UBC Faculty Research and Publications

Changes in net ecosystem productivity and greenhouse gas exchange with fertilization of Douglas fir:.. Grant, Robert F.; Black, T. Andrew; Jassal, Rachhpal S.; Bruemmer, Christian 2010-10-31

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


52383-Black_AGU_2009JG001094.pdf [ 831.3kB ]
JSON: 52383-1.0041931.json
JSON-LD: 52383-1.0041931-ld.json
RDF/XML (Pretty): 52383-1.0041931-rdf.xml
RDF/JSON: 52383-1.0041931-rdf.json
Turtle: 52383-1.0041931-turtle.txt
N-Triples: 52383-1.0041931-rdf-ntriples.txt
Original Record: 52383-1.0041931-source.json
Full Text

Full Text

Changes in net ecosystem productivity and greenhouse gasexchange with fertilization of Douglas fir: Mathematical modelingin ecosysR. F. Grant,1T. A. Black,2R. S. Jassal,2and C. Bruemmer2Received 6 July 2009; revised 19 November 2009; accepted 1 April 2010; published 8 October 2010.[1] The application of nitrogen fertilizers to Douglas fir forests is known to raise netecosystem productivity (NEP), but also N2O emissions, the CO2equivalent of which mayoffset gains in NEP when accounting for net greenhouse gas (GHG) exchange. However,total changes in NEP and N2O emissions caused by fertilizer between times ofapplication and harvest, while needed for national GHG inventories, are difficult toquantify except through modeling. In this study, integrated hypotheses for soil and plantN processes within the ecosystem model ecosys were tested against changes in CO2andN2O fluxes recorded with eddy covariance (EC) and surface flux chambers for 1 yearafter applying 20 g N m−2of urea to a mature Douglas fir stand in British Columbia.Parameters from annual regressions of hourly modeled versus measured CO2fluxesconducted before and after fertilization were unchanged (b = 1.0, R2= 0.8, RMSD =3.4 mmol m−2s−1), indicating that model hypotheses for soil and plant N processes didnot introduce bias into CO2fluxes modeled after fertilization. These model hypotheseswere then used to project changes in NEP and GHG exchange attributed to the fertilizerduring the following 10 years until likely harvest of the Douglas fir stand. Increased CO2uptake caused modeled and EC‐derived annual NEP to rise from 443 and 386 g C m−2inthe year before fertilization to 591 and 547 g C m−2in the year after. These gainscontributed to a sustained rise in modeled wood C production with fertilization, whichwas partly offset by a decline in soil C attributed in the model to reduced root Cproductivity and litterfall. Gains in net CO2uptake were further offset in the model by arise of 0.74 g N m−2yr−1in N2O emissions during the first year after fertilization, whichwas consistent with one of 1.05 g N m−2yr−1estimated from surface flux chambermeasurements. Further N2O emissions were neither modeled nor measured after the firstyear. At the end of the 11 year model projection, a total C sequestration of 1045 g C m−2was attributed to the 20 g N m−2of fertilizer. However, only 119 g C m−2of this wassequestered in stocks that would remain on site after harvest (foliage, root, litter, soil). Theremainder was sequestered as harvested wood, the duration of which would depend on useof the wood product. The direct and indirect CO2‐equivalent costs of this application,including N2O emission, were estimated to offset almost all non‐harvested C sequestrationattributed to the fertilizer.Citation: Grant, R. F., T. A. Black, R. S. Jassal, and C. Bruemmer (2010), Changes in net ecosystem productivity andgreenhouse gas exchange with fertilization of Douglas fir: Mathematical modeling in ecosys, J. Geophys. Res., 115, G04009,doi:10.1029/2009JG001094.1. Introduction[2] Most temperate and boreal forest ecosystems areconsidered to be nitrogen limited, so that N fertilizerapplication may increase forest CO2uptake and consequentC storage [Aber et al., 1989; Johnson and Curtis, 2001],slowing the rise in atmospheric CO2concentration andincreasing wood supply. However, N fertilizer productioninvolves the emission of CO2, and its application may alsoincrease emission of N2O[Jassal et al., 2008; Matson et al.,1992; Sitaula et al., 1995b], an important greenhouse gas(GHG). These emissions offset the effects of increased CO2uptake on net GHG exchange and thereby on the radiativeproperties of the atmosphere