 TABLE OF CONTENTS (Continued) FOREST DYNAMICS 200 ENVIRONMENTAL RELATIONSHIPS 214 Ocean Spray 215 Soil Moisture 220 Soil Chemistry 224 Throughfall Cations 227 Discussion 233 VI HIERARCHIC SYNOPSIS OF SITKA SPRUCE FOREST BI0GE0C0EN0SES 238 VII BIBLIOGRAPHY 242 VIII APPENDICES 269 vi LIST OF APPENDICES Appendix Page I CHECK LIST OF SPECIES 270 II EXPLANATORY NOTES FOR SYNTHESIS TABLES 280 III A VEGETATION CLASSIFICATION FOR PHASE I OF PACIFIC RIM NATIONAL PARK 283 IV SOIL PROFILE DESCRIPTIONS AND CHEMICAL ANALYSES 309 V PHYSICAL CHARACTERISTICS OF SOILS 354 VI OCEAN SPRAY AND THROUGHFALL DATA 367 VII THE ALLUVIAL SITKA SPRUCE FOREST 381 v i i LIST OF TABLES Table 1. Long-Term Climatic Records for Two Stations on the West Coast of Vancouver Island Seasonal Amounts of Ocean Spray Collected at Six Sites for a One-Year Period Seasonal Wind Frequency, Velocity, and Number of Hourly Recordings of 25 mph or Greater Mean Weekly Amounts of Sodium Collected at Ocean Spray Measuring Stations During the 1967 Growing Season Cox Bay Transect Showing Relationships of Canopy Height and Type of Forest to Ocean Spray and the Sodium Content of Throughfall Page 76 86 92 104 116 6. Forest Types of the Coastal Sitka Spuce Forest 124 7. Picea sitchensis - Maiantherrrum dilatatum plot data 126 8. Picea sitchensis - Maianthemum dilatatum soil data 127 9. Picea sitchensis - Maianthemum dilatatum vegetation data 128 10. Picea sitchensis - Gaultheria shallon plot data 136 1 1 . Picea sitchensis - Gaultheria shallon soil data 137 12. Picea sitchensis - Gaultheria shallon vegetation data 138 13. Picea sitchensis - Eurhynchium oreganum plot data 145 14. Picea sitchensis - Eurhynchium oreganum soil data . 146 15. Picea sitchensis - Eurhynchium oreganum vegetation data 147 16. Picea sitchensis - Rubus spectabilis plot data 156 17. Picea sitchensis - Rubus spectabilis soil data 157 18. Picea sitchensis - Rubus spectabilis vegetation data 158 19. Tsuga heterophylla - Picea sitchensis plot data 166 20. Tsuga heterophylla - Picea sitchensis soil data o 167 21. Tsuga heterophylla - Picea sitchensis vegetation data 168 v i i i LIST OF TABLES (continued) Table Page 22. Picea sitchensis - Carex obnupta plot data 177 23. Picea sitchensis - Carex obnupta soil data 178 24. Picea sitchensis - Carex obnupta vegetation data 179 25. Picea sitchensis - Polystichum munitum plot data 188 26. Picea sitchensis - Polystichum munitum soil data 189 27. Picea sitchensis - Polystichum munitum vegetation data 190 28. Condensed Table of Forest Types Indicating Characteristic Combination of Species and Relationships 201 29. Comparison of Tree Growth and Volume Per Acre for the Forest Types of the Coastal Sitka Spruce Forest 202 30. Sitka Spruce Forest Types of the Coastal Western Hemlock Zone on Vancouver Island 239 o ix LIST OF FIGURES Figure Page 1. The Sitka spruce forest along Wickaninnish Bay 1 2. Instrument used to measure incoming ocean spray 19 3. Slide used to catch incoming ocean spray 19 4. Device used to collect throughfall precipitation 21 5. Coastal study area plot locations 59 6. Winter storm line showing the accumulation of beach logs and wind-blown sand 68 7. View of a forest transect along Wickaninnish Bay 68 8. Vegetation - Pacific Rim National Park 71 9. Mean daily temperature and mean precipitation by months for two stations on the west coast of Vancouver Island 77 10. Mean monthly relative humidity for four daily recording periods for two stations on the west coast of Vancouver Island 78 11. Monthly means of visibility and wind speed for two stations on the west coast of Vancouver Island 81 12. Seasonal amounts of ocean spray collected at six sites for a one-year period 88 13. Cumulative totals of ocean spray for six sites over a one-year period 89 14a. Ocean spray damage on current year leaves of Sitka spruce 97 14b. Shriveled leaf tips on current year's growth of Sitka spruce 97 15a. Delayed bud burst on Sitka spruce 99 15b. Stunted leaves of Sitka spruce following late bud burst 99 16. Severely deformed western hemlock contrasted by the normal growth form of Sitka spruce 100 X LIST OF FIGURES (continued) Figures page 17. Map of the study area showing the relationship between the width of the coastal Sitka spruce forest and the amounts of incoming ocean spray 103 18. Rocky headland site used in the study of vertical ocean spray distribution 110 19. Beach site used in the study of vertical ocean spray distribution 110 20. Vertical distribution of ocean spray on a rocky coast-line site and a beach site 111 21. The process of ocean spray generation along the rocky shoreline as i t occurs on Box Island 112 22. The process of ocean spray generation on beaches as i t occurs in Florencia Bay 112 23. A western hemlock - western redcedar stand damaged by ocean spray 119 24. Death of Sitka spruce due to ocean spray 119 25. Comparison of the mean weekly amounts of ocean spray received by Sitka spruce, western hemlock - Sitka spruce, and western hemlock - western redcedar forests for the 1967 growing season 121 26. The Picea sitchensis - Maianthemum dilatatum forest type 129 27. A Lithic Regosol 129 28. A close-up view of the Picea sitchensis - Maianthemum  dilatatum forest type 134 29. Polypodium scouleri growing on rock outcrop 134 30. The Picea sitchensis - Gaultheria shallon forest type 139 31. Orthic Ferro-Humic Podzol 139 32. The Picea sitchensis - Eurhynchium oreganum forest type 148 33. Orthic Regosol formed in sandy beach plain deposits 148 34. Interior view of the Picea sitchensis - Eurhynchium oreganum forest type 154 xi LIST OF FIGURES (continued) Figure Page 35. Herb layer in the Picea sitchensis - Eurhynchium oreganum forest type 154 36. Orthic Ferro-Humic Podzol formed in an aeolian sand deposit 160 37. The Picea sitchensis - Rubus spectabilis forest type 160 38. An Orthic Humic Podzol 170 39. The Tsuga heterophylla - Picea sitchensis forest type 170 40. A Gleyed Regosol 182 41. The Picea sitchensis - Carex obnupta forest type 182 42. A midden deposit associated with a grassland opening 192 43. A Orthic Humic Gleysols formed under seepage conditions 192 44. A "wolf tree" 197 45. Herb layer in the Picea sitchensis - Polystichum munitum forest type 197 46. Comparison of tree species' coverage by strata in the coastal Sitka spruce forest 204 47. Sequence and relationships of biogeocoenotic units in Cox Bay 206 48. Sequence and relationships of biogeocoenotic units on Long Beach 210 49. Sequence and relationships of biogeocoenotic units in Schooner Cove 212 50. Comparison of incoming ocean spray received by the coastal Sitka spruce forest types during the 1967 growing season. 216 51. Comparison of sodium content in throughfall collected in the coastal Sitka spruce forest types during the 1967 growing season 218 52. Comparison of soil moisture in the coastal Sitka spruce forest types during August, 1967 222 x i i LIST OF FIGURES (continued) Figure , Page 53. Comparison of pH, percent base saturation and cation exchange capacity for the coastal Sitka spruce forest types 226 54. Comparison of soil cation contents for the coastal Sitka spruce forest types 228 55. Comparison of throughfall amounts and ten centimeter replacement time of four cations for the coastal Sitka spruce forest types 230 e x i i i ACKNOWLEDGEMENT I should like to express my sincerest thanks to Dr. V. J. Krajina e for his expert and invaluable supervision of this work in its entirety. From the conception of the problem to the final writing of this thesis, he has provided encouragement and help when it was most needed. I am indebted to my Graduate Committee, Dr. K. Beamish, Department of Botany, Dr. P. Haddock and Dr. J. P. Kimmins, Department of Forestry, and Dr. G. E. Rouse and Dr. R. F. Scagel, Department of Botany, who provided further guidance and valuable criticism of the manuscript. Thanks are extended to Dr. V. J. Krajina for his assistance in identifying many of the vascular plants and most of the bryophytes, to Dr. C. Bird, University of Calgary, for identification of the lichens, to Dr. W. B. Schofield, Department of Botany, for identifying some of the bryophytes, to Mr. B. von Spindler, Department of Soil Science, for carrying out chemical analyses on the soils, to Mr. L. Armstrong, Mr. J. Masyk, Miss M. Medwid and Mrs. M. Croot for drafting the maps and graphs, to Miss N. Kariel, Miss D. Reidlinger and Miss. S. Thompson for data reduction, to Dr. A Kozak, Department of Forestry, and Mr. S. Borden for writing computer programs, to Mrs. E. Wittig and Mrs. M Burge for typing and Mr. N. Babey for photography and technical advice on the maps and graphs. I wish to thank my field assistants, Mr. John Cordes, Mr. Don Hansen and Mr. Don MacKenzie, who were always willing and able to carry on. I am grateful to the many residents of the West Coast for their xiv assistance, hospitality and friendship. In this regard, special mention should be made of Mr. and Mrs. E. Bach, Mr. Z. Borsos, Mr. D. Flintoff, Mrs. MacKay, Mr. and Mrs. R. MacKenzie, Mr. J. Martin, Mr. G. Pownell, Mrs. W. Saxton, Mr. J. Toovey and Mrs. P. Wittington. I should like to thank my collaborators on the Pacific Rim National Park study, Dr. J. G. Nelson, University of Western Ontario, and Mr. G. MacKenzie and Mr. N. Roe, University of Calgary, for their assistance and many fruitful discussions. Lastly, I would like to thank my wife, Janet, for her help, encour-agement and patience. This research was supported by a National Research Council of Canada Grant (No. A - 9 2 ) to Dr. V. J. Krajina, a National Research Council Studentship, a University of British Columbia Graduate Fellowship, and the Department of Botany, University of British Columbia. Figure 1. The Sitka spruce forest along Wickaninnish Bay. CHAPTER I 2 INTRODUCTION Initial interest in this study was stimulated by the unique distributional pattern of the Sitka spruce (Picea sitchensis) forest along the west coast of Vancouver Island. This region is almost com-pletely covered by de'nse stands of western hemlock (Tsuga heterophylla) , western redcedar (Thuja plicata) and Pacific silver f i r (Abies amabilis); however, this extensive forest is replaced by one composed of almost pure stands of Sitka spruce in a narrow strip along the coastline and on the flood plains and terraces of the major river valleys. The presence of the Sitka spruce forest in well-defined physio-graphic situations leads one to suspect that there is something very different about the environmental complex of these areas to allow a forest to develop which is so distinct from the typical forests of the region. One begins to search for hypotheses that might produce an explanation, or at least a partial explanation, for this interesting spatial pattern. Can i t be related to some climatic or edaphic factor or by some historical event related to geomorphic processes, some past human influence or to some catastrophic event? In this con-text, one wonders i f the presence of the Sitka spruce forest is a result of primary or secondary succession or i f . i t is a self-perpetuating system, a climax forest. This study was undertaken with several major objectives in mind. One of these was to analyze and compare the environment of the coastal Sitka spruce forest with that of the more inland western hemlock - western redcedar forest. This was approached on the assumption that consistent 3 differences between the two types of forest in the expression of certain environmental factors may lead to an explanation for the existence of the coastal Sitka spruce forest. In other words, the physical environment of the western hemlock - western redcedar forest was used as a standard or "norm" to which the environment of the Sitka spruce forest could be compared. However, due to the large amount of environmental variability present within both types of forest, this phase of the study tended-to concentrate on general trends rather than on specific site measurements since i t would have been impossible to sample, or even make observations, on the range of different habitats present. At the outset of the study, i t became apparent that wind-borne ocean spray was a factor deserving special attention. A review of the available literature revealed that, not only had very l i t t l e research been carried out on this factor along the west coast of North America, but that there was also a general paucity of knowledge on both the dynamics and ecological effects of ocean spray. Therefore, the author undertook a fairly detailed study of the ocean spray regime in an attempt to gain an understanding of its ecological role in the coastal Sitka spruce forest. It is hoped that this part of the study has also added to the general body of knowledge on the subject of ocean spray dynamics. The other major objective of this thesis was to carry out an ecological investigation of the coastal Sitka spruce forest as a complex entity in it s e l f . More specifically, the objectives were: 1) to obtain both quantitative and qualitative information on the coastal Sitka spruce-dominated plant communities, their soils and other elements of the physical environment; 2) to incorporate this information into an ecosystematic 4 classification scheme; and 3) to describe and interpret the vegetation-environmental relationships of these ecosystematic units. Common to a l l synecological studies is the question of what form of classification system will be both practical and philosophically acceptable for the characterization of plant communities and an evaluation of their environmental relationships. Comprehensive discussions and reviews of various approaches to these problems have been presented by Braun-Blanquet (1932), Krajina .,(1933, 1960a, 1965), Sukachev (1945, 1960), Major (1951), Billings (1952), Daubenmire (1966), Poore (1955, 1956, 1962), Becking (1957), Hills (1960), Rowe (1960a), Whittaker (1962), Waring and Major (1964) , and Sukachev and Dylis (1964). In recent years the ecosystem concept and its application in the classification of natural systems has been steadily gaining recognition. This concept has been used in the present study in an attempt to classify and characterize the Sitka spruce forest and as a basis for evaluating environmental relationships and successional, dynamics. While commonly understood that the components of the ecosystem are a l l interrelated and interdependent upon one another, i t is also realized that the number of factors involved is far too great to a l l be used as classification criteria. For this reason, Hills (1960) has delimited individual ecosystems on the basis of soil material and topo-graphy within a regional framework based on macroclimate. Once the units are established using these criteria, the plant communities and other components are described to produce the total ecosystem. At the opposite end of the scale, Sukachev (1960), also realizing that i t is impossible to deal with a large number of factors, has chosen 5 to delimit individual ecosystems, or biogeocoenoses, primarily by using the plant community (phytocoenosis). He then goes on to describe the remaining components of the ecosystem, particularly the edaphic elements (edaphotope), to obtain a complete characterization of the unit. Although these two approaches are quite different, both are correct in assuming that even though the total system is classified in terms of one or more of its parts, i t is the whole system and not the part which is classified (Rowe 1960a). Since, sooner or later, a l l essential parts of the ecosystematic units will be defined, the point of most concern in the classification methodology rests on identifying the component or components that will best serve as ini t i a l classification criteria. It is the author's belief that vegetation is the single best indicator of the sum total of elements making up the ecosystem and, therefore, basically followed the approach developed by Sukachev. However, certain structural elements of the environment such as physiography and spatial position in reference to the configuration of the coastline are strongly correlated with the plant communities and have considerable value in classification due to the ease with which they can be recognized in the field. These elements along with others of the physical environment can serve as important checks or as verification of a system based primarily on the vegetation. As proposed by Sukachev and Dylis (1964) each sample plot (stand) is considered to represent an individual ecosystem or biogeocoenosis characterized in terms of vegetation and environmental measurements obtained from the field analysis. Classification units were formed by 6 arranging the plots into groups on the basis of similarities and differences in their vegetation and habitat characteristics. The abstract unit so formed is referred to as the "type of the biogeocoenosis" or simply as "forest type." This unit is considered to closely approximate the plant association as -defined by Krajina (1959, 1960a). The forest types have also been named in accordance with the Zurich-Montpellier School for purposes of developing a hierarchical classification of the Sitka spruce forest. In order to properly develop this system at the higher levels of generalization, i t was necessary to incorporate some forest types that do not belong to the coastal Sitka spruce forest. These types belong to a group designated by the author as the "alluvial Sitka spruce forest" and were also studied during the course of the field work for this thesis (Appendix VII). 7 CHAPTER II REGIONAL DESCRIPTION , TOPOGRAPHY The west coast of Vancouver Island is a rugged area covered with mountains and dissected by U-shaped valleys with steep sides. Gently sloping and flat land is largely restricted to the valley bottoms. The Vancouver Island Range runs most of the length of the island and contains many peaks and ridges with elevations between 4000 and 5000 feet and some which reach to slightly over 7000 feet. The mountains extend right to the coast, or nearly so, for almost the entire length of the island. The only major area which differs from this description is the northwestern end where a dissected plateau occurs between Quatsino Sound and the tip of the island. The maximum elevations in this section are around 2300 feet and the slopes are not nearly so steep. PHYSIOGRAPHY The rocks which make up the island ranges are primarily volcanic and sedimentary in origin and are Paleozoic or younger in age (Heusser 1960). The Vancouver group which consists of andesite volcanics with intercalated beds of argillite and limestone is the most widely distributed (DoImage 1920). These rocks have been invaded by diorite and grano-diorite intrusives which are believed to be related to the Coast Mountains -batholith (Heusser 1960). Sandstones and fine conglomerates of Cretaceous age are prominent along the coast in some places (Dolmage 1920). The region has been strongly modified by glaciation. Much of the 8 surface is covered with a mantel of glacial t i l l , particularly on the middle and lower parts of the slopes. The major valleys were flooded by the sea during glacial recession resulting in many of the present fiords which cut deeply into the mountain ranges in Barkley, Clayoquot, Nootka, Kyuquot, and Quatsino sounds. These sounds along with several inlets are the only major indentations into the otherwise fairly regular coastline. The coast, in general,is rocky and the land rises abruptly from the waters edge. However, there are some rather extensive areas of beaches which have formed at the base of rocky c l i f f s or along raised marine terraces. Long Beach and the beaches in the vicinity of Estevan Point are noteworthy examples. THE COASTAL WESTERN HEMLOCK ZONE Most of the land is covered with dense coniferous forest. Whitford and Craig (1918) estimated that approximately 757o of this area was well timbered with the remaining 25% being either too high or too rocky to support such stands. According to Krajina (1965) the west side of the island can be divided into three biogeoclimatic zones. The Alpine Zone occurs at elevations above 5500 feet and makes up only a small part of the total area. The Mountain Hemlock Zone occupies areas beneath the Alpine Zone and extends down to around the 3000 foot contour. The major tree species in this zone are Tsuga mertensiana, Abies amabilis, and Chamaecyparis  nootkatensis. Picea sitchensis is rarely encountered in the Vancouver Island and Southern Mainland portion of this zone (Peterson 1964). o The forest between sea level and 3000 to 3500 feet belongs to the Pacific Silver Fir-Western Hemlock Subzone of the Coastal Western Hemlock 9 Zone. This zone is by far the most extensive and contains most of the merchantable timber. It is within this subzone that the Sitka spruce forest reaches its best development and where the research for this study was carried out. Krajina (1969) has described this zone and subzone-as follows: Macroclimate^: mainly Cfb (and the mildest Dfb) Annual total precipitation (inches): (44) 65 - 262 Annual snowfall (inches): 5 - 295 Driest month precipitation (inches): 1.2 6. Wettest month precipitation (inches): (6) 11 - 46 Mean annual temperature (°F): 41 - 49 Number of months above 50°F: 4 6 Number of months below 32°F: 0 3 Number of frost-free days: 120 - 250 Elevation (feet): British Columbia: North : 0 - 1000 South : 0 - 3000 1500 - 3500 Soil: Prevailing pedogenic processes: (a) mor formation, (b) podzolization, (c) gleization. Mesic (zonal) soils (in the wetter subzone): Orthic Humic Podzols. The Wetter (Perhumic) or Pacific Silver Fir-Western Hemlock Subzone is characterized by a perhumid or rainy climate with an annual total precipitation of 110 - 262 inches and by the following coniferous trees: Pseudotsuga menziesii var. menziesii, Tsuga  heterophylla, Thuj a plicata, Abies amabilis, Picea sitchensis, Pinus contorta, Pinus monticola, Chamaecyparis nootkatensis and. Tsuga mertensiana. Western hemlock, Sitka spruce, Pacific silver f i r and yellow cedar have habitats with their best site index in this subzone. Alnus rubra and Populus balsamifera ssp. trichocarpa are frequent here after logging in suitable habitats. Macroclimate symbols (Koppen): C climates: warm temperate rainy climates (humid mesothermal); average temperature of coldest month below 64.4°F, but above 26.6°F; average temperature of warmest month over 50°F. f : no distinct dry season; driest summer month with more than 1.2 inches precipitation. ° b : cool summer; average temperature of warmest month under 71.6°F. 10 According to Orloci (1961) the mesic (zonal or climatic climax) plant community in the Wetter Subzone is the Abieteto-Tsugetum heterophyllae association. He l i s t s the following species as being characteristic for this association: Major Tree Species Western hemlock followed by western redcedar are the two most abundant tree species of the subzone. Hemlock along with lesser amounts of Pacific silver f i r form extensive stands on mesic sites. Xeric to submesic sites support western hemlock, Douglas-fir, and lodgepole pine either singularly or in combination. Subhygric and wetter sites are usually dominated by western redcedar along with considerable amounts of western hemlock and Pacific silver f i r . Douglas-fir may occur on most sites as a pioneer species, but since neither fires nor logging were very prominent forces in the past the area supporting mature stands of this species is very small. In recent years i t has been planted extensively after logging where i t grows very well on most sites. The Sitka Spruce Forest Within the Wetter Subzone stands of Sitka spruce are almost completely For a definition of constant dominant and constant species see page 23. Constant dominant species : Tsuga heterophylla  Abies amabilis  Blechnum spicant  Rhytidiadelphus loreus  Hylocomium splendens Thuja plicata  Vaccinium alaskaense  P1agiotheeium undulatum  Mnium glabrescens Constant species (not dominant): Hypnum circinale Isothecium stoloniferum Scapania bolanderi Dicranum fuscescens 11 restricted to a narrow strip along the open coast and to the flood plains and terraces of the major valleys. The cumulation of these stands makes up what is here referred to as the "Sitka spruce forest." For purposes of this study the Sitka spruce forest has been divided into the "coastal Sitka spruce forest" and the o"alluvial Sitka spruce forest." Although both the coastal and alluvial forests were studied in the field, only the coastal Sitka spruce forest is considered in the body of the thesis. However, descriptions of the alluvial spruce forest types have been included in Appendix VII to facilitate comparison with the coastal spruce forest types for those who may wish to do so. The inclusion of the alluvial spruce forest types was also necessary for the formation of a comprehensive hierarchical classification of the Sitka spruce forest (Chapter Vl). While most of the Sitka spruce on the west coast occurs in one or the other of these two major groups, there is a minor spruce component over the remainder of the area. This may take the form of scattered trees throughout the extensive western hemlock, western redcedar, and Pacific silver f i r dominated forests or of almost pure stands. Stands with a high percentage of spruce are occasionally found on seepage sites i f spruce becomes established before western redcedar. These two species are sometimes found growing together, especially in subhydric habitats. Sitka spruce is unique among Pacific Coast conifers in its ability to withstand slightly saline soils (Krajina 1959, 1969). Scattered spruce may be found growing under these conditions along estuaries and tidal flats therefore, its other common names are "coast spruce" and "tideland spruce." 12 CHAPTER III METHODS This study basically employs the methods used by V. J. Krajina and his students in a l l of their synecological studies of British Columbia ecosystems ('see Krajina 1969) . The author believes that the consistency in methodology used throughout these studies has been an important contributing factor to the advancement of plant ecology in western Canada. That part of this methodology related to the vegetation is more commonly associated with European phytosociologists and has been described in detail by Braun-Blanquet (1932). Excellent discussions of the philosophy are given by Poore (1955, 1962) and Whittaker (1962). Methods used to study the physical environment were taken from a number of sources. For certain facets of the study, particularly those related to the examination of the ocean spray regime, new methods were devised. It is felt that these methodological developments and the information derived through their application represent some of the more important contributions of this study. ANALYSIS1 Prior to sampling, a considerable amount of time was spent in the field to gain familiarity with the flora, vegetation and physical Field work was carried out between June, 1965 and September, 1971. 13 environment. During this period, locations for the first group of sample plots were selected. A deliberate attempt was made to spread these plots throughout the study area as well as to choose sites which represented a wide range of vegetation, parent material and topography. No attempt was made to form a preliminary classification at this time. Prior to sampling, a decision was made to limit the study to stands in which Sitka spruce made up a significant part of the forest 3 cover. At the same time i t was decided to restrict the study to stands exceeding 50 years in age. This was done in an attempt to keep the scope of the study within manageable limits. In addition to meeting the above criteria, for a site to be acceptable for sampling, i t had to be rela-tively uniform in both vegetation composition and structure and in environment. The author believes the subjective sampling, i f carefully applied, has certain advantages over the method by which plots are randomly located. Plots located subjectively can largely avoid heterogeneity in vegetation and environment. This is particularly important when only one set of sample data per plot is obtained. In this case the data represent averages for the entire plot and, therefore, i f a high degree of variability exists, these data may be very misleading. Selective sampling also allows one to adequately sample the less frequently occurring ecosystems which may have significant ecological implications. At the same time, the more common V. J. Krajina accompanied the author during part of this period. Significant is defined here as 30% or more of the forest canopy. 14 ecosystems are not oversampled, allowing the researcher to make the best use of his time. The normal sample plot size was 66 x 132 feet (approximately 2 800 m ) or 1/5 acre. However, for several of the ecosystem types, stands were small or irregular in shape necessitating the use of plots with 2 dimensions of 33 x 132 (approximately 400 m or 1/10 acre), or 33 x 66 feet 2 (approximately 200 m or 1/20 acre). Once a suitable location for sampling had been selected, a plot was laid out using a steel tape and a Brunton compass. The first step in data collection was to describe the habitat in terms of land form, relief shape, microrelief, aspect, slope, exposure to the ocean, distance from 4 the coastal tree line and wind exposure. Estimates were also made on the percentage of the ground surface covered by humus, rock, decayed wood and mineral soil. Vegetation Vegetation was analyzed using the techniques of the Zurich-Montpellier School adapted by Krajina (1933). Visual cover estimates were made on each of the following strata: Tree layer: Shrub layer: Herb layer: Moss layer: - dominant trees A£ - co-dominant trees A^ - intermediate and suppressed trees over .30 feet in height B^  - saplings and shrubs between 6 and 30 feet in height B2 - saplings and shrubs between 6 inches and 3 feet in height C - herbaceous plants and woody plants less than 6 inches in height - bryophytes and lichens cn the humus, or mineral soil See Appendix II for explanation of terms. 15 - bryophytes and lichens on rock - bryophytes and lichens on decayed wood Epiphyte layer: E - bryophytes, lichens and ferns on aerial parts of woody plants The next step was to l i s t a l l species of vascular plants, bryophytes and lichens present in the plot according to the vegetation strata in which they occurred."* Each entry on this l i s t was rated according to the Domin-Krajina scale which combines both abundance and cover. This scale is as follows: + - Species very sparse, cover negligible 1 - Seldom, cover negligible 2 - Very scattered, cover negligible 3 - Scattered, cover up to 5%> of plot 4 - Common, cover 5-10% of plot 5 - Often, cover 10-20% of plot 6 - Very often, cover 20-35%. of plot 7 - Abundant, cover 35-50%, of plot 8 - Abundant, cover 50-75% of plot 9 - Abundant; cover 75-95%, of plot 10 - Abundant, cover 95-100% of plot Vigor was also estimated using a seven-part scale (see Appendix II). Tree mensurational analysis included a tally of a l l trees over 6 feet tall in the coastal spruce forest and a l l trees over 30 feet tall in the alluvial spruce forest. Height and diameter at breast height were recorded for a l l individuals. Age was determined for a minimum of 12 trees in most plots, although considerably more were sometimes taken depending on the complexity of the forest. Trees were selected to represent the range of species and diameter classes present. From this ^Nomenclature and authorities for the vascular plants, with few exceptions, follows Hitchcock et al. (1955, 1959, 1961, 1964, 1969), Schofield (1969a, 1969b) for bryophytes, and Hale and Culberson (1960) for lichens. 16 data density, basal area, gross volume and site index were calculated for each species.^ Soil A soil pit was located in each plot and dug to the depth of unweathered parent material. Profiles were described by horizon according to proposals of the National Soil Survey Committee of Canada (1963) and the United States Soil Survey Manual (1951). These descriptions included the following: horizon depth and thickness, structure, consistence, moisture, presence and type of mottling, stoniness, root distribution, depth of water table, and presence of charcoal. Bulk density was taken for each horizon by means of a brass ring. Samples were collected for laboratory analyses. Prior to chemical analyses, each sample was passed through a 2 mm sieve to remove the coarse fraction. The less than 2 mm fraction was hand-textured by Mr. B. von Spindler, Department of Soil Science, University of British Columbia. Determination of carbon, total nitrogen, total phosphorus and cation exchange capacity were also done by Mr. von Spindler. The determinations of pH and exchangeable sodium, potassium, calcium and magnesium were carried out in the Department of Botany. Using the results of these analyses, values for per cent organic matter, carbon-nitrogen ratio and per cent base saturation were calculated. The analyses follow standard methods used by the Department of Soil Science, University of British Columbia and are described by Beil (1969). The results of the chemical analyses are presented in the synthesis ^Site index at 100 years was determined using curves from Meyer (1937). 17 tables according to the humus layer, the major mineral horizon and g total profile. Exchangeable cations and cation exchange capacity are presented on 3 both a weight basis (meq/lOOg) and a volume basis (meq/lOOcm ). Conversion to the volumetric measure was made by using the bulk density. The volumetric measure is of considerable use in comparing horizons and in calculating the exchangeable cation values and cation exchange capacity for a given depth of soil or for the total profile. Measurements of water retention properties of the soil were made using a porous plate extractor and following methods proposed by Richards (1939). Available moisture retention was calculated as the difference in moisture retention between 1/10 and 15 atmospheres negative pressure (Richards and Weaver 1944, Lehane and Staple 1960). All data were converted to water per cent by volume (volume of water/volume of soil = water per cent by volume). The per cent pore space was calculated using mean particle densities of 1.34 for organic matter and 2.65 for the mineral soil. All results are expressed for the surface horizon, major mineral horizon and the total profile. Samples for field moisture determination were taken by horizon immediately after digging the soil pit. The fresh samples were weighed, oven dried at 105°C for 24 hours and weighed again. Moisture was calculated as a percentage of the soil volume and on a depth basis ^Major mineral horizon is defined as the thickest mineral horizon in the rooting zone. It is either the B horizon or the C horizon in regosols. g The total profile has a depth equivalent to the depth of the main concentration of roots. 18 (cm soil water = cm soil depth x water per cent by volume). Available field moisture was obtained by subtracting the 15 atmosphere moisture retention from the field soil moisture. The corresponding figure for volume of air-filled pore space was calculated by subtracting the field moisture from the total porosity. Soils were classified according to the National Soil Survey Committee of Canada (1970) and based on field descriptions, soil color using Munsell soil color charts and chemical analyses (Appendix IV). Ocean Spray To measure ocean spray i t was necessary to first design and build instruments since none were available on the market. A modified version of an instrument described by Edwards and Claxton (1964) was designed and built. Following field tests, further modifications were made to overcome problems created by heavy precipitation. The instrument consists of a 3 inch diameter tube and a one foot square tail mounted on opposite ends of a shaft (Figure 2). The shaft rotates on top of a 4 foot high post so that the open ends of the tube are always parallel to the direction of the wind. A wire frame holding a one square inch piece of filter paper is mounted inside the tube and oriented in a plane perpendicular to the long axis of the tube (Figure 3). Ocean spray passing through the tube is deposited on the fi l t e r paper. Upon removal from the instruments, the f i l t e r papers were placed in 100 ml of distilled water, shaken for 30 minutes and allowed to stand for 12 hours or more. The resulting solutions were analyzed for sodium, magnesium, potassium and calcium on a Perkin-Elmer, model 303, atomic adsorption spectrophotometer. The results were expressed in grams of incoming cations per square meter Figure 2. Instrument used to measure incoming ocean spray. Figure 3. Slide with f i l t e r paper target used to catch incoming ocean spray. 20 2 orientated in a plane perpendicular to the wind, per week (g/m /wk) for each of the four cations. Measurements were taken on a weekly basis for the 1966 growing season and for the 1966-67 winter and spring. Two-week periods were used for the 1967 growing season. Ocean spray was also studied by measuring the concentration of different cations in rainwater or fog drip which had passed through the tree canopy. Throughfall was collected in small pots which were placed directly on the ground (Figure 4). The shrub layer around the pots was removed so that i t did not interfere with the throughfall. Four pots were placed at each station and the water collected from them was combined to give one sample. Determination of cations followed the same methods used for the ocean spray samples. Results were expressed in grams of each cation per square meter of ground surface per week. The throughfall method gives a relative measure of the amount of ocean spray intercepted by the tree canopy in different stands, as well as an absolute value for the weight of cations deposited on the soil . surface. SYNTHESIS OF BI0GE0C0EN0TIC UNITS The synthesis stage included two basic steps: (l) the formation of the classification units and (2) a comparison of these units on the basis of the major components studied. Formation of the classification units involved grouping similar plots together into biogeocoenotic units, or forest types. This was done using vegetation composition and structure as the primary criteria. Secondary criteria included soil morphology, parent material, topographic position in reference to drainage and spatial location. The secondary criteria were generally used too verify plot groupings made on the basis of vegetation. In most cases there was close correspondence of all criteria and in many cases major breaks in 21 Figure 4. Device used to collect throughfall precipitation. 22 the secondary criteria correlated with strong differences in vegetation. One of the major advantages of this classification method is that i t uses structural elements which can be easily observed and measured without time-consuming techniques and sophisticated instrumentation. This means that, hopefully, the classification system devised can be understood and used by someone besides the author. Although most plot groupings and forest types became evident during,the sampling period, the classification was not finalized until complete analytical data for a l l plots was available. Using a synthesizing procedure very similar to the process of "successive approximation" as described by Poore (1955, 1962) and Brooke et al. (1970), plot lists of the analytical data were compared for similarities and differences. Groups of plots were formed and tested for conformity of the data. This process went on until, in the author's judgement, the degree of consistency of the data within groups had been maximized. At this point, the classification process was considered to have been completed and each group was designated as a forest type. Presence, per cent cover and dominance were the main criteria for classifying and describing the forest types. Fidelity (Braun-Blanquet, 1932) was of limited value because of the relatively poor flora of the area. Five presence classes were used as follows: Class Percentage Occurrence I 0-20 II 21 - 40 III 41 - 60 IV 61 - 80 V 81 100 23 Species which occurred in 81% or more of the sample plots of a forest type are described as constants. Mean cover for each species was calculated by averaging the midpoints of the cover values in the species significance scale (Appendix II). Constant species with a mean cover of 10% or greater in any one strata are described as constant dominant species for that forest type. Certain species with a presence of less than 81%, but which have high cover values or are largely restricted to one or several forest types are designated as important non-constants. This designation may be subdivided as follows (Braun-Blanquet, 1932): Exclusive species - are completely or almost completely confined to one forest type. Selective species - are found most frequently in a certain forest type. Preferential species - are found most frequently in several forest types. As a basic rule, forest types were described and compared strictly on the basis of strata. The designation of constant, constant dominant and non-constant species also followed this rule. The second step in synthesis involved a comparison of the forest types on the basis of mean values and ranges of the major components studied. This was particularly useful in bringing out similarities and differences in components which required instrumentation for their analysis and were, therefore, not used as classification criteria. This procedure led to a better understanding of the dynamics of the forest types, since these components are dynamic in nature themselves. Components of particular value in this regard were soil moisture, soil chemical parameters, ocean spray and throughfall. 