thought to drive climatechange. Therefore, a full accounting of net GHG exchange1DepartmentofRenewableResources,UniversityofAlberta,Edmonton,Alberta, Canada.2Biometeorology and Soil Physics Group, Faculty of Land and FoodSystems, University of British Columbia, Vancouver, British Columbia,Canada.Copyright 2010 by the American Geophysical Union.0148‐0227/10/2009JG001094JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 115, G04009, doi:10.1029/2009JG001094, 2010G04009 1of17from fertilizer application in forests needs to include effectson net CO2uptake by trees as well as on CO2and N2Oemissions from N fertilizer production and application.[3] Gains in net CO2uptake from fertilization of foreststands are thought to be caused by root uptake and root‐shoot transfer of fertilizer N, resulting in increased foliar Nconcentrations that raise photosynthetic capacity and henceCO2fixation for several years after application [Hopmansand Chappell, 1994]. However, these gains are highly var-iable because they are affected by the biochemical andhydrologic characteristics of the fertilized stands. Gains maybe smaller with lower C/N ratios in tree foliage or the soilLFH layer when N limitations are less severe [Edmonds andHsiang, 1987; Hopmans and Chappell, 1994]. Gains mayalso be smaller when uptake of fertilizer N is reduced byrapid nitrification of fertilizer products [Matson et al., 1992]and subsequent leaching of nitrate [Flint et al., 2008].[4] Gains in net CO2uptake from fertilization are alsoaffected by changes in C allocation to organs with differingturnover rates within trees. With improved N supply, treestypically allocate more resources to shoot growth, much of itin boles with slower turnover, than to root growth with morerapid turnover. This reallocation is based on a theory offunctional equilibrium between shoots and roots in whichnutrient allocation is based on proximity to sites of acqui-sition and on rates of consumption [Ericsson et al., 1996].Reduced root allocation is consistent with reduced soilrespiration frequently measured in fertilized forest stands[Giardina et al., 2004; Olsson et al., 2005]. Changes inroot‐shoot allocation with fertilization have been shown inconiferous seedlings [e.g., Iivonen et al., 2001; Kaakinen etal., 2004]; however, evidence for such changes in maturestands remains inconclusive. Iivonen et al. [2006] foundsmall decreases in allocation of C and N to fine roots withfertilization, but little change in allocation of C and N toaboveground organs or coarse roots in spite of a largeincrease in growth. Earlier studies have found both increasesand no changes in allocation to roots with fertilization. In afurther complication, some studies have indicated thatimproved N supply causes shoot growth to be allocatedmore to foliage and branches with more rapid turnover thanto boles with slower turnover [Gower et al., 1992; Valinger,1993]. These changes in allocation require that gains in netCO2uptake and C storage attributed to fertilization beevaluated in comprehensive ecosystem studies rather thanfrom simple estimates of gains in wood C.[5] Increases in N2O emissions from fertilization of foreststands are thought to be caused by nitrification and deni-trification of fertilizer N and its mineral products[Martikainen and de Boer, 1993; Matson et al., 1992]. N2Oemissions from nitrification are known to be favored by lowpH found in acidic forest floors that develop under conif-erous stands [Martikainen and de Boer, 1993; Sitaula et al.,1995b], suggesting that these emissions may be supple-mented by chemodenitrification [e.g., Mørkved et al., 2007].Estimates of N2O emissions derived from surface fluxmeasurements in forests are highly variable. Although theseemissions are thought to be small [Matson et al., 1992], thegrowing use of N fertilizer may cause them to rise. Jassal etal. [2008] measured substantial N2O emissions (5% of ad-ded N) during several months after a fertilizing a Douglas firstand in British Columbia, although other researchers havemeasured less (e.g., 0.5% by Sitaula et al. [1995b] during 1month after fertilizing Scots pine, 0.35% by Matson et al.[1992] during a growing season after fertilizing Douglasfir, and 0.2% by Bowden et al. [1991] during 1 year afterFigure 1. Conceptual model of key C and N transfers in ecosys. Numbers in brackets refer to equationsin Appendix A.