24 CHAPTER IV OCEAN SPRAY: A LITERATURE REVIEW This literature review is an attempt to bring together the important published works on the subject of ocean spray as an environmental factor. To the author's knowledge, this is the first attempt at such a review. It is realized that some works may have been missed due to the wide range of disciplines in which material on the subject has been published. GENERAL CHARACTERISTICS OF OCEAN SPRAY Ocean spray refers to droplets of salt water which are derived from sea water and are present in the air above the ocean and nearby land surfaces. Wilson (1959) describes i t as quite noticeable "haze." Edlin (1943) and many others refer to i t as a "salt-laden wind" blowing inland across the ocean. Daubenmire (1959) as well as a number of others have used the term "wind-borne salt spray." Droplets of ocean spray may vary in size, depending upon the mode of formation as discussed by Knelman, Dombrowski and Newitt (1954), by Woodcock (1962) and by Stuhlman (1932); and upon the consequent evaporation of the droplets when exposed to air (Boyce 1954, Edwards and Claxton 1964). The droplets may be evaporated to some extent upon exposure to air: the rate being dependent on the size of the original droplet and the relative humidity. The concentrated droplets of sea water are referred to as "salt nuclei" (Edwards and Claxton 1964). Since some of the water from the droplets is evaporated, the concentration of dissolved salts is higher in 25 spray than in sea water. Salts need not be dissolved in water but may be carried in the air as salt crystals (Karschqn 1958). L i t t l e attempt has been made to determine or analyze the chemical composition of ocean spray beyond recognizing that i t contains salts. Many workers such as Oosting (1945), Karschon (1958) and Boyce (1954) do not make any distinction between sea water and ocean spray in terms of composition. Oosting (1945) used "ordinary sea water" to spray his experimental plants in order to simulate ocean spray without the wind factor. Since ocean spray originates from sea water, i t can be assumed that most, i f not a l l , of these constituents should be present in the spray.^ It might also be assumed that the composition of the various constituents in the spray should be somewhat similar to those in ocean spray, at least in a relative sense. Although some measurements of the composition of ocean spray have been made, they have been very limited as well as incomplete. Analyses have been limited to measuring one or several of the constituents and using this as an indicator of the quantity of ocean spray. Oksbjerg (1952) used the chlorine ion concentration as an indicator of the amount of ocean spray for his study on coastal vegetation in Denmark. In his studies on the east coast of the United States, Boyce (1951b, 1954) also used the chlorine ion. Edwards and Claxton (1964) used sodium in studying the distribution of ocean spray along the coast of Ireland. Other workers including Oosting (1945) and Karschon (1958) have limited their analysis *See Harvey, 1960 for a description of the chemical composition of sea water. to measuring the quantity of sodium chloride. Ingham (1950) measured nitrogen and potassium in water that dropped off trees in a South African coastal forest. In New Zealand, Wilson (1959) analyzed samples from the snow surface and concluded that potassium and nitrogen were deposited here by ocean spray. Formation of Ocean Spray Sylwan (1923) cited in Oksbjerg (1952) believed spray to be formed by the friction of strong winds blowing across the ocean. A number of refinements since this time have revealed that i t is the effervescence of air bubbles on the surface of the ocean which forcefully eject droplets of sea water into the air (Woodcock 1952, Boyce 1954, Edwards and Claxton 1964). Not all droplets are carried upwards by turbulence either: due to such factors as droplet diameter, settling velocity of the droplets, and the carrying capacity of the wind (Boyce 1954). The mechanics of droplet formation will determine, in part, the size of the droplet. Droplet size is important in dispersal, as only particles in a certain size range can be transported by a given wind velocity. Woodcock (1952) has carried out a number of studies on the formation of ocean spray. He concluded that the formation of ocean spray was initiated by the momentary trapping of air in the form of bubbles as a result of meteorologically induced disturbances; such as the breaking of wind-waves, the impact of raindrops and the melting of snow and hail. Upon reaching the surface, these bubbles burst and several droplets of sea water are projected into the air from the tip of a jet which forms as the bubble cavity collapses and is f i l l e d with water. Most of these droplets have a radius of 5 to 100 microns and are ejected to a height 27 of 0.5 to 5 centimeters above the surface of the water. Several other mechanisms have been described for the formation of ocean spray. Boyce (1951a, 1954) measured droplets produced by breaking waves and found that they ranged in size from 0.5 to 20 millimeters. He concluded that these droplets were produced by the splashing effect of the waves rather than by the bursting of bubbles. He also measured the size of ocean spray droplets at varying distances from the waves and concluded that the droplets produced by wave splash were carried only very short distances under normal wind speeds. In their studies of bursting bubbles, Knelman, Dombrowski, and Newitt (1954) concluded that, in addition to the bubbles formed by the collapsing of the bubble cavity, smaller droplets were formed from the shattered surface film of the bubble. However, Junge (1957) found that these very small droplets made up less than one per cent of his ocean spray samples. Boyce (1954) suggests that any disturbance on the ocean which forms small bubbles would be a source of ocean spray. Edwards and Claxton (1964) agree on this point, but suggest that the most frequent occurrence of bursting bubbles is associated with the surf line, thereby implying that this region is a more important source of spray. In his earlier work, Boyce (1951a) discusses the relative importance of certain regions in the zone of the breaking waves as sources of air-borne spray which may be carried inland by winds of "normal" velocity. His experi-ments showed that: (l) l i t t l e or no spray is found on the seaward side of the region of the breaking waves; (2) the largest droplets are found immediately above the breaking waves; (3) numerous small air-borne droplets are found above the swash and spume between the breaking waves and the strand; and 28 (4) the smallest and most numerous droplets are found above the strand. It was also noted by Boyce (1954) that bursting bubbles in the foam which is deposited along the shore do not forecefully eject droplets into the a i r . He concluded that the main source of air-borne spray was formed in the swash region and not in the region of cresting and breaking waves. While many large droplets were formed by breaking waves, they were considered to be too heavy to become air-borne and, therefore, did not contribute substantially to spray carried inland by winds of "normal" velocity. This is not to discount the possibility that the large droplets could not reach land. They may be carried inland by winds of high velocity, or be thrown sufficiently far when a wave breaks on a rocky shoreline. These droplets may, in fact, be a significant source of spray received by vegetation in' the v i c i n i t y of a rocky shore. A factor which may influence the potential supply of ocean spray i s wind. An increase in wind velocity w i l l result in larger waves which, in turn, produce a greater number of both large and small droplets as these waves break along the coastline. Breaking waves beyond the influence of the coastline occur during periods of high wind velocities, resulting in an additional source of small spray droplets (Boyce 1954). The previous discussion leads logically to the conclusion that the most important factor in the formation of ocean spray is the zone of breaking waves along the coast. The characteristics of these waves, particularly their size and form, w i l l determine both the number and size distribution of the spray droplets. In general, larger waves w i l l produce a greater number of droplets, although wave form may be more 29 important under certain conditions. Wave form can be broken into two major groups: surf waves along beaches and waves which break onto the rocky coastline. The former will usually produce a much greater propor-tion of small droplets because of the large area of swash while the latter will produce a greater proportion of large droplets which are generated by the process of wave splash. Dispersal of Ocean Spray Air movement, or wind, plays the major role in the dispersal of the ocean spray droplets once they have been ejected into the air. The mechanism by which larger droplets are thrown inland by the physical force of the cresting and breaking waves is of minor importance in com-parison. In order for the spray droplets to be carried any significant distance by the wind, they must first be ejected above the air-sea boundary. The process by which this takes place has already been dis-cussed. It might be added that pockets of low pressure, or partial vacuums, at the surface created by air turbulence accentuate this process. It is unlikely that spray droplets can be picked up from the surface of the water because of the low wind velocity which occurs here (Boyce 1954). However, once the droplets have been ejected above the air-sea boundary, they are quite easily carried by turbulence created by wind moving over the rough ocean surface. In general, winds of "normal" velocity are not sufficient to transport the larger droplets any significant distance. The air turbulence is not great enough, the carrying capacity of the wind is too 3 0 low, and the settling velocity of the droplets is too great. Boyce (1954) and others discuss the physics behind this statement. It is usually only the smaller, lighter droplets which are carried the distance necessary to reach land. The intensity of the spray is normally measured in terms of the concentration and intensity of the salts carried by the wind. Some devices used to "trap" and measure the incoming salts are as follows: Woodcock and Gifford (1949) used glass slides treated with a chemical preparation called Dri-film. The slides were enclosed in a special apparatus so that the slide surface was oriented normal to the wind. Boyce (1951b and 1954) used oiled glass slides after Haughton and Radford (1938), to collect spray. He also used a cardboard "leaf" to collect spray and to measure deposition under."normal" conditions. Oosting (1954) set up cheesecloth "salt traps" at various loca-tions on the beach and distances from the sea. They consisted of cheesecloth stretched across a frame. A standard size piece was soaked and the salt content determined by t i t r a t i o n . Ambler and Bain (1955) cited in Edwards and Claxton (1964), constructed a "wet candle" apparatus which, according to Edwards and Claxton, was less satisfactory than their method. The method of Edwards and Claxton (1964) consisted of a collecting surface oriented normal to the wind. The device consisted of a long cylinder with a glass slide attached across the inside diameter. It is kept properly oriented by a wind vane attached to a pivot. 31 The most satisfactory devices are those which are sufficiently protected from the rain which could significantly change the readings. From the measurements of salts deposited on these devices or washed from vegetation, Wells and Shunk (1938), Moss (1940), Oosting and Billings (1942), Edlin (1943), Boyce (1954) and Edwards and Claxton (1964) concluded that the concentration of salts and hence, ocean spray, is greatest during storms and gales when the wind velocity is high. The spray is also carried a greater distance inland during these periods of exceptional winds, as the above authors have also verified. Boyce (1954) found that at wind velocities between 5.5 and 7.5 meters per second, the intensity of ocean spray increased greatly. This was attributed to the increased formation of bubbles, increased ejection of droplets and the increased carrying capacity of the wind. The amount of ocean spray being carried inland is greatest when the winds are "strong and dry," and, of course, when they are blowing inland from the ocean (Edlin 1943, Oksbjerg 1952). This is partly a function of evaporation. "Evaporation reduces the weight of the droplets and hence, their settling velocity, so that in dry air, a wind of a given force will carry a higher concentration of salt a greater distance than would a wind of the same strength under conditions of higher humidity" (Edwards and Claxton 1964). Precipitation accompanying a strong wind would "wash out" the salts according to Oksbjerg (1952). It is also important to consider the intensity of ocean spray which may reach a coastline under "normal" circumstances as this factor 32 has considerable consequences for the vegetation and soils of the affected area. The distance or zone over which spray is carried was referred to as the "spray zone width" by Wells and Shunk (1938). It was taken to be the distance from the sea over which salt spray had some deleterious effect on a given species. This measure seemed to vary greatly from one area to another. The reason for this has been explained as being a function of the topography and direction of the prevailing wind (Wells and Shunk 1938), and of the orientation and character of the coastline (Boyce 1954). Two examples adapted from Boyce (1954), may help to c l a r i f y the preceding statement. If the prevailing winds are from the ocean to the land—and thus carry ocean spray—and a particular section of coast-line is sheltered from the winds by a large rocky headland or some other topographic feature, then the coastline w i l l not receive the intensity of spray i t might i f i t was not protected from the salt-laden wind. Another example may be the case of a shore not protected by topographic features from the winds. Yet, because of the orientation of the coast-line and the alteration in i t s trend, only small waves are produced, and therefore, a relatively small amount of surf and, consequently, spray is produced for transport. The shore therefore receives a relatively low intensity of spray and s a l t s . It is reasonable to assume that a section of coastline which is orientated perpendicular to the direction of the prevailing winds and which has an unlimited reach i n the direction of the ocean w i l l receive o the highest possible amounts of ocean spray under a given set of climatic conditions. Any section of coastline which deviates from the 33 above stated conditions, either in orientation or reach, should receive a lesser amount of ocean, spray. This generalization, of course, assumes that other factors which might affect the ocean spray are similar in each case. Deposition of Ocean Spray Once the ocean spray reaches shore, i t can either be intercepted by the land surface, the vegetation or other obstacles. Karschon (1955) studied the process of spray deposition and provides the following explanation: Air currents are deflected by objects such as a coastal forest; the velocity of the wind is subsequently slowed; the less turbulent air is capable of holding fewer and smaller spray droplets, that is, its carrying capacity is reduced and the spray droplets are deposited on contact. Oksbjerg (1952) and Sylwan (1923) believe that the wind is filtered by the vegetation or topography which causes friction with i t . Sylwan (1923) further explains that the salt is removed from the air as "lees create turbulence in which mixing of lower air layers of high salt concentrations with upper, salt free, air layers occurs." The distribution of spray-deposited salts on vegetation and soils has been examined by a number of researchers. The following observations have been made and reconfirmed several times: The highest content of salt is deposited on the windward side of obstacles, in this case vegetation. Wells and Shunk (1938) offer the following explanation as to why the deposition of salts is less on the 34 leeward side of vegetation and less on vegetation which is slightly depressed: "A strong wind playing over a porous surface would set up a suction effect...which would result in a slight elevation of the spray laden winds above the growth...a similar effect would keep the spray from falling behind." Boyce (1954) confirms this to some extent but suggests that shoots immediately behind twigs or branches receiving spray, do receive some spray, but those further back receive progressively less. The effect of shelter or some form of protection from winds carry-ing ocean spray has been examined by the following: Wells and Shunk (1938), Moss (1940), Edlin (1943), Boyce (1954), Karschon (1958), Karschon and Heth (1958) as cited in Edwards and Claxton (1964), and Edwards and Claxton (1964). Others, such as Hofstra and Hall (1971) and Westing (1969), have made similar findings in their work on the effects of highway salting on vegetation. Many of these studies indicate that topographic features or windbreaks significantly affect the amount of salts deposited. Vegetation fully exposed to the salt-laden wind will generally receive a greater intensity of salt deposition than that which is protected or sheltered from the wind, given that they are the same distance from the source of spray. Vegetation which is a consider-able distance from the sea but exposed to the winds can accumulate large quantities of salts, perhaps even more than vegetation which is closer but more sheltered. The seaward vegetation or topography acts as a fi l t e r or strainer, and removes salts from the winds. Thus the inner trees of a stand, the inner branches and leaves of canopies, and the leeward branches or leaves of plants will receive less salt than those which are more exposed to the wind. The concentration of salts deposited apparently decreases inland. Measurements have been made on the relative external and internal con-centrations of salts in plants at given distances from the sea (Wells and Shunk 1938, Moss 1940, Edlin 1943, Karschon 1958, Boyce 1951b and 1954, Edwards and Claxton 1964. It was discovered that while the "spray zone widths" varied quite markedly, the relative concentrations of salts did decrease away from the sea. Edwards and Claxton (1964) measured the distribution and concentration of salts along a transect perpendicular to the coastline about 4 miles in length. They found that less salts were deposited at a distance of about 2 miles from the sea than at a distance of only about 1 mile. There was no significant f a l l in salt deposition at greater distances. Fujiwara and Umejima (1962), performing a similar experiment along a mile long transect, found that there was a rapid f a l l of the CI ion content in the first 500 meters from the sea. Over the next 1000 meters the f a l l in the CI was slower but more irregular. There did not seem to be any measurable pattern of decline in CI content beyond 3000 yds. from the coast. Boyce (1954) has studied the efficiency of certain plants and plant organs as collectors of salt from salt spray. He found that long narrow leaves and twigs were more efficient collectors of small salt spray droplets than were broader leaves. Hence, the concentration of salts per unit area was greater. The spray was not concentrated around the tips and margins of the leaves and twigs, but was evenly distributed on the surface of the leaf. Where the salt is deposited on the plant organs, depends on the orientation of the plant to the incoming wind. 36 Boyce, in the same a r t i c l e , investigated the matter of deposition of salts on the canopy as a whole. An irregular canopy or crown which has leaves, twigs or shoots projecting above the average level of the canopy i s a more efficient collector of salt spray particles than i s a regular or more uniform oanopy. The projecting leaves and twigs collect the spray and offer protection to the more leeward foliage. On a broad uniform canopy, which i s rather hedge-like, there is less deposition of salt on leaves which are just below the average level of the canopy. Even the upper leaves do not receive the intensity of salt that the upper projecting leaves of an irregular canopy do. In other words, the irregular canopy has a higher "impact deposition efficiency" than a large uniform canopy (Boyce 1954). The previous discussion has been limited to the process of deposi-tion of ocean spray on vegetation. The amount of salt which accumulates on the surface of the vegetation may or may not be closely related to the intensity of the ocean spray. Martin (1938) states that, "the maximum amount of spray... which can be carried by a vertical (collecting) surface i s determined mainly by the wetting properties of the spray upon the particular surface." According to Edlin (1943), bright sunshine concentrates the salts in a salt film as the droplets are evaporated. In fact, any factor such as low humidity, high temperature or increased wind speed which contributes to an increased evaporation rate can effectively influence the accumulation of salts_pn vegetation (Harding, Miller and Fireman 1956). If the droplets are evaporated rapidly, then the surface of the vegetation w i l l not remain or become wetted. 37 Additional salt spray will not drip off but will form a "light bloom" of crystals on the foliage. This has been observed by Moss (1940), Edlin(1943) and Oksbjerg (1952). It has long been recognized that precipitation will wash off most of the salts which have accumulated on the foliage (Edlin 1943, Boyce 1954, Oksbjerg 1952, Edwards and Claxton 1964). Boyce (1954) noted that heavy fog tends to remove salts which had accumulated on the vegetation. It is quite possible that water droplets could form on the vegetation by condensation during periods of high humidity. If conden-sation was sufficiently high to cause droplets to form and drip to the ground, salts could be removed in the process. The hydrophylic nature of the salts could be expected to increase the rate of condensation. The deposition of salts on soils may occur directly, by the mechanisms described; indirectly, from tree' drip which contains the salts deposited directly on to the foliage from the spray (Beall 1934, Ingham 1950, Long, Sweet and Tukey 1956, Westing 1969) which washes salt particles out of the air (Oksbjerg 1952). The distribution of salts deposited directly as a "salt cake" follows the general pattern as described previously. However, i t is rather more difficult to measure the amount of salts deposited on soils that i t is to measure salt deposition on vegetation or other objects, because the soluble salts tend to become part of the soil solution quite rapidly. It is then a question of whether salts are contributed to the soil by ocean spray or whether some other source is responsible, for example tidal inundation. 