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G040092of17fertilizing in red pine). However, even small emissions ofN2O could partially offset gains in CO2uptake when esti-mating net GHG exchange from N fertilizer application.[6] The variability and duration of changes in CO2andN2O exchange caused by fertilizer N application in differentforest stands complicate efforts in GHG accounting forfertilizer use directly from site studies. Such accountingcould benefit from mathematical models based on a com-prehensive set of integrated hypotheses for the biologicaland physical processes driving changes in GHG exchange.In this study, integrated hypotheses for soil and plant Nprocesses within the detailed ecosystem model ecosys aretested against changes in CO2and N2O fluxes recorded witheddy covariance (EC) and surface flux chambers during thefirst year after applying 20 g N m−2of urea to a matureDouglas fir stand near Campbell River in British Columbia.These hypotheses are then used to project changes in GHGexchange and in C and N storage from this applicationduring the following 10 years until the likely harvest of thestand.2. Model Development2.1. General Overview[7] Key algorithms for C and N transformations that driveecosystem productivity and greenhouse gas exchange inecosys are described in detail elsewhere [Grant andFlanagan, 2007; Grant et al., 1993a, 1993b, 2005, 2006a,2006b, 2006c, 2007a, 2007b, 2007c; Grant, 2004; Grantand Pattey, 2003]. Algorithms that directly govern thetransformation, uptake, and assimilation of fertilizer N aredescribed in summary form below with reference to equa-tions and definitions listed in Appendix A and representedin Figure 1.2.2. Urea Hydrolysis[8] Urea fertilizer first undergoes hydrolysis to NH3(equation (A1a)) at rates calculated as the product of specifichydrolysis activity and total microbial C (as a proxy forurease), constrained by an Arrhenius function of soil tem-perature [Moyo et al., 1989; Vlek and Carter, 1983] and aMichaelis‐Menten function of urea concentration [Lal et al.,1993] incorporating a competitive inhibition term driven bysoil water content to simulate effects of soil drying [Vlekand Carter, 1983]. The hydrolysis product NH3is main-tained in equilibrium with NH4+according to soil or residuepH (equation (A1b), in which states they undergo otherreactions described below.2.3. Mineralization‐Immobilization[9] Each kinetic component j (j is labile or resistant) ofeach microbial population m (m is obligately aerobic bac-teria, obligately aerobic fungi, facultatively anaerobic deni-trifiers, anaerobic fermenters plus H2‐producing acetogens,acetotrophic methanogens, hydrogenotrophic methanogensand methanotrophs, NH3and NO2−oxidizers, and non-symbiotic diazotrophs) in each substrate‐microbe complex i(i is coarse woody residue, fine nonwoody residue, partic-ulate organic matter, or humus) in the surface residue andeach soil layer l seeks to maintain a set population‐specificC/N ratio by mineralizing NH4+(equation (A2a)) or by im-mobilizing NH4+(equation (A2b)) or NO3−(equation (A2c)).Changes in microbial C and N arise from changes in organicsubstrate availability and quality and in soil temperature andwater content. Provision is made for C/N ratios to riseabove set values during immobilization but at a cost tomicrobial function. Under these conditions, provision is alsomade for internal recycling of microbial N. These transfor-mations control the exchange of N between organic andinorganic states in soil. Equations representing these trans-formations are given in greater detail by Grant et al. [1993a,1993b].2.4. Nitrification[10] Rates of NH3and NO2−oxidation are calculated fromspecific NH3or NO2−oxidizer activities multiplied by NH3or NO2−oxidizer biomasses, constrained by an Arrheniusfunction of soil temperature and Michaelis‐Menten func-tions of aqueous NH3[Stark and Firestone, 1996] or NO2−[Blackburne et al., 2007] concentrations and aqueous CO2concentrations (equations (A3a) and (A3b)). Rates of NO2−oxidation are inhibited by aqueous NH3and HNO2con-centrations [Blackburne et al., 2007] (equation (A3c)).Reduction of NO2−to N2O by nitrifiers (equation (A3d)) isdriven from demand for electron acceptors