38 The pH value has been used as an index to determine the relative amounts of ocean spray present in the s o i l at different locations. Oosting (1954) sampled soils along the strand and beach and found that a greater amount of soluble salts were deposited on the windward side of dunes as compared to the leeward side. The windward side of the foredunes had the highest concentration of soluble salts in the soil while the leeward side of the rear dunes had the lowest concentration. After a rain, concentrations were reduced in a l l sites and the leeward side of the foredune had the highest value. Doutt (1941) found a uniform pH in the soils which was independent of the distance from the sea. Salts from ocean spray are believed to be added to the soil by tree drip, or throughfall as i t is sometimes called. Beali (1934); Ingham (1950), cited in Edwards and Claxton (1964); Long, Sweet and Tukey (1956); and Westing (1969) a l l suggest that a light rain or spray can wet foliage sufficiently so that any additional moisture w i l l drip o f f . Thus, salts deposited on foliage beforehand may be picked up by the additional moisture and drip to the ground, thereby contributing these salts to the s o i l s . Long, Sweet and Tukey (1956), in particular, put forth the point that salts and other nutrients absorbed by the leaves, may be leached out of the leaves and then are dripped to the ground by the process described. Effects of Ocean Spray on Plants A number of studies have been carried out concerning the particu-lar injuries sustained by coastal vegetation; the morphological abnor-malities resulting from these injuries and the plant distribution and 39 successional patterns peculiar to coastal plants. It has now been generally accepted that the high levels of salt contributed by ocean spray, are wholly or partly responsible for.the peculiar types of injuries known predominantly to coastal plants, and for the character-i s t i c vegetation patterns of coastal plant communities. In general, salt spray, either from the ocean or a r t i f i c i a l l y applied through sprinkling or splashing, in some way affects the physio-logy and morphology of certain plants. Some of the documented injuries, growth abnormalities, and adaptations are as follows. Injury to leaves is manifested in a condition known variously as necrosis, leaf burn or brown leaf scorch. It has been observed numerous times on a wide range of plants but is perhaps most obvious on deciduous trees or broadleaf herbs. The inverted V-shaped pattern of scorch on broadleaves is quite characteristic (Boyce 1954). Karschon (1958) describes other patterns such as tip burn which is often combined with margin burn. Browning of the leaf i s believed to start along the margin and/or t i p , never in the interior, and can progress towards the midrib (Boyce 1954, Karschon 1958). The needles of conifers have been observed to be a reddish colour when affected with leaf burn (Oksbjerg 1952). It is universally observed that the highest degree of injury occurs on the windward side of foliage where i t is particularly exposed to the salt-bearing winds. Necrosis initiated by ocean spray has. been...observed by Wells and Shunk (1938), Moss (1940) and Wells (1939) on dune and strand trees and shrubs; by Edlin (1943) on such British coastal trees as spruce, Dine, 40 larch and elm; by Oksbjerg (1952) on coastal spruce forests in Jutland; b y Boyce (1954) on dune herbs, grasses and trees; and by Karschon (1958) on coastal Eucalyptus stands. Additional observations of necrosis on roadside trees caused by spray from salted highways have been made by Walton (1969), Westing (1969) and Hofstra and Hall (1971). That "salt" is the probable cause of injury has been repeatedly confirmed since Oosting and Billings (1942) observed the effects of spraying dune plants with sea water. Salt spray, and specifically the "salts" carried in the spray, is also believed responsible for the death of exposed leaves, twigs, shoots, buds, branches and occasionally an entire plant. Dead leaves, twigs and branches are most apparent on the windward side of trees and shrubs, on the lower branches and trunks (Edlin 1943) and above the average level of the canopy (Boyce 1954). Those individuals or plant organs which are more sheltered, for example, by other trees or branches such as those in the interior of stands or canopies suffer less injury (Wells and Shunk 1938, Moss 1940, Wells 1942, Edlin 1943, Oosting and Billings 1942, Boyce 1954).. The following example of these two major types of injury will serve to illustrate them. Wells and Shunk (1938) describe the injuries of affected trees and shrubs as "the killing of the upper younger tissues of the stem, the small immature leaves and the margins of the half developed leaves lower down, and necrosis along the tip and margins of said leaves." It is important to note here that i t is the young plant organs which are particularly susceptible. Trees in bud or with mature 41 foliage are not damaged to such a high extent (Edlin 1943). Moss (l940)> however, describes damage of trees which only became clear in the spring, several months after a "salt storm" of hurricane force struck the New England coast in the f a l l of 1938. Apparently, the over-wintering buds which were exposed to the spray were killed. Other buds less exposed or even completely sheltered opened later in the spring than usual. The trees experienced an overall lag of growth. A lag of growth was also noted between the windward and leeward sides of individual trees: the buds on the windward side opened up several weeks later than those on the leeward side. Oksbjerg (1952) and Karschon (1958) also mention that trees affected by salt spray have a reduced or decreased growth compared to those which are not exposed to salt spray, or are exposed to a lesser degree. In general, injuries on vegetation were observed to be greatest after severe storms, especially when accompanied by l i t t l e or no rain and under conditions of high evaporation (Edlin 1943; Harding, Miller and Fireman 1956). These conditions allow a build up of salts on the vege-tation. Storms which occurred in spring usually incurred the most damage but only i f the young shoots and leaves had broken bud. Westing (1969) studying the effects of highway applied salt on roadside vegetation, suggests the following order of effects on afflicted trees. First, the overall affect is a depression of growth expressed as fewer and/or smaller plant organs; second, leaf burn and defoliation; thirdly, shoot die-back which, i f sufficiently severe, may lead to death of the tree. A l l of these, of course, need not occur to one individual; i t is a generalized statement. 42 The injuries are explained as being a consequence of certain physiological changes in the plant organs. These physiological changes are explained in terms of either excessive water loss or "salt poisoning." Of the two, the theory that "necrosis or death is due to excessive water loss from the osmotic action of high salt concentration on the unprotected surface" (Wells and Shunk 1938, p. 489) is losing favour to the second theory. Oksbjerg (1952) would support the theory of Wells and Shunk and further explains that the "hygroscopic salt deposited results in a high osmotic pressure and water is actually drawn out of the leaves to the point of desiccation if unchecked." Others dispute or ignore this theory as a possible explanation for necrosis or death. Very l i t t l e seems to have been written on the desicca-tion theory regarding the plants which would be affected in this way compared to the immense amount of information available on salt toxicity and tolerance. Oksbjerg (1952) suggests that'the moisture requirements of a plant might be a factor in' the degree of injury sustained through desiccation. Physiological studies since the early 1950's have revealed that the salts are actually absorbed by the aerial parts of plants. It was suggested that the ions of the dissolved salts may, therefore, be toxic to some plants (Boyce 1951b and 1954, Westing 1969, Walton 1969). Certain plants, however, can adjust to the intake of salt by means of certain morphological features. Others are injured or die as a result of the salt intake (Curtis 1943 cited in Karschon 1958, Boyce 1951b and 1954). The mechanisms of uptake of salts and other elements from atmosphere of 43 high humidity or from applied sprays are discussed in articles by Breazeale, McGeorge and Breazeale (1950); Swanson and Whitney (1953); Fisher and Walker (1955); Oberlander (1956); Oland and Opland (1956); Van Overbeek (1956); Gustafson (1957); Koontz and Biddulph (1957); Proebsting (1957); Harding, Miller and Fireman (1956); Schropmeyer (1961); and Clayton (1972). In general, salts may enter leaves through either the epidermis, stomata, or traumata. It is quite reasonable to assume that salts would have to be in solution i f they are to enter the leaves by either of these three routes. The water necessary to produce this solution could come from ocean spray, rain or fog, direct condensation on the leaves, or from guttation. It is quite possible that a l l these sources of water could produce a hypertonic solution on the surface of leaves. Entrance of salts directly through the epidermal cells is probably important only for young leaves or mature leaves of species which have a very thin cuticle. Both Curtis (1943) as cited in Karschon (1958) and Harding, Miller and Fireman (1956) believe that the stomata provide the most important route for salt entry. However, experiments by Boyce (1954) revealed that entry of salt through the stomata and cuticle was too slow to account for the rapid injury of most plants when subjected to simulated ocean spray. Boyce (1954), supported by Oosting (1954), has noticed that small "traumata," which are cuts or tears in the cuticle and epidermal cells produced by the mechanical abrasion due to winds, are present on many of the branches and leaves of coastal plants. These are considered as 44 the primary entry points for salts. The wind-damaged leaves and branches were found to uptake a greater amount of salt; and were considered to be more susceptible to injury than others which were intact, because of that fact. The entry of salts may be affected by the external morphological features of the plant organs. The entry of ions may be reduced by such features as a hard or impermeable waxy cuticle, a very thick cuticle, resinous deposits on the surface of the organs (eg. on over-wintering buds), hairy stomatal openings, and rigidity of structure (Edlin 1943, Breazeale, McGeorge and Breazeale 1950, Boyce 1954, Oland and Opland 1956, and Wittwer and Teubner 1959). These features either physically block the passage of the ions or reduce the rate of intake of ions by providing mechanical resistance to wind which would produce many more entry points or traumata (Benecke 1930, cited in Boyce 1954). This may explain in part the varying degrees of injury among species. Further on the topic of tolerance or resistance to salt spray, i t has been observed by Boyce (1954) and Westing (1969) that woody species are more susceptible to injury by salt spray than are herbaceous species, particularly the grasses. Boyce attributes the particular resistance of grasses to the nature and the "wetting properties" of the leaves. Salt droplets deposited on the leaf surface may be unable to enter the leaf in sufficient quantities to „ cause... injury. The droplets tend to "bead" on the surface; this apparently has an adverse affect on entry of the water droplet. Morphological properties of the leaf itself may also result in the lack of movement of those salts which do enter the leaf. 45 Westing (1969) also includes species which are not indigenous to a natural saline environment as being especially susceptible to injury from salt spray. This is due to the low tolerance to internal accumula-tions of salts and lack of the special adaptive morphological features which would restrict entry of the ions. The two ions believed to be responsible for the injuries sustained by plants are sodium and chloride as they are present in the highest con-centration in the droplets of salt spray. That is to say, they make up the largest part of the total number of ions i n the spray droplet (Boyce 1954). Harding, Miller and Fireman (1956) have found no other ions present in the leaves of citrus i n significant quantities. They suggest the following reason for this. As the water droplets from sprays evaporate "the f i r s t salts to precipitate from solution of relatively low concen-trations are those such as calcium carbonate or calcium sulphate ... sodium chloride remains in solution u n t i l most of the water is absorbed." Hence, the salts are i n relatively high concentrations on the leaves and are absorbed. There seems to be considerable disagreement cn the topic of absorbed sodium and chlorine movement within the plant. Wittwer and Teubner (1959) believe that these ions are relatively immobile and> therefore accumulate in the leaves. Harding, Miller and Fireman (1956) found no evidence that the ions were translocated from leaves where absorption took place to leaves where no absorption occurred. Boyce (1954), on the other hand, did find considerable evidence that the ions were translocated once absorbed. His experiments have revealed that upon 46 entry the chloride ion is translocated to the apices and tips of leaves, shoots and twigs. This is;believed to explain the patterns of injury of leaves suffering from leaf scorch. Westing (1969) would agree that chloride is accumulated in high quantities and adds that the sodium is o translocated out of the absorbing organ and into other parts of the plant. Hence the concentration of sodium in these organs is lower than that of chloride. Karschon (1958), Boyce (1954), Westing (1969) and Walton (1969) have found that the sodium ion is generally present in lower concentra-tions than the chloride ion. This does not mean that a l l agree that chloride is the most injurious of the two. Researchers tend to disagree over which ion, sodium or chloride, is the most important in causing injury to coastal plants. Boyce (1951b and 1954) considers chloride to be responsible for both injury and salt hypertrophy. Westing (1969) would also consider chloride to be the more important ion. He states that only, "very sodium sensitive plants ... are a f f l i c t e d with sodium toxicity sensu s t r i c t o . The others are usually injured f i r s t by such indirect means as the deleterious effects of sodium on the root environment (referring to the high salinity of ditches adjacent to salted highways) ... or chlorine toxicity." Walton's experiments in 1969 also showed that there was a closer relationship between injury and the chloride ion concentration than between injury and the sodium ion concentration. On the other hand, La Casse and Riche-(1964) as cited in Walton (1969) have related sodium content and f o l i a r damage. 47 The fact that foliar absorption of ocean spray has deleterious effects on coastal vegetation has been verified by a number of studies as the previous discussion demonstrates. This does not eliminate the possibility that vegetation may benefit from foliar absorption of spray under certain conditions, although there does not appear to be any published research to support this idea. For example, certain ions necessary for the growth of plants could be taken into the plant by foliar absorption. This idea seems feasible since a number of essential nutrients are known to exist in ocean water. There is no reason to expect that these ions should not enter the plant by foliar absorption just as sodium and chlorine ions do-Foliar absorption of plant nutrients, both cations and anions, has been verified for a wide variety of agricultural species. Some studies on this subject include those by Swansoh and Whitney (1953), Fisher and Walker (1955), Oland and Opland (1956), Koontz and Biddulph (1957), Proebsting (1957), and Schopmeyer (1961). A review of the extensive literature available on this subject is beyond the scope of this thesis. For a detailed discussion of foliar absorption of plant nutrients the excellent review by Wittwer and Teubner (1959) is recommended. The complexity of the subject is reflected in one of their main conclusions which states that the mechanisms involved in foliar absorption such as active uptake, ion exchange, and diffusion may be essentially the same as for roots. The distinct asymmetrical or flag-form growth exhibited by coastal woody plants has been attributed to death or damage of plant tissue by ocean spray (Wells and Shunk 1937, Moss 1940, Oosting and Billings 1942, Wells 1942, Oosting 1945, Boyce 1954, and Karschon 1958). These trees or shrubs usually have more and healthier foliage on the leeward, or landward, side; branches on the windward side may be dead, either bare or broken; and the asymmetry may be apparent over the length of the individual or may be restricted to the upper or lower part, depending on the relative exposure. Boyce (1954) noted the stems of several coastal woody plants were curved inland. These stems were found to be "an amalgam of inter-nodes produced by meristems of many different lateral branches." He suggested that this growth form was the result of the following series of events. "The terminal and seaward meristems are killed by salt spray while the leeward meristems continue to function." Leeward meristems which take over the function of the dead terminal grow vertically until they too are killed. The leeward meristem which was killed, is in turn replaced as leader by another leeward meristem and so on. The result of this differential growth is a "composite stem curved in a leeward direction." Boyce (1954) ascribed the peculiar shape of certain canopies to the action of salt spray on exposed branches, shoots and leaves. The canopies of some trees and shrubs tend to assume an escalier appearance with the highest point being the furthest inland. The development of this shape is believed to occur in the following way. The exposed terminal shoot is killed, owing to a high deposition of salt particles from the spray onto the shoot. Lateral meristems take over the aqtion 49 of the dead terminal and produce many shoots. Any shoot which projects above the average level of the canopy will be killed by the high level of salts which they will collect. Dead branches will eventually be "pruned" by the wind, thus giving the canopy a rather uniform appearance. The seaward branches will grow to a certain height and will offer pro-o tection to the more leeward branches behind. In turn these leeward branches will be able to grow up a l i t t l e higher because of the increased protection and so will protect even more leeward shoots. The effect of this protection is cumulative in a direction away from the sea. Only the protected lateral and leeward terminals develop, resulting in the characteristic "compact, repressed, sloping form" (Wells and Shunk 193 7). The angle which the canopy will develop is dependent upon the intensity of the spray as the angle is such that salts will be deposited uniformly upon the canopy (Boyce 1954). Woody plants which display one or more of these form modifications are referred to as "spray forms" by Wells and Shunk (193 7), since i t is obvious to them that the form is a consequence of the injuries sustained as a result of salts deposited on vegetation from ocean spray. They also consider the degree of form modification to be strictly correlated with the extent of injury. Effects of Ocean Spray on Soils Up to this point the discussion has been limited to ocean spray as i t effects the above ground parts of coastal vegetation. In general, these injuries and adaptations are believed to be primarily a result of the aerial deposition of salts on vegetation rather than a result of 50 salts in the soil. A discussion of the effects of ocean spray on salts in the soil may help to clarify the above statement. The concentration of salts from ocean spray in the soil is not simply a function of the amount or rate at which salts are added to the soil. Since most of these salts are highly soluble, they may be quite easily leached from the soil. Therefore, both precipitation and soil moisture will have an important effect in determining the soluble salt concentrations in any soil. Oosting (1954) has suggested that the salts, which are deposited, are very rapidly leached, since he noted the pH of the soils to be greatly reduced after a rain. He also mentioned that salts, contributed to soils with a high moisture coiitent, will not be as concentrated as in those which have a lower moisture content. Even though the same amount of salts may be deposited on two different soils, the concentration of salts in both may be quite different. There seems to be very l i t t l e support for the supposition that ocean-spray-deposited-salts in soils could have extreme effects, or that salts are deposited and retained in the soils to a high degree. Since low concentrations of chlorides were found in the root zone of dunes by Boyce (1954); since most salts deposited on soils by ocean spray are rapidly leached (Boyce 1954, Oosting 1945 and 1.954); since Oosting's extensive experiments in 1954 revealed no relation which would explain the distribution of certain species on the basis of soil salinity data; and finally, since injuries of a comparable nature and. extent were found on vegetation growing on both saline and non-saline soils (Harding, Miller and Fireman 1956; Karschon 1958), i t was concluded that salts 51 from ocean spray do not significantly affect soils and hence, the salt deposition on foliage must be responsible for much of the observed damage. This is not to deny that damage to plants may be caused by saline s o i l solutions. On the contrary, a great deal of work confirming the damaging capacities of saline s o i l solutions has been done by Geraldson (1954, 1957), Monk and Wiebe (1961), La Haye and Epstein (1969), Walton (1969) and Westing (1969). Although very l i t t l e research has been done on the subject, i t is quite possible that the addition of certain ions to the soil by ocean spray may influence the level at which these ions are maintained in the so i l and thereby have a beneficial effect on certain plant species. This possibility seems logical for the following reasons: (1) ocean water contains a number of ions such as magnesium, calcium and potassium which are essential for the growth of a l l plants (Harvey 1960); and (2) the ab i l i t y of the s o i l colloids to hold the above named cations is greater than their a b i l i t y to hold either sodium or chlorine ions (Millar, Turk and Foth 1965). This statement goes along with the thinking of Wilson (1959) and Ingham (1950). Both of these authors feel that ocean spray, deposited either directly or indirectly from tree drip, can add valuable plant nutrients such as potassium and nitrogen to the s o i l , thus enrich-ing i t and contributing to plant growth. The addition of sodium, potassium, magnesium and calcium to the s o i l from ocean spray has been documented by 22 Clayton (1972). By using Na he also demonstrated that cations absorbed by the foliage can be transported through the plant and released in the s o i l . Etherington (1967) studied potassium cycling on sandy soils with low cation exchange capacities and concluded that potassium was not de-ficie n t because of the income from salt spray. 52 Ocean Spray and the Distribution of Vegetation One of the f i r s t to recognize that ocean spray might be an important factor in the distribution of coastal vegetation was Bowman (1918). He suggested that zonation might be related to the prevailing wind which "carries dense salt spray or mist inshore... and certain strand plants are better able to withstand this drenching." For the next 20 years most of the thinking went back to the drying effects of wind and environmental factors associated with the sand so i l as explanations for the distribution of coastal communities as well as individual species. Wells and Shunk (1937) contradicted these ideas by pointing out that coastal growth forms commonly called "wind forms" are more widely the product of salt spray. Their work along the North Carolina coast (Wells and Shunk 1938) was the f i r s t detailed study correlating the distribution of coastal vegetation and ocean spray. They recognized three groups of plant species, each occupying a particular part of the coastal strand, apparently in response to different intensities of ocean spray. Water bush (Baccharis halimifolia), yaupon (Ilex  vomitoria), and wax myrtle (Myrica cerifera) are the dominant species in the high spray zone nearest the ocean. Occurring to the landward side of these communities, and receiving some protection by them, are stands of live oak (Quercus virginiana). A third type characterized by l o b l o l l y pine (Pinus taeda), the most sensitive of the arborescent species surviving in the spray zone, dominates the inner and weaker spray zone. Observa-tions on a few representatives of more inland woody species which by chance were close enough to the sea to be injured, showed such a severity 53 of injury that the explanation of their almost complete absence from the spray zone was apparent. On the other hand, the complete absence of characteristic dune species on dunes located very near the coast was shown to be related, in certain circumstances, to extreme intolerance to ocean spray. Wells and Shunk (1938) considered that the loblolly pine was the most sensitive of the woody plants able to survive in the weak or landward side of the spray zone and, therefore, could be used as an indicator species in determining the width of the spray zone at various points along the coast. Wells (1939) used the term "salt spray climax" to describe forests of live oak on Smith Island, North Carolina. He reasoned that the normal climax species of the region such as white, black and southern red oaks were intolerant to ocean spray, leaving coastal habitats to the spray-tolerant live oak. From studies of the live oak forest, he concluded that these stands were at the "climax" stage of succession and that the soils were mature and "mesic." He f e l t that the regional climax species would dominate the coastal forest rather than live oak i f i t were not for the ocean spray The hypothesis put forth by Wells and Shunk "that the distinctive composition of the (coastal) dune community is based primarily upon the species adaptation to the spray factor" was not based on detailed measurements of the coastal environment. This lack of data, particularly measurements on ocean spray, tends to weaken their argument somewhat. Setting out to test this hypothesis, Oosting and Billings (1942) 54 established a series of transects on Bogue Bank, North Carolina from the beach to the top of the rear dune and made measurements of a number of "pertinent" vegetation and environmental factors along these transects. No significant correlations could be found between the vegetation zona-tion indicated in the transects and the environmental factors measured, with the exception of ocean spray. Simple "salts traps" consisting of cheesecloth stretched on a frame, and set up at right angles to the prevailing winds were located along the transects. Regardless of weather, the highest ocean spray values were on the windward side of the foredune; the crest of the foredune showed the next highest value, and the crest of the rear dune was next in order, with the depression between receiving much less. The distribution of the two major dune species, sea oats (Uniola paniculata) and wire grass (Andropogor. scoparius var l i t t o r a l i s ) , showed a close zonal relationship to the deposition of spray: sea oats is dominant where ocean spray is high, wiregrass where spray is at a minimum. Phytometer experiments in which plants were sprayed with sea water corroborated the fi e l d observations. Boyce (1954), using methods similar to those of Oosting and B i l l i n g s , studied other North Carolina dune areas and obtained similar results. Following up his earlier work, Oosting (1945) studied the tolerance of a number of species common to the North Carolina strand by spraying with seawater. Species tolerance ranged from death after a single spray-ing to completely unaffected by spraying four times daily for nine days. Oosting concluded that "between these extremes were a variety of responses which lead to the conclusion that most of the strand species are 55 relatively tolerant to spray, but when storms are severe and of long duration, injuries or death will occur among the least tolerant". This study also confirmed some of the earlier work of Wells and Shunk (1938); live oak, the dominant tree in the maritime forest, was resistant to spray while loblolly pine and turkey oak (Quercus laevis), common inland dune species, were intolerant. Consequently their occurrence marks the inland limit of effective spray and the limit of the truly maritime forest. Martin (1959), working on the New Jersey coast, sprayed a number of coastal species with ocean water and was able to classify these plants into five groups according to the degree of resistance exhibited. The classes ranged from complete resistance (no visible effect) to complete lack of resistance (plants killed). He found that herbacous plants either showed almost complete resistance or no resistance while woody plants were spread throughout the five classes. In relating these classes to coastal vegetation zonation, he found that in plant communities which occur where the intensity of ocean spray is extreme, species which exhibited absolute tolerance to ocean spray were definitely favored. However, some species with low tolerance but a low growth-form adapted to escaping high salt spray intensity can also survive. Some species with a high resistance rating were not found here, apparently because they required higher levels of soil moisture. In the zone of high to moderate intensity, the effect of ocean spray is to favor low growth-forms and to reduce the height of less resistant species by surpressing terminal growth. Therefore, the canopy formed by 56 the tallest species exhibit a molded or hedge-like growth form. In the zone of low spray the vegetation is dominated by species which have a low tolerance for ocean spray. Most species show l i t t l e or no modification of growth .form by ocean spray. Spray-molded forms are exhibited by only a few species which are extremely intolerant and which occupy a position in the upper canopy. It seems as i f almost a l l of the research on the effect of ocean spray on the distribution of vegetation has taken place on the Atlantic coast of North America. A literature review failed to locate any studies from the Pacific coast of the continent that attempt to relate ocean spray measurements to the pattern of coastal vegetation. This is not due to a lack of coastal vegetation studies per se: a considerable number of studies have been carried out including those by Purer (1936), Cooper (1936, 1967), Dasmann (1959), Johnson (1962), and Clayton (1972) in California; Egler (1934), Byrd (1950), Cooper (1958), Hanneson (1962), Green (1965), Wiedemann (1966), Davidson (1967), and Kumler (1963, 1969) in Oregon; Jones (1936), Lotspeich et a l . (1961), Cooper (1958) in Washington; and Day (1957) and Kuramoto (1965) in British Columbia. The lack of interest in ocean spray is rather puzzling since many of the papers recognize the importance of this factor, particularly those which were published prior to most of the east coast papers on the subject. The only ocean spray data located are in papers by Byrd (1950) and Clayton (1972). Byrd measured spray by the cheesecloth method along a transect that extended over several dune ridges. However, he did not correlate these findings with the vegetation pattern. Clayton measured 57 the concentration of a number of ions in rainwater at two sites along the California coast as part of a mineral cycling study. Although the species composition was not markedly different at the two sites, he did note that the growth-foorm of the dominant species, Bacchar is pilularis, was different. 58 CHAPTER V THE COASTAL SITKA SPRUCE FOREST The coastal study area included most of the open coast between Clayoquot and Barclay sounds. More specifically, i t included that section of coastline starting at Emerald Beach, on the Esowista Peninsula, one mile south of the village of Tofino, and running southeastward to the headlands south of the Ucluelet townsite (Fig. 5). This stretch of coastline was a logical choice for the study area being the only part of the west coast which was both exposed to the open ocean and easily accessible by roads. The section includes a variety of coastal landforms as well as a number of different exposures to the ocean and appeared to be quite representative of the west coast as a whole. Strictly speaking, the study area encompassed only the narrow strip along the coast occupied by the coastal Sitka spruce forest. However, since one of the objectives of the thesis was to attempt an explanation for the existence of the spruce forest in a region in which the normal climax forest (zonal or climatic climax) is dominated by western hemlock, i t would not have been logical to completely ignore the hemlock forest. Instead, i t seemed reasonable to concentrate on the coastal spruce forest and, at the same time, to take a more general look at the area immediately inland from the spruce forest where western hemlock as well as western redcedar are well represented. In other words, the hemlock or hemlock-redcedar forest was, in a general sense, Figure 5. 60 considered to be a base mark for which to compare the environment of the coastal spruce forest. PHYSIOGRAPHY AND PLEISTOCENE EVENTS The oldest land surfaces within the study area are made up of andesite and dacite volcanics of the Vancouver group which are Triassic in age (Dolmage 1920), and, as far as is known, are the only rocks which outcrop here. They are exposed along almost the entire length of the headlands as well as on a l l of the small offshore islands. The largest continuous exposures are between Point Cox and Portland Point and between Wickaninnish and Florencia bays. Several cone-shaped hills of volcanic rock east of Point Cox rise to slightly over 400 feet and are the highest points in the study area. The volcanic rocks originated from a series of lava flows laid down one on top of the other beneath the ocean. The resulting surface is very irregular and forms a very jagged shoreline. These irregular-ities are intensified in the area immediately adjacent to the shore to the extent that foot travel was very difficult in many places. Multidirectional jointing in the rocks and wave action are largely responsible for this. The rocks are very resistant to weathering except along the joints; therefore, erosion has resulted in increased surface roughness. The land above the zone of storm wave activity has a surface organic layer in direct contact, or nearly so, with the volcanic rock. Mineral soil is almost completely lacking except for shallow deposits in depressions. Wave action in the past would have probably removed any 61 material of this kind. It is quite likely that sea level was at one time considerably higher than i t is now and that the land has undergone gradual uplift since the last major glaciation. Such changes have been documented for a number of locations along the coast of British Columbia and Alaska; the uplift in most cases being attributed to isostatic rebound (Kerr 1936, Twenhofel 1952, Heusser 1952, 1960). This premise is further supported by the occurrence of a number of raised wave-cut benches on Echachis, Frank, and Box islands that are ca. 10 to 20 feet above high tide. Aside from the Jurassic volcanics, the remainder of the land surface in the study area consists of Pleistocene or Recent deposits. A l l of Vancouver Island, with the exception of some of the highest mountain peaks, was almost certainly covered with a thick ice sheet around 15,000 years B.P. which is considered to be the time of maximum ice advance during the Vashon Stade of the Fraser glaciation (Mullineaux et al. 1965). Ice retreat began about that time, marking the beginning of the Everson Interstade. Melting of large masses of ice resulted in a considerable rise in sea level (Nasmith 1970). Radiocarbon dates from the southern end of Vancouver Island indicate that the latter was free of ice by 12,800 B.P. (Dyck et al. 1966). At that time, sea level on the Saanich Peninsula was 275 feet higher than i t is today (Halstead 1968). Valley glaciers remained for some time after 12,800 B.P., and probably retreated slowly. This was followed by a period in which some valley glaciers apparently advanced, or at least maintained their positions (Halstead 1968). The period of advance was short-lived, and was followed by a rapid retreat and disappearance of most of the 62 glaciers by 9,000 B.P. (Nasmith 1970). Retreat was again accompanied by high sea levels followed by a period during which relative sea level gradually dropped due to rebounding of the land after the load of ice was removed (Nasmith 1970). Surface deposits formed during the Pleistocene can be classified into four groups on the basis of their composition and probably mode of origin. In order of youngest to oldest these are: 1. Sandy beach plains 2. Sand and gravel glacial marine outwash 3. Marine clays 4. Boulder clay t i l l Boulder clay t i l l has the most limited distribution of any of the four groups. Small pockets of t i l l occur around the flanks of volcanic h i l l s and erratics can be found on their summits. Such deposits appear to be restricted to hi l l s which are a minimum of several hundred feet high. T i l l at lower elevations may have been removed by the sea during periods of higher relative sea level. The remaining three groups of glacial deposits occur in a continuous sequence in some locations and, together, make up an extensive plain which rises to slightly over 100 feet above sea level. Here, the plain, part of the Estevan Coastal Plain, extends from the northwest end of Schooner Cove to the Ucluelet townsite and inland to Kennedy Lake, a distance.of six to ten miles. Dolmage (1920) gave the name Wreck Bay Formation to these deposits and considered them to be of marine origin and Pleistocene in age. His observations of this formation were probably restricted to the eroding c l i f f s along Florencia Bay. Wade (1965), also working with the Florencia Bay exposure, recognized two sequences: a lower section of glacial marine t i l l composed of blue-grey clay; and an upper section of glacial outwash or stream deposits. The investigator's observations of this exposure, as well as a number of others, suggest the following interpretation. The lowest exposed sequence of the Wreck Bay Formation consists of thin horizontal beds of a s t i f f grey clay. Along Florencia Bay its upper surface varies from just above sea level to a height of about 30 feet. Near Green Point in Wickanninish Bay i t reaches to the surface of the plain, an elevation' of almost 100 feet, and extends down to sea level or lower. Grey clay has also been observed at the bottom of a gravel pit near Kennedy Lake. The presence of shell fragments along with the fine bedding suggests a marine origin, which occurred possibly during ice recession when relative sea level was much higher. The presence of cobbles in the clay indicates that there might have been ice-rafted t i l l in the sea during the time of formation of these beds. Along Florencia Bay and extending inland to Kennedy Lake the clay beds are overlain by unconsolidated sand, gravel, and cobbles. These well-sorted horizontal beds are similar to normal glacial outwash deposits except for the almost complete lack of cross-bedding. The extremely regular horizontal bedding can best be explained as out-wash material which has been rexrorked and deposited in a shallow standing water body. Such conditions may have occurred during ice retreat, when relative sea level was much higher than i t is today. During this time, meltwater from the Alberni Canal-Barkley Sound area 64 would have been moving large quantities of course material down valley. A large stagnant ice block probably existed in the Kennedy Lake basin at this time and kept the basin from being f i l l e d by outwash. The uppermost deposits of the Wreck Bay Formation consist of horizontal beds of sand that almost everywhere form the surface of the terrace plain. This material was li k e l y deposited in a shallow sea during the latter stages of glacial retreat when the flow of meltwater was reduced to a point where large amounts of gravel and cobbles could no longer be moved. After the terrace plain had risen above the sea there was apparently some movement of sand by wind. The surface of the terrace plain i s flat-to-gently-rolling and has been dissected by streams to a very limited extent. Surface drainage is very poor except in the immediate vi c i n i t y of small streams and along the l i p of the terrace where i t drops off to the beach. This problem is further intensified by the presence of an iron hardpan which greatly hinders the downward movement of water. As a result, soils are waterlogged for much of the time and standing water is prevalent during the winter and spring. The only portion of the terrace plain of direct concern to this study was that lying within the coastal spruce forest: a strip 600 feet or less in width along the coast from the west end of Schooner Cove to the southeast end of Florencia Bay. Stands of Sitka spruce cover the ocean-facing scarp of the terrace from the beach up to the plain and inland from this point for a distance of no more than 200 feet. Along the south-facing part of Florencia Bay the terrace has been eroded by winter storm waves resulting in a slumping c l i