to oxidize NH3(equation (A3a)) unmet by O2because of diffusion limita-tions, and is constrained by Michaelis‐Menten functions ofaqueous NO2−and CO2. Equations for these processes aregiven in greater detail by Grant et al. [2006b].2.5. Biological Denitrification[11] Demand for electron acceptors from denitrifier Coxidation unmet by O2because of diffusion limitations[Grant et al., 2006b] drives the sequential reduction of NO3−,NO2−, and N2O, constrained by Michaelis‐Menten functionof NO3−,NO2−,andN2O concentrations [Yoshinarietal.,1977] (equations (A4a), (A4b), and (A4c)). All gaseousproducts undergo convective‐dispersive transfer and aque-ous and gaseous phases. Equations for these processes aregiven in greater detail by Grant et al. [2006b].2.6. Chemodenitrification[12]NO2−from nitrification and denitrification is indynamic equilibrium with HNO2depending on soil or res-idue pH (equation (A5a)). HNO2concentration drives first‐order decomposition [Cleemput and Samater, 1996](equation (A5b)) to N2O[Mørkved et al., 2007] (equation(A5c)) and other N products [Cleemput and Samater, 1996].2.7. Root and Mycorrhizal Uptake[13]NH4+and NO3−uptake by roots and mycorrhizae iscalculated from mass flow plus radial diffusion through soilbetween adjacent roots and mycorrhizae (equations (A6a)and (A6c)) coupled with active uptake at root and mycor-rhizal surfaces (equations (A6b) and (A6d)) in multilayeredsoil profiles. Both fluxes depend on root length densityderived from a root growth model, as described in greaterdetail by Grant [1998]. Uptake products in each soil layerare added to nonstructural N pools in roots and mycorrhizae.If N/C ratios of these nonstructural pools rise above thoserequired for growth, inhibition of root and mycorrhizal Nuptake is invoked to keep uptake in balance with CO2fix-ation [Grant, 1998].GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G040093of172.8. Plant Assimilation[14] Nonstructural N pools, generated from root andmycorrhizal uptake, are coupled with nonstructural C pools,generated from CO2fixation, in mycorrhizae, roots, andbranches. Transfers among these pools (equations (A7a) and(A7b)) are driven by concentration gradients generated byacquisition versus consumption of nonstructural N and C inmycorrhizae, roots, and branches [Grant, 1998] (equations(A7c) and (A7d)). Ratios of nonstructural N and C inbranches govern CO2fixation (equations (A7f), (A7g), and(A7h)) [Farquhar et al., 1980] by (1) setting ratios ofstructural N and C in leaves (A7e) and hence maximumcarboxylation rates (equations (A7i) and (A7j) and (2)determining rubisco activation through product inhibition(A7k). Nonstructural C pools also drive autotrophic respi-ration (Ra) to meet maintenance and growth requirements(Rmand Rg)[Grant et al., 2007c, equations (C12)–(C16)]. IfRafalls below Rm, the deficit drives the withdrawal ofremobilizable C and N from older leaves and supportingstructures into C and N nonstructural pools, the former ofwhich is used to sustain Rmand the latter of which can betranslocated to newer foliage. All model equations for C andN are fully coupled to counterparts for phosphorus. Equa-tions for these processes are given in greater detail by Grantet al. [2007c] and in earlier references cited therein.3. Field Experiment3.1. Site Description[15] The field experiment was conducted in a standdominated by Douglas fir (Pseudotsuga menziesii), but alsoincluding western red cedar (Thuja plicata), western hem-lock (Tsuga heterophylla), and red alder (Alnus rubra). Thisstand started to regenerate in 1949 after fires and logging in1939 and 1943 on a well‐drained Humo‐Ferric Podzol(Quimper sandy loam, Table 1) about 10 km southwest ofCampbell River (49°52′7.8″N, 125°20′6.3″W) on the eastcoast of Vancouver Island, Canada (mean annual tempera-ture and precipitation: 8.6°C and 1450 mm). This stand isfurther described in the works of Humphreys et al. [2006]and Morgenstern et al. [2004].3.2. Fertilizer Treatments[16] A 1115 ha area of the stand surrounding the eddycovariance (EC) flux tower was aerially fertilized with200 kg N ha−1urea on 13 January 2007, following standardcommercial practice. A 17 ha experimental area on thesoutheast side of the fertilized area remained unfertilized fora field plot study to measure changes in soil N