f f with l i t t l e or no 65 vegetation cover. Here the spruce forest is limited to the plain on top of the c l i f f , the only location observed where this type of erosion was s t i l l active. However, tree ring dating indicated that wave erosion was active along parts of Wickanninish Bay and Schooner Cove as recently as 500 years ago. Subsequent to wave erosion, the cliffs had slumped resulting in stable slopes which have been covered with stands of Sitka spruce for 300 to 450 years or more. In the vicinity of Green Point, the whole slope is made up of clay with a small proportion of sand, and the texture remains constant along the slope for a distance of several miles to the northwest. From this point on i t contains an increasingly greater proportion of sand, and is predominantly composed of sand in Schooner Cove. Southeast of Green Point the sand component also increases. For the last 500 years or more most of the geomorphic activity has been directed towards the establishment of wide beaches that are present in almost every bay in the study area. In most bays the upper beach has aggraded to a height sufficient to insure protection from the winter storm waves allowing Sitka spruce to invade and form dense stands over the extensive beach plains. The beach plains have an undulating surface consisting of gentle ridges and depressions orientated parallel to the shoreline. Local relief is generally no more than about five feet. In some places the beach plain is capped by a series of dune ridges that form long crescents conforming to the shape of the bay. Dune ridges are best developed in bays receiving the maximum force of the summer onshore winds such as Cox Bay and the southeast end of Wickanninish Bay. Crests 66 of beach ridges may be uniform in height or very irregular and cut by a series of blowouts. They vary in height from several feet to a maximum of about 40 feet with the taller ones having the most irregular crests. The beach plains and beach plain-dune ridge deposits have undoubtedly resulted from progradation of the coastline in these bays. Along Wickanninish Bay and Schooner Cove, their development must have coincided with a period when the amount of sand deposited on the beach during the summer was greater than the amount removed during winter storms. This was in marked contrast to the preceding period when almost the entire length of the terrace had been undermined by winter storms as is presently taking place in Florencia Bay. Thus far, the discussion of beach plain development has been limited to areas where a continuous shoreline existed prior to progradation. A very different pattern existed in that part of the study area lying between Cox Bay and the tip of the Tofino Peninsula. At some time in the past the volcanic headlands along this section of the coast must have comprised a chain of islands. These islands were gradually connected by sand bars and converging spits, resulting in the formation of the present-day crescent-shaped bays with their accompanying beaches. Once the connections between the headlands were complete, the beaches began to advance in the direction of the sea, a process active at the present time. Once the land connections had been completed between the headlands and islands, conditions on the inlet side of the peninsula were greatly changed due to protection from the open ocean. Tidal currents on the inlet side of the peninsula deposited large amounts of 6 7 s i l t and clay forming extensive t i d a l flats in the area extending from Emerald Beach to within about one and one-half miles of the Tofino airport. Subsequent u p l i f t has raised these flats to ca. 10-15 feet above high tide. The present coastline i s one of relative s t a b i l i t y when compared with some periods of the past. The very erosion-resistant volcanic headlands and islands have provided most of the resistance to change. The only location where the land is receding at a noticeable rate i s along the south-facing section of Florencia Bay. One of the few areas of active beach plain development is located southeast of Green Point near the middle of Wickanninish Bay. Here, wind-blown sand as well as sand moving by longshore d r i f t i s trapped in the lee of the point. The surface so formed is well stabilized and covered with young spruce. An active dune area at the southeast end of Wickanninish Bay has been studied by Kuramoto (1965). Both active and stabilized dunes are present while the shoreline in front of them appears to be stationary. Along most of the bays a wall of beach logs along the winter storm line has been responsible for small to moderate amounts of progradation. The logs have formed barriers to erosion by winter storms and have also trapped the wind-blown sand. After a number of years, the sand has built up between the logs and has become stabilized by vegeta-tion (Fig. 6). The protection afforded the seedlings of Sitka spruce and species of herbs and shrubs appears to be very important for their survival. The large accumulation of logs and the resultant effect on progradation has likely been important only since the beginning of extensive logging operation on the west coast. Beach logs were not 63 F i g u r e 6. W i n t e r s t o r m l i n e s h o w i n g t h e a c c u -m u l a t i o n o f b e a c h l o g s a n d w i n d -b l o w n s a n d . F i g u r e 7. V i e w o f a f o r e s t t r a n s e c t a l o n g W i c k -a n i n n i s h Bay e x t e n d i n g f r o m t h e c o a s t -l i n e ( l e f t o f p h o t o ) , i n l a n d . F r o m l e f t t o r i g h t , t h e c a n o p y c o m p o s i t i o n i s S i t k a s p r u c e , S i t k a s p r u c e - w e s t e r n h e m l o c k , w e s t e r n h e m l o c k - w e s t e r n r e d c e d a r ( t o t h e r i g h t o f b r e a k i n c a n o p y ) , a n d w e s t e r n r e d c e d a r - s h o r e p i n e ( m u s k e g f o r e s t ) . 69 nearly so abundant before this time. For additional information on the surficial deposits of the Estevan Coastal Plain, one should consult Nelson and Cordes (1972). A comparison of the surficial deposits of the coastal spruce forest with those of the interior hemlock-redcedar forest indicates that, although the percentage area covered by the different deposits varies between the two, both types of forest contain a complete range of the deposits found in the area as a whole. During the last 500 years or more, geomorphic processes have been more active in the coastal spruce forest resulting in both active deposition and erosion. Development of beach plains and dune complexes as well as erosion and slumping of sea clif f s has provided new land for colonization by Sitka spruce. Nevertheless, extensive areas of much older surfaces such as rock outcrops, boulder clay t i l l s , marine clays, and ancient beaches and outwash plains occur within the coastal spruce forest. On the other hand, not a l l the surfaces in the interior hemlock-redcedar forest are old: sand plains and raised tidal flats probably of recent origin are common, particularly in the northern parts of the study area. In conclusion, a study of the physiography of the area failed to turn up any strong differences between the coastal spruce forest and the interior hemlock-redcedar forest in either geomorphic history or the range of surficial deposits present. Therefore, i t does not appear as if the physiographic differences can provide an explanation for the existence of the coastal Sitka spruce forest. 70 VEGETATION As previously stated, the coastal Sitka spruce forest occupies a narrow but continuous strip along the coast which is a maximum of 600 feet wide, and in most places considerably less (Fig. 7). The area inland from the spruce forest is dominated by western hemlock, western redcedar, and shore pine (Fig. 8). In general, the range of forest communities found in this "interior" forest were similar to those reported for the Coastal Western Hemlock Zone in British Columbia by Krajina (1959, 1965), Orloci (1961, 1964), Lesko (1961), and Eis (1962); and in Washington and Oregon by Fonda and Bliss (1969) and Franklin and Dyrness (1969). The interior forest can be broken into two major types on the basis of landforms and surfical deposits: the rock outcrop land type and the coastal plain land type (Cordes and MacKenzie 1972). The rock outcrop land type consists of bedrock knolls as well as inland extensions of the rocky headlands. Western hemlock, western redcedar, and shore pine are a l l present throughout this land type, however, the proportion of these three species varies substantially from place to place. In the coastal plain land type surface drainage, which is controlled by both topography and parent material, is the single most important factor in determining the pattern of vegetation types. The vegetation types, listed in order of good to poor drainage, are: the high forest (hemlock - redeedar), the redcedar - hemlock forest, the mus forest (redcedar), the (shore) pine bog forest, and the treeless bog. These types have been described by Cordes and MacKenzie (Appendix III). V E G E T A T I O N PACIFIC RIM N A T I O N A L PARK Coastal Zone Rocky headlands and islands subzone A l rocky outcrop A2 shrub and herb A 3 spruce forest red cedar-hemlock-spruce transitional forest Beaches, dunes, and near coast terraces and plains subzone | j Bl unvegetated B2 herb and log B3 shrub and herb B4 scrub spruce forest B6 spruce forest B7 hemlock- red cedar- spruce transitional forest Coastal plain subzone =^= C2 cedar - hemlock forest lllllllll C3 muskeg forest C4 pine bog forest C5 bog s Rocky outcrop subzone Ijjjjljjj DI rocky scrub forest Tidal flats subzone El unvegetated | j E2 vegetated Figure 8. Source: Cordes and MacKenzie, 1972. (See Appendix III for descriptions of the classification units). 72 Although a sharp break exists between the coastal spruce forest and the interior forest in some places; in other locations there is a transitional belt where both Sitka spruce, western hemlock, and some-times western redcedar grow together, a l l as dominant or co-dominant trees (Fig. 7). The transitional forest has important ecological implications since i t represents, in a general sense, both the inland boundary of the coastal spruce forest and the coastal boundary of the interior western hemlock - western redcedar forest. For these reasons, the transitional forest was given forest type status (Tsuga heterophylla - Picea sitchensis forest type) in the classification system and was studied in the same detail as the Sitka spruce forest types. The vegetation of the coastal spruce forest will not be discussed here since a detailed description is provided later in the thesis. SOILS Soils of the coastal spruce forest belong to the Podzolic, Regosolic, Gleysolic, and Organic Orders while podzolization, gleization, and mor formation are the dominant pedogenic processes. A majority of the soils found on well-drained sites have a thick mor humus layer, a well-defined eluviated A horizon and a thick podzolic B horizon, and belong to the Podzolic Order. Normally there is a very sharp break between the humus layer and the Ae horizon except in seepage sites where a melanized Ah horizon is present on the surface or beneath a layer of raw humus. All three great groups of the Podzolic Order are well represented and, as a group, are far more common than any other order. Soils belonging to the Regosolic Order occur on rocky sites and 73 sand plains. Lithic Regosols consist of l i t t l e more than a thick layer of raw humus over bedrock. Even though these soils are classified as regosols, they are mature profiles and show no signs of further pedogenic development. Orthic Regosols, consisting of mor humus over slightly altered sand, are common on beach plains of fairly recent origin. Given time, these soils develop into podzols. Soil belonging to the Gleysolic Order form on poor drainage sites such as flat areas and depressions. Gleysols also form on better drained sites, particularly those with coarse-textured parent material, as a result of iron pan formation leading to a deterioration of internal drainage. Organic soils are of limited occurrence in the area being restricted to very poorly drained sites where the water table remains close to the surface throughout the year. Sites of early Indian habitation with extensive midden deposits often have chernozemic-like rendzina soils. These sites are gradually invaded by Sitka spruce resulting in a trend towards podzolization and mor humus formation. The range of soil types in the interior hemlock - redcedar forest is similar to that of the coastal spruce forest (Cordes and MacKenzie 1972). It appears as i f the process of gleization is more prominent in the interior forest, judging by the more frequent occurrence of gleysols and gleyed podzols. This is undoubtedly related to the presence of extensive flat areas with poor surface drainage. 74 CLIMATE An examination of the climatic data from several west coast stations was undertaken with two main objectives: (1) to provide a basic description of the climate for the study area; (2) to compare the climate of the coastal spruce forest with that of the western hemlock - western redcedar forest immediately inland from the spruce forest. Since climatic measurements were not recorded as part of this investigation, i t was necessary to use established Canada Department of Transport weather station records. Weather records from the Tofino airport, located approximately 2,500 feet inland from the coastline, were used to represent the climate of the interior hemlock - redcedar forest. The two weather stations closest to the study area and within, the coastal spruce forest were located at Clayoquot, two miles northwest of the Tofino townsite, and at Amphitrite Point, several miles south of the study area. One of these stations would have been used; however, suitable data on wind, humidity, and visibility were not available. The closest station that was both within the spruce forest and for which suitable data were available was Estevan Point. This station was located on an exposed headland at the southern end of Hesquiat- Peninsula, approximately 30 miles northwest of the study area. It was considered to be typical of the more exposed portions of the coastal spruce forest. The west coast climate is typified by a lack of temperature extremes and an abundance of precipitation. The difference between the mean daily temperatures for the warmest and coldest months is only 17°F 75 at Estevan Point (Fig. 9, Table 1). As one would expect, the difference in temperature between the two stations is very small: the mean daily temperatures for the warmest and coldest months are 57.3 and 40.4°F at Estevan Point and 58.4° and 40.4°F at the Tofino airport. At Estevan point, the mean daily maximum temperature for the warmest month is 62.7°F while the mean daily minimum for the coolest month is 35.6°F. The mean daily temperature range for the entire year at this station i s 10.6°F. The area averages 36 days of freezing temperatures on record for the area are 7 and 91°F. Precipitation is very high with an annual mean of 115.4 inches at Estevan Point and 125.8 inches at the Tofino airport (Fig. 9, Table 1). The major portion of this occurs during late f a l l , winter, and early spring. Although there is a sharp drop in precipitation during the summer, no month averages less than 3 inches at either station. Nevertheless, lengthy periods with l i t t l e or no precipitation do occur; for example, in 1967 the Tofino airport recorded only 0.4 inches i n June and 0.6 inches in August. Snowfall is very limited with both stations receiving less than 11 inches of snow as an annual average. Due to the high precipitation, moderate temperature, and close proximity to the ocean the humidity remains high throughout the year (Fig. 10, Table 1). In fact, the stations are among the highest in North America for this factor. The relative humidity seldom f a l l s below 70 per cent and is consistently between 75 and 95 per cent. The high humidity along with cool temperatures and lengthy period of cloudiness keep the evaporating power of the atmosphere at a low l e v e l . In comparing the two stations, the Tofino airport has slightly higher values 76 Table 1. Long-term climatic records for two stations on the west coast of Vancouver Island Data are from ten-year records or longer. Mean Daily Temperature (°F) Estevan Point Tofino Airport Precipitation (in.) Estevan Point Tofino Airport Oct. 50.5 51.1 12.9 14.0 Nov. Dec. 45.3 44.0 16.0 16.6 42.4 41.6 17.1 16.6 Monthly Means Jan. Feb. Mar. 40.4 40.8 15.0 17.9 41.1 40.4 12.0 14.8 42.3 41.7 11.7 13.8 Apr. May June July Aug. Sept. Annual Mean 45.9 45.4 8.5 11.2. 50.3 50.8 5.1 4.1 54.0 54.7 4.0 3.7 56.6 57.9 3.6 3.7 57.3 58.4 3.1 3.5 55.2 55.8 6.4 5.8 48.4 48.5 115.4 125.8 Relative Humidity (Z) Estevan Point 4 A.M. 10 A.M. 4 P.M. 10 P.M. Tofino Airport Visibility (mi.) Estevan Point Tofino Airport 4 A.M. 10 A.M. 4 P.M. 10 P.M. 0~h h-5 0-5 0-h h-5 0-5 Wind Speed (m.p.h.) Estevan Point Tofino Airport Maximum Hourly Wind Speed (m.p.h.) Estevan Point Tofino Airport 91 89 90 89 88 89 87 89 91 93 94 93 90 87 89 88 88 86 82 78 78 83 84 87 86 85 84 83 86 84 80 76 76 76 80 81 84 84 81 91 88 89 88 86 86 84 85 89 81 84 91 87 96 94 94 93 94 93 93 94 95 96 97 97 95 90 91 92 91 88 84 81 81 83 85 88 86 87 84 85 89 86 81 78 75 74 74 74 80 79 80 95 93 92 91 91 90 90 90 91 91 94 94 92 :.7 0.6 0.9 0.7 0.3 0.4 0.6 0.6 1.1 1.1 3.6 2.2 14.8 i.4 3.2 5.3 6.8 5.7 4.9 2.7 2.1 3.7 3.7 4.2 4.2 51.9 1.1 3.8 6.2 7.5 6.0 5.3 3.3 2.7 4.8 4.8 7.8 6.4 66.7 :.3 1.1 1.7 1.4 0.8 0.7 0.7 0.4 0.4 1.9 2.9 2.6 16.9 .2 5.3 8.0 7.8 7.6 6.7 3.4 3.9 4.9 5.7 7.1 8.1 75.7 '.5 6.4 9.7 9.2 8.4 7.4 4.1 4.3 5.3 7.6 1.0.0 10.7 92.6 1.5 11.2 11.8 11.4 12.0 11.5 12.2 11.0 11.1 10.1 8.6 9.1 10.9 -.5 7.9 8.6 8.6 8.7 8.8 8.7 8.6 8.4 7.3 6.6 6.5 8.1 37 40 43 42 39 41 36 34 32 28 26 34 36 37 36 38 37 36 36 31 31 30 27 26 33 33 Source of data: Canada Department of Transport, Meteorological Branch (publications as listed in the bibliography, as well as data abstracted from the stations' records by the author). 77 O N D J F M A M J J A S MONTHS Figure 9. Mean daily temperature and mean precipitation by months for two stations on the west coast of Vancouver Island. Data are running ten-year averages. (Source: Canada Department of Transport.) • 78 s i t Q ZD I UJ > UJ at O N D J F M A M J J A S 90 80-70- • Estevan Point . . . . - . - n Tofino Airport 4 AM 90< 80-70 10 AM 4 PM 10 PM g • . i > > t . . • O N D J F M A M J J MONTHS Figure 10. Mean monthly relative humidity for four daily recording periods for two stations on the west coast of Vancouver Island. Data are ten-year means (1967-66). (Source: stations' records.) 79 for three of the four daily time periods, with the magnitude of these differences being greatest at night and during the late summer. The Estevan Point station has higher relative humidity during the late afternoon for the spring and summer months. However, considering the general high level of relative humidity, the moderately small difference between the two stations probably has l i t t l e effect on evapotranspiration rates. No description of the west coast climate would be complete without mention of fog. Fogs along the coast are initiated when air passing over warm offshore ocean currents is heated as well as saturated with water vapour. As this air moves landward i t passes over colder water, is cooled, and condensation takes place. A fog bank may result, especially during periods of strong atmospheric stability when vertical mixing of the air over the ocean is minimal. This fog bank is present along the coast for much of the year; i t may be 20 or more miles off the coast or as much as five miles or more inland. Fogs most frequently move inland during the morning when the air is relatively s t i l l except for very light onshore winds. Although fogs were not recorded as a standard weather station observation, i t was found that observations on visibility could be used as a measure of the frequency of fogs. During most periods of fog, visibility was recorded as being between 0.5 and 5 miles; during very dense fogs, the visibility was less than 0.5 miles. Using this method for determining for frequency, i t was found that the greatest number of fogs are recorded at the 10 a.m. observation period. The maximum number of fogs as well as the maximum 80 number of heavy fogs occur during the later summer and early f a l l ; although during the December-January period fogs are almost as high (Fig. 11). During the month of August, the Tofino airport records a mean of 10 fogs while Estevan Point has 8. The Tofino airport averages slightly more than two fogs a month more than Estevan Point throughout the year. Although the data show that the inland-most station has a greater frequency of fogs than the coastal station, i t is highly unlikely that this would stand up as a generalization of the west coast. Personal observations indicate that fogs generally extend inland in bays while s t i l l being offshore at headlands. This is likely due to stronger onshore winds in bays created by the greater land mass-shoreline ratio of bays as compared to headlands. More significant in regard to this study is the fact that the data do not show a relationship between the coastal spruce forest and a greater frequency of fogs in comparison to the hemlock - western redcedar forest. In general terms, the west coast has winds of moderate speed which may occasionally reach gale force, especially during winter storms. The wind speed varies l i t t l e throughout the year: at Estevan Point the maximum and minimum mean monthly values are 12.0 mph for February and 8.6 mph for August (Fig. 11). Probably because of its occupying a less exposed topographic position, the Tofino airport has mean monthly wind speeds consistently two to three miles per hour less than Estevan Point. ° The west coast does not have the extreme winds, either in magnitude or frequency, of some coastal areas. At Estevan Point the 81 VISIBILITY n' • 1 1 1 1 1 1 1 * O N D J F M A M J J A S WIND SPEED MEAN MONTHLY MONTHS Figure 11. Monthly means of v i s i b i l i t y and wind speed for two stations on the west coast of Vancouver Island. V i s i b i l i t y recordings are from the 10 AM observation period. Data are ten-year means (1957-66). (Source: stations' records.) 82 maximum hourly wind speed on record was 75 mph while the maximum gust was 103 mph. For the 10-year period, 1959 to 1968, only 19 hours of wind speed over 39 mph were recorded. Nevertheless, winds are of sufficient intensity to break and throw trees on exposed sites along the coast. Prevailing winds are from the west-northwest during the summer and from the east-southeast during the winter. In conclusion, a comparison of the weather records of two west coast stations failed to find climatic discrepancies of a magnitude that might suggest an explanation for the existence of the coastal spruce forest. The differences in the climatic parameters of air temperature, precipitation, humidity, and frequency of fog between the two stations are very small. Mean wind speed is slightly higher at the coastal spruce forest station, however, i t seems doubtful that this discrepancy could have much of an effect on the distribution of vegetation. SOME ASPECTS OF OCEAN SPRAY Earlier in this thesis a literature review on the ecology of ocean spray revealed that very l i t t l e is known about many aspects of the subject. For instance, only a handful of measurements on the amount of ocean spray reaching the coastline have been recorded in the literature, and those that have been made cover only very short time periods. This is to say nothing of the much more complicated problems of the spatial distribution of ocean spray, both vertical, inland and along the coast; and its relationship to such factors as wind speed and direction, coastline configuration, weather patterns, and surf 83 generation. The seasonal variation of ocean spray as well as its chemical composition are aspects of which practically nothing has been reported in the literature. Although several studies have been carried out on the effect of ocean spray on the distribution of vegetation, they have been rather limited in scope and the results are difficult, i f not impossible, to apply in other areas where environmental conditions and vegetation are different. The paucity of data on a l l aspects of ocean spray from the west coast of North America is particularly noteworthy. This section will present data from the coastal study area originating from the continuous monitoring of ocean spray over a 17-month period from May, 1966 to September, 1967. Although some of this information is not directly related to the main thrust of this thesis, because of the meagre amount of information available on ocean spray, i t was felt that this material was worthy of presentation. Ocean spray data directly related to 'the distribution of the coastal spruce forest types and to the nutrient balance of the soils will be left to a later section of this chapter. Seasonal Pattern of Ocean Spray and its Relationship to Wind and Surf This section presents data from six sites for a one-year period beginning on October 2, 1966 and ending on September 23, 1967. Although a larger number of sites were monitored during the spring and summer, because i t was necessary to use voluntary help to service the stations during the f a l l and winter, i t was possible to obtainocomplete measurements for a whole year for only six sites. This problem also 84 limited the choice of locations so that the sites used may not be as representative of the study area as might have been. Regardless of these limitations, the six stations (Fig. 5) are characteristic of a range of coastal sites and can be used to show some of the similarities and differences between them as well as the general pattern for the area. Some general characteristics of the six sites are as follows: Site 20 was located on the west-facing beach in Cox Bay and on the seaward side of the Sitka spruce forest. This beach is almost completely exposed to the open ocean, being orientated parallel to the general trend of the coastline and not protected by any headlands or islands. As a result, there is a constant influx of large waves which produce a high surf upon reaching the beach. Site 37 was located on the west-facing beach at Chesterman Bay which is slightly protected by several small islands and reefs to the west reducing the surf to a moderate level. Site 05 was on the beach at Wickaninnish Bay (Long Beach) which faces southwest and is exposed to winds from the southeast, south, and southwest. The surf is slightly lower than i t is at site 20 because the orientation of the beach is not quite parallel to the general trend of the coastline. Site 36 was located on the south side of a west-facing rocky headland between Chesterman and MacKenzie Bays. This site is somewhat sheltered from west and northwest winds by the point of the headland. Incoming waves are impeded by both offshore islands and reef, and by the irregular shoreline. 85 The Tofino airport site was located at the northwest end of Wickaninnish Bay in a clearing that was once western hemlock - redcedar forest. This site was approximately 2,500 feet from the shoreline and exposed to the ocean to the south and southeast. The surf is very similar to that at site 05. Finally, site 16 was located on west-facing MacKenzie Beach which is well protected from the open ocean by Wickaninnish and Echachis islands and several smaller islands. These obstructions tend to reduce the surf to a very low level. Measurements of ocean spray at the six sites were made for one-week periods from October 2, 1966 to April 30, 1967, and for two-week periods from May 1, 1967 to September 23, 1967. However, a l l data are expressed as weekly amounts of ocean spray. The quantity of ocean spray is expressed in grams of sodium passing through a vertical square meter perpendicular to the wind direction for the one-week time period. The yearly ocean spray totals for the six sites ranged from 42.9 g/m2 at site 16 to 248 g/m2 at site 20 while the mean for a l l sites was 108.2 g/m2 (Table 2). For the combined sites, the amount of spray measured in f a l l , as a percentage of the total measured for the entire year, was slightly less than that recorded during the winter—34 per cent compared to 39 per cent. In spring the amount of spray dropped to 16 per cent while the summer received only 11 per cent of the yearly total. Thus, there was a very uneven distribution between f a l l and winter with 73 per cent of the total compared to spring and summer with only 27 per cent of the total. 86 Table 2. Seasonal Amounts of Ocean Spray Collected at Six Sites for a One-Year Period (Oct. 2, 1966-Sept. 23, 1967) Na+ Per. Cent Na+ Per Cent Na+ Per Cent Season (g/m2) of Total (g/m2) of Total (g/m2) of Total Site 20 Site 05 Site 37 Fall 85.4 34 32.9 31 36.9 33 Winter 82.4 33 36.0 34 45.4 41 Spring 48.9 20 23.2 22 14.9 13 Summer 31.8 13 14.2 13 14.1 13 Total 248.5 106.3 111.3 Site 36 Airport Site 16 Fall 21.0 27 26.4 42 17.1 40 Winter 43.6 56 24.2 39 20.1 47 Spring 7.2 9 4.9 8 4.0 9 Summer 6.2 8 6.9 11 1.7 4 Total 78.0 62.4 42.9 All Sites Sites 20, 05, 37 Sites 36, A.P., 16 Fall 36.6 34 51.7 33 21.5 35 Winter 42.0 39 54.6 35 29.3 48 Spring 17.2 16 29.0 19 5.4 9 Summer 12.5 11 20.0 13 - 4.9 8 Total 108.3 155.3 61.1 87 Two basic patterns of spray reception emerged among the sites. One group, consisting of sites 20, 05, and 37, closely followed the pattern just described. In this group 65 per cent to 74 per cent of the year's total spray was carried in during the f a l l and winter; the remainder was distributed between spring and summer with 19 per cent and 13 per cent, respectively. Site 05 deviated slightly from this pattern with 22 per cent of the total being received during the spring. The second group, consisting of sites 36, 16, and the airport, differed significantly from the general pattern: between 81 per cent and 87 per cent of the year's total was recorded during the f a l l and winter with the remaining 13 per cent to 19 per cent recorded during the spring and summer. The proportional distribution of ocean spray throughout the year at each site is not sufficient to describe the complete picture of the yearly distribution of ocean spray. The actual amounts of spray received at the individual sites (Fig. 12) are a better indicator of the seasonal and site variations of ocean spray. Sites 20, 05, and 37 a l l received larger actual amounts of spray in f a l l and winter than sites 36, 16, and the airport (a mean of 106 g/m2 compared to 51 g/m2). Therefore, the quantity of spray received by the first group was much greater than that received by the second group for both the f a l l and winter period and the spring and summer period. Cumulative weekly values for the six sites (Fig. 13) illustrate more clearly the differences in the pattern of spray reception among the seasons and among the sites. The general seasonal pattern was as follows. In f a l l and winter the variation in the weekly intensity of F A L L W I N T E R 9 0 8 0 7 0 6 0 ^ 5 0 00 + « 3 0 2: 2 0 10 0 2 0 3 7 0 5 3 6 A . P . S i t e s 16 9 0 8 0 7 0 6 0 00 4 0 3 0 2 2 0 10 0 2 0 3 7 0 5 3 6 A . P . 16 S i t e s CM 5 0 4 0 3 0 i 2 0 cfl Z 10 00 S P R I N G M. 2 0 3 7 0 5 3 6 A . P . S i t e s M. 16 5 0 4 0 3 0 00 ~ 2 0 h + a! Z 10 h 2 0 3 7 SUMMER 0 5 36 S i t e s A . P . 16 Figure 12. Seasonal amounts of ocean spray collected at six sites for a one-year period (Oct. 2, 1966 - Sept. 23, 1967). co oo 89 Figure 13. Cumulative totals of ocean spray for six sites over a one-year period (Oct. 2, 1966 - Sept. 23, 1967). 90 spray was very great. The graph shows a general upward trend, the increase being largely due to a number of high spray weeks. These high spray weeks are very evident in the sites which have the lower yearly totals. In contrast, during spring and summer the line of the graph flattens off considerably. During this time period there was a small to moderate weekly increase and an almost complete lack of high spray weeks. Relatively small steady influxes of spray each week accounted for most of the ocean spray received during spring and summer. The frequency of high spray weeks was one of the major differences between the f a l l and winter as compared to the spring and summer. During the f a l l and winter approximately 50 per cent of the weeks can be considered high spray weeks, while during the spring and summer no high spray weeks occurred in some sites and only one or two in the others. The sites which appeared to be similar, previously, in the proportion of spray received i n each season may not in fact be similar when the weekly components are examined. Since a l l sites basically follow the overall pattern, differences in totals between the sites as shown by the divergence of the lines, depend on•the size and frequency of the high spray weeks and in the size of the more or less "normal" weekly influxes of ocean spray. The high spray weeks seem to be particularly important in accounting for a high proportion of the spray received at sites 36, 16, and the airport. This is not to say that the high spray weeks are higher at these sites than at sites 20, 05, and 37, i t simply means that they form a higher proportion of the yearly total for these s i t e s . During "normal" f a l l and winter periods and for the entire spring and summer when very few high spray weeks occurred, sites 91 36, 16, and the airport received low amounts of ocean spray in comparison to sites 20, 05, and 37. One of the factors which i s very important in determining the ocean spray regime at any site i s wind. Although i t was not possible to measure wind at each s i t e , data from the Department of Transport station at the Tofino airport was available. Hourly data from this source were extracted and used to compile frequency and mean wind speed values for eight points of the compass. These values have been grouped according to the four seasons along with the number of hourly readings with a velocity of 25 mph or greater (Table 3). This latter parameter was included as a measure of the frequency of high velocity winds. The winds varied markedly from season to season in both direction and velocity and in the number of recordings of high velocity winds. During the f a l l , the prevailing winds were from both the east and the southeast. The combined frequency of these winds was 49 per cent, and the average velocities were 10.5 and 16.4 mph respectively. High velocity winds totalled 142; of these 95 were from the southeast and 24 were from the south. In winter the prevailing wind was from the east with a frequency of 28 per cent and a mean velocity of 9.1 mph. During this season, there were 37 high velocity winds from the south, 35 from the northwest, and 33 from the southeast. In spring the prevailing wind direction changed to the northwest, with a frequency of 21 per cent and a mean velocity of 11.4 mph. The number of recordings of high velocity winds dropped to 20; most of these were from the southeast and the northwest. The prevailing wind direction remained the same during the summer with a frequency of 31 per cent and a mean 92 Table 3. Seasonal Wind Frequency, Velocity, and Number of Hourly Recordings of 25 mph or Greater (Data from Canada Department of Transport, Tofino Airport) (Oct. Fall 2-Dec. 24) (Dec. Winter 25-Mar. 19) (Mar. Spring , 20-June 17) Summer (June 18-Sept. 23) Frequency (%) Velocity (mph) Hours of 25+ mph Frequency (%) Velocity (mph) Hours of 25+ mph Frequency (%) Velocity (mph) Hours of 25+ mph • Frequency (%) Velocity (mph) Hours of 25+ mph i Wind Direction North 9 6.4 0 11 6.8 0 10 6.4 0 4 5.1 0 Northeast 6 6.8 0 10 6.8 1 5 5.6 o 1 3.9 0 East 25 10.5 10 28 9.1 1 12 6.4 0 4 4.9 0 Southeast 24 16.4 95 14 14.6 22 10 11.7 8 13 20.5 5 South 14 14.6 24 12 12.6 11 15 2.7 2 20 8.2 0 Southwest 5 10.9 3 6 12.7 3 9 9.6 0 5 8.6 0 West 6 12.9 6 6 12.7 3 14 12.3 15 10.8 1 Northwest 11 10.3 9 12 11.5 10 21 11.4 6 31 10.1 5 Mean (all directions) 12.2 10.5 9.0 9.9 Total hours of 25+ mph winds 14.7 54 20 11 Onshore Winds (S. , S.W., W., N.W.) Mean frequency 36 36 59 71 Mean velocity 12.5 12.3 10.9 9.6 Total hours of 25+ mph winds 42 31 12 6 93 velocity of 10.1 mph. There were only 11 recordings of high velocity winds; most of these were from the southeast and northwest. In general, the wind data are complicated and d i f f i c u l t to interpret. Although the seasonal frequencies and velocities for the eight directions differed considerably, the mean seasonal wind speed for a l l directions did not vary that much; these values were 12.2, 10.0, 9.0, and 9.9 for the f a l l , winter, spring, and summer respectively. The directional frequency and velocity form a definite pattern for each season, however, prevailing wind directions are not sharply defined. For example, during the f a l l there were two directions with frequencies greater than 25 per cent and three others greater than 9 per cent. Winter was even less well defined with six directions having frequencies greater than 9 per cent. It would seem that the seasonal differences in wind speed do not show a very close relationship to the seasonal pattern of ocean spray. Even i f only the onshore wind directions are used, the relationship, i f there i s one, is not apparent. Winds from the south, southwest, west, and northwest had mean velocities which range from 9.6 in the summer to 12.5 in the f a l l . This range is no larger than that incorporating a l l eight directions. On the other hand, using the onshore winds, the directional frequency is reversed in comparison to seasonal trends of ocean spray: onshore winds had a frequency of 71 per cent during the summer and only 36 per cent during the f a l l . Of a l l the wind parameters investigated, the closest relationship to the seasonal pattern of ocean spray is found in the number of hours of onshore wind with a velocity of 25 mph or greater. For the f a l l , 94 winter, spring, and summer these values were 42, 31, 12, and 6 respectively. These figures correspond quite closely to the seasonal proportions of ocean spray. Although there is no intention of implying that these are the only winds responsible for carrying ocean spray inland, i t seems as i f they may be important in producing high spray readings. Probably the best explanation for the seasonal differences in ocean spray is found in the general pattern of weather systems in the northern Pacific region. During the f a l l and winter low pressure systems are predominant (Maunder 1968). These air masses, which are often unstable, originate far out to sea commonly in the Gulf of Alaska (Chapman 1952). Before reaching the coastline, the winds associated with these storms w i l l have been blowing for several days over the open ocean. Consequently, the wave fetch may be hundreds of miles during this time and a high swell is built up, resulting in a much increased level of surf by the time the storm has reached the coast. Therefore, two of the main factors responsible for the production of ocean spray, wind, and surf, are both high at this time. Also of considerable importance during these storms is the wind direction; as the storm approaches the coast, high winds begin blowing from the east. As the storm passes through the area, these strong winds shif t in a clockwise direction and end up blowing from the northwest. This means that a l l locations along the coast, w i l l be exposed to strong, onshore winds during a normal winter storm. . The spring and summer weather pattern is quite different: high pressure systems resulting in stable conditions are predominant during 95 these seasons. Onshore winds are localized along the coast being the result of convection currents caused by differential heating of the land and water surfaces. Consequently, winds, even when strong, are usually of much shorter duration than those of the f a l l and winter and are not normally accompanied by a high surf. Also of importance is the fact that the stronger winds during the spring and summer are generally perpendicular to the trend of the coastline. This means that locations which are not orientated in this direction are not likely to receive much ocean spray from these winds. The differences in the actual amounts and the proportional distribution of ocean spray between the six recording sites is another matter which can only be very generally explained in terms of the wind and surf. Sites 20, 05, and 37 have been grouped together on the basis of receiving greater yearly amounts of ocean spray and a higher proportion of this during the spring and summer in comparison to the other sites. These sites are exposed to the southwest, a direction perpendicular to the general trend of the coastline, in locations where obstructions to the incoming waves, such as headlands and islands, is minimal. Therefore, during periods which lack storms or high winds, the surf remains moderately high. This undoubtedly is the most important factor in producing high annual levels of ocean spray at these sites as well as comparatively high summer and f a l l levels... The .higher spring and summer levels (in comparison to the other sites) is also related to their direction of exposure: these sites are exposed to the west and northwest winds which have frequencies of 35 per cent and 46 per cent 96 during these two seasons. The fact that site 20 receives considerably more spray than the other two sites in this group can probably be explained by the much higher surf at this site. Sites 36, AP, and 16 have been grouped together on the basis of receiving smaller yearly amounts of ocean spray as well as a smaller proportion of this during the spring and summer in comparison to the first group. These sites are located such that the surf is very low except during periods when storms are in the general area. Low surf may be the result of offshore islands (site 16) or by lack of exposure in a direction approximately perpendicular to the general trend of the coastline (airport site). Therefore, these sites receive high amounts of ocean spray only during winter storms when the surf is high and when strong winds from the south, a wind very common during these storms, blow spray inland. The low spray values for the spring and summer are a result of the low surf and the fact that these sites are not exposed to the frequent west and northwest winds. Foliar Damage by Ocean Spray Although no attempt was made to study the physiological effects of ocean spray on vegetation, some observations were made on spray damage to foliage of Sitka spruce and other species. Throughout most of the study area there was no apparent damage to the foliage of Sitka spruce, except perhaps for earlier than normal f a l l of needles which is difficult to determine without detailed studies. Much more obvious injury was found on young needles from both young and old trees on sites exposed to high intensities of ocean spray (Fig. 14). Injury Figure 14a. Ocean spray damage on current year leaves of Sitka spruce. Note almost complete loss of previous year's leaves. I Figure 14b. Shriveled leaf tips on current year's growth of Sitka spruce. 98 f i r s t became evident at the leaf tip which turned brown and then shriveled. This condition proceeded towards the base k i l l i n g the needle. This type of injury was observed between mid-May, the normal time of bud burst, and the end of July. It seemed to be restricted to leaves of the current year. Leaves that survived u n t i l the end of July did not appear to be susceptible to this type of injury after this date. Another condition associated with needle k i l l was delay in bud burst followed by the appearance of leaves which never grew to normal size (Fig. 15). Bud burst may be delayed u n t i l the end of June and, in some cases, the buds never opened. Late bud burst was often followed by shriveling and death of the leaves. Foliar damage apparently due to ocean spray was observed on a considerable number of species. Western hemlock, where i t occurs near the coast, often showed injury similar to that described for Sitka spruce. However, i t was much less tolerant than the spruce being badly damaged on somewhat protected sites where spruce showed no signs of injury (Fig. 16). One point that deserves further discussion is the time of year during which injury occurred and i t s relationship to the seasonal pattern of ocean spray intensity. In the previous sections weekly increments of ocean spray were shown to be much greater in f a l l and winter in comparison to spring and summer. On the other hand, injury to vegetation appeared to be restricted to the late spring and early summer. In other words the period during which vegetation injury occurred does not correspond to that time of year when ocean spray was the highest. The single most important reason for this discrepancy must 99 Figure 15b. Stunted leaves of Sitka spruce following late bud burst. July 14, 1967. 100 Figure 16. Severely deformed western hemlock is contrasted by the normal growth form of Sitka spruce in a site receiving moderate to low amounts of ocean spray. 101 be the greater susceptibility of the current year's foliage during the early part of the growing season. This lack of tolerance is quite likely related to the immature nature of protective structures such as the leaf epidermis. However, the fact that young leaves are very active in photosynthesis and transpiration during this time may also be important. The climatic conditions during the early part of the growing season may also be important in causing leaf k i l l . Although much larger amounts of ocean spray are deposited along the coast during f a l l and winter, this is also the time of very frequent rains. The accumulation of salts on the foliage is hindered, being washed off at frequent intervals by rain. During May, June, and July both the frequency and amount of precipitation are greatly reduced (Fig. 2). For example, the number of days with measurable precipitation averages 21 for the months of November through March and only 11 for May through July, while the figures for mean monthly precipitation for the same time periods are 14.4 and 4.2 inches respectively (Canada Department of Transport 1967). The above data suggest that the actual amount of salts on the foliage may be as high or higher during the early part of the growing season than during the f a l l and winter. Distribution of Ocean Spray During Growing Season  The previous section suggests that the effect of ocean spray on the vegetation of the study area is at a maximum during the growing season. For this reason, i t seems that the distribution of ocean spray during this period is worthy of some discussion. Data used for this 102 part of the study were collected from 1 5 sites for the period from May 7 to September 2 3 , 1 9 6 7 . Where possible, the number used to identify a site was taken from the closest vegetation sampling plot. Throughout the growing season, there were notable differences in the mean intensity of ocean spray received at different points along the coast (Fig. 1 7 , Table 4 ) . In general, sites which are protected from the prevailing winds, from the west and northwest, had lower spray values than those sites which are fully exposed to these winds. For example, sites 3 9 , 2 6 , and 8 6 , which are protected from these winds by headlands, a l l had mean spray intensities of less than 1 g Na+/m2/wk. On the other hand, sites 2 0 , 2 9 , and 1 0 are fully exposed to the west and northwest winds; consequently, they had much higher mean spray values ( 2 . 2 8 - 2 . 9 5 g Na+/m2/wk) . Several other sites are orientated in the same direction as sites 2 0 and 1 0 but received low amounts of spray. For example, sites 1 6 , 3 5 , 3 6 , and 4 2 had mean spray values of less than 0.4 g Na+/m2/wk. Obviously, as explained earlier, exposure itself does not provide the whole answer or sites with similar orientations, such as the west-facing beach sites, would have similar mean spray values. Differences in the size of the surf at these similarly orientated sites helps to explain the variation in the amount of spray received. A series of west-facing beach sites, 3 0 , 2 , 1 6 , and 3 5 , show a progressive decrease in spray intensity corresponding to a similar decreasing trend in the size of the surf (Fig. 1 7 , Table 4 ) . The difference in the size of the surf throughout this series can be explained by the degree of protection provided by offshore islands. Site 103 Figure 17. Map of the study area showing the relationship between the width of the coastal Sitka spruce forest and the amounts of incoming ocean spray. Spray measurements are weekly averages for the 1966 growing season. Table 4. Mean Weekly Amounts of Sodium Collected at Ocean Spray Measuring Stations During the 1967 Growing Season (Station Numbers Correspond to the Closest Sample Plot in Most Cases). Station Location Na (g/m2/wk) Station Location Na (g/m2/wk) 02 Chesterman Bay 1.09 36 Chesterman Bay 0.36 03 Frank Island 0.86 39 Florencia Bay 0.17 05 Wickanninish Bay 1.30 42 Wickanninish Bay 0.21 10 Wickanninish Bay 2.28 56 Frank Island 0.98 16 MacKenzie Bay 0.13 86 Schooner Cove 0.91 20 Cox Bay 2.40 87 Portland Point 2.06 26 Schooner Cove 0.64 91 Cox Bay 0.06 29 Box Island 2.95 94 MacKenzie Bay 0.21 35 Emerald Bay 0.11 100 Airport 0.38 105 20 received the highest amount of spray, had the highest surf, and was completely exposed to the open ocean. At the other extreme, site 35 received the lowest amount of spray, and the lowest surf, and was well protected from the open ocean by Wickaninnish Island. The width of the surf zone at the various sites is clearly evident on the air photograph (Fig. 17). On rocky headlands and islands the process of surf formation and the production of ocean spray is quite different from that just described for beaches. Along the rocky coastline the transition from sea to land is very sharp; there being almost a complete lack of shallow water except for the occasional small reef. Therefore, the wide surf zone associated with the beach is lacking because the waves do not break repeatedly as they approach the coastline. Instead, they break very close to, or right on, the shoreline forming a very active but narrow surf zone. This type of surf tends to produce large amounts of spray; however, much of this appears to be in the form of large droplets that are not carried very far inland. In many places the rocky headlands rise very steeply from the shoreline. In such situations, much of the spray is intercepted by the bare rock "walls" and never reaches the vegetation. Even though the rocky headlands do not have the wide surf zones common to the more exposed beaches, the amount of ocean spray received by the forest vegetation closest to the coastline is s t i l l very high. For example, sites 87 and 29 occupy headlands that are well exposed to the prevailing winds and the open ocean, and received spray amounts comparable to the most exposed beach sites. In both of these headland 106 sites, spray production is enhanced by offshore reefs and small islands which increase the surf area in the direction of the prevailing winds. On the other hand, sites that are not located downwind from reefs received less spray than their beach counterparts with similar exposures to prevailing winds and the open ocean. Sites 3, 36, and 42 are good examples of such sites. In general, the pattern of coastal distribution of ocean spray during the growing season can be summarized as follows: sites that are well exposed to both the open ocean and the prevailing west and north-west wind receive the highest amounts of spray. This group includes sites 10, 20, 29, and 87; a l l had mean spray values between 2 and 3 g Na+/m2/wk. Sites receiving moderate amounts of spray (0.6-1.3 g Na+/ m2/wk) are exposed to the prevailing winds in most cases but are partially protected from the open ocean resulting in a reduction in surf activity. This group includes sites 2, 3, 5, 26, 56, and 86. The remaining sites (16, 35, 36, 39, and 42) received less than 0.4 g + 2 Na /m /wk. These sites are extremely well protected from either the open ocean or the prevailing winds, or both. Ocean Spray Dynamics and the Coastal Spruce Forest  The relationship between the width of the spruce forest and ocean spray can be demonstrated by using the beaches in the northwestern part of the study area. The beaches in Emerald, MacKenzie, Chesterman, and Cox bays have widths of 10, 90, 210, and 560 feet respectively 1 (Fig. Measurements of the width of the Sitka spruce forest do not include the width of the Tsuga heterophylla - Picea sitchensis forest type. 107 17). Ocean spray values for the four beaches, listed in the same order, are: 0.11, 0.13, 1.09, and 2.40 g Na+/m2/wk (Table 4). Therefore, the progression of the spruce forest widths on the four beaches corresponds to a similar progression in ocean spray values. The relationship between the width of the spruce forest along beaches and ocean spray can also be demonstrated by grouping the beaches according to the two factors. On beaches where the width of the spruce forest is less than 100 feet, the amount of spray received is less than + 2 0.4 g Na /m /wk. On beaches with a spruce forest width between 100 and 400 feet, the ocean spray values ranged from 0.6 to 1.30 g Na+/m2/wk. Sites 2, 05, and 26 f a l l in this group. Beaches with the widest spruce band, sites 10 and 20, have widths of 400 to 600 feet and had ocean spray values greater than 2.2 g Na+/m2/wk. On rocky headlands and islands the relationship between the width of the spruce forest and ocean spray is much more difficult to interpret. One of the biggest problems is measuring the width of the spruce forest. On the beaches this is a simple matter: one simply follows a line set at right angles to the beach and measures the distance between the seaward side of the spruce forest (the tree line) and the beginning of the transitional forest. However, on rocky headlands this measurement becomes a real problem. The difficulty is largely due to the extreme irregularity of the coastline which makes i t very difficult to decide which direction to use in measuring the width of the spruce forest. Also, many headland sites are exposed to the ocean in several different directions; for example, those on narrow points and small islarfds. If the vegetated area is not large, the spruce forest may cover a l l of i t 108 so that there is no transitional forest to mark the inland boundary. Even though there were considerable inaccuracies involved in measuring the width of the spruce forest on rocky headlands, the data obtained suggest a relationship between this width and the spray intensity. The width is greatest at sites 29 and 87 where i t is 105 and 120 feet respectively. These sites received between 2 and 3 g Na+/m2/wk, which is approximately the same amount of spray received at the highest beach s i t e s . The remaining headland sites have widths of 100 feet or less and had spray values between 0.36 and 0.98 g Na+/m2/wk. A comparison of the spruce forest on beaches and headlands indicates that while both types of sites definitely show a progressive decrease in width with a decrease in spray intensity, the beach sites have a much wider spruce forest for comparable levels of spray. For example, sites 20 and 29 both have extremely high levels of spray, yet the width of the spruce forest at site 20 is 560 feet compared to 105 feet at site 29. These widths represent the maximum recorded on the two types of sites indicating that the spruce forest on headlands has a much more narrow range than that on the beaches. This fact i s easily verified by comparing the width of the two types in Figure 17. The discrepancy in the width of the spruce forest on beaches and headlands presents an interesting problem. A possible explanation might be found in the vert i c a l distribution of ocean spray. The rationale for this lies in the assumption that ocean spray must f i r s t reach the vegetation before i t can affect i t . The f i r s t line of trees or the f i r s t forest stand inland from the coastline represents an obstacle over which the spray must pass i f i t is to continue inland. 109 Several stations were set up to study the vertical distribution of ocean spray. One of these was located on a section of rocky coastline (Fig. 18) and the other on a beach (Fig. 19). Each station consisted of two instruments: one of these was placed on the seaward side of the forest four feet above ground level while the other was placed at the top of the tree canopy. Measurements were taken for one-week time periods from July 2 to September 9, 1966. At the rocky headland site (plot 29), the ground-level instrument recorded much greater amounts of spray than the tree canopy instrument (Fig. 20). This is believed to be due to the short distance between the forest and the rocky shoreline on which the ocean spray is generated. This distance is not great enough to allow adequate vertical mixing of the spray necessary to produce a high spray content in the air passing over the top of the canopy (Fig. 21). Instead, the spray remains concentrated near the ground and much of i t is intercepted by the rocks and by the lower strata of trees and shrubs. The greater frequency of the larger spray droplets associated with headlands is also a contributing factor to the poor vertical mixing. At the beach site (plot 20) the results were just the opposite with the canopy-level instrument recording more spray than the instrument at ground level. On beaches the ocean spray is generated by the surf which is a considerable distance from the forest. This distance is apparently great enough to allow for good vertical mixing of the spray by the time i t reaches the forest along the upper edge of the beach (Fig. 22). It can be concluded that a much greater proportion of the F i g u r e 18. R o c k y h e a d l a n d s i t e u s e d i n t h e s t u d y o f v e r t i c a l o c e a n s p r a y d i s t r i b u t i o n . ( B o x I s l a n d ) . F i g u r e 19. B e a c h s i t e u s e d i n t h e s t u d y o f v e r t i c a l o c e a n s p r a y d i s t r i b u t i o n C o x B a y ) . VERTICAL DISTRIBUTION OF OCEAN SPRAY ROCKY COASTLINE (Box Island) \ E + o Z cn MEAN VALUES GROUND LEVEL 2.62 TOP OF CANOPY 1.04 July9 July 18 July 24 July 30 August 6 August 13 August 20 August 27 Sept. 3 Sept. 9 6 -\ E Z O) i — GROUND LEVEL TOP OF CANOPY BEACH (Cox Bay) MEAN VALUES GROUND LEVEL 1.10 TOP OF CANOPY 1.29 July 9 July 18 July24 July 30 Aug.6 Aug.13 Aug.20 Aug.27 Sept 3 Sept.9 1966 Figure 20. Vertical distribution of ocean spray on a rocky coastline site and a beach site. Figure 21. The process of ocean spray gener-ation along the rocky shoreline as i t occurs on Box Island. 113 incoming ocean spray passes over the top of the canopy on beaches than i t does on rocky headlands. Actually, the sites used for this part of the study are not as typical as they might have been. This is particularly true of the headland site which occupied a narrow point (actually an island at; high tide) so that the canopy-level instrument could receive spray from any direction while the ground-level instrument was protected by the forest from some wind directions. If a site had been chosen along a more regular section of coastline, the amount of spray recorded at canopy level would likely have been somewhat less. It is unfortunate that comparable data for the two stations is not available for the early part of the growing season. Data are available for site 20 but the canopy-level instrument at site 29 was not installed until early July. During the earlier part of the season, which was marked by lengthy periods of fairly strong west and northwest winds, the canopy-level instrument at site 20 recorded ocean spray values greater than two times that at ground level. For the week with the highest spray values, the top of the canopy received more than four times more spray than at 4 feet above ground level: 13.2 compared to 3.3 mg Na +/in 2. Up to this point several relationships concerning ocean spray and the coastal spruce forest have been suggested. These are: (1) the spray received during the growing season has the greatest effect on the vegetation; (2) the width of the spruce forest corresponds to the amount of ocean spray measured on the seaward side of the forest during the growing season; and (3) of the total incoming spray, that component passing over the top of the canopy must have the greatest effect on the 114 width of the spruce forest stand. If, in fact, the width of the spruce forest stand is related to ocean spray, and i f the spray component passing over the top of the canopy is the most important part of the total incoming spray; then i t follows that the next step should be to examine this component as i t moves inland from the first line of trees along the coast. To do this, a transect was established on Cox Bay starting at the beach and running inland through the spruce forest, the transitional forest (Tsuga heterophylla - Picea sitchensis forest type) and ending in a western redcedar - western hemlock stand immediately inland of the transitional forest (Fig. 47, page 207). Several of the wind-vane-type of ocean spray measuring instruments were placed along this transect: one at 4 feet above ground level on the beach, one at the top of the tree canopy near the beach, and one at the top of the canopy in the redcedar - hemlock stand. Although i t would have been helpful to have had more of these instruments located along the transect, the height of the trees (and the lack of courage on the part of the author) prohibited this. To overcome this problem, throughfall precipitation was collected at six stations: four in the spruce, forest, one in the transitional forest, and one in the redcedar - hemlock forest. Water was collected at one-week intervals when rainfall permitted and analyzed for sodium concentration. The values so obtained were then converted to + 2 weight per unit area (g Na /m /wk). Since rain water washes sodium from ocean spray off the leaves, the weight of the sodium ion contained in the throughfall rain water should be indicative of the amount of sodium deposited on the leaves. The throughfall data is used here only 115 as a relative value in comparing the intensity of ocean spray at the tree canopy level of the different sites. Measurements were taken from May 16 to September 10, 1966. Data collected by the canopy height ocean spray instruments show a considerable difference in spray intensity between the two ends of the transect (Table 5). Near the beach a mean value of 2.89 g Na+/m2/wk was measured while the station at the inland end of the transect recorded only 0.03 g Na+/m2/wk. The throughfall data for the same two sites also showed a strong difference: 1.52 g Na+/m2/wk at the beach end of the transect and 0.13 for the inland end. Starting at the beach, the series of throughfall stations show that the amount of spray at canopy height was relatively consistent for a distance of 450 feet inland. These stations a l l recorded values between 1.29 and 1.52 g Na+/m2/wk. However, station 5 at a distance of 560 feet from the beach recorded only 0.18 g Na+/m2/wk, a value very close to that recorded at station 6 (0.13 g Na+/m2/wk) at the inland end of the transect. The most logical explanation for the pattern of ocean spray intensity in the upper part of the tree canopy lies with the profile of the tree canopy along the transect. Starting from the beach, the canopy gradually rises over the first 450 feet. This part of the transect corresponds to the first four stations, a l l of which recorded close to the same amounts of sodium. The canopy height is at a maximum at station 4 and begins to f a l l off inland from this point. At station 5 the canopy height is 20 feet less than at station 4. This decline corresponds to a very sharp drop in the sodium content of the through-f a l l . Inland from station 5 the canopy height continues to decrease Table 5. Cox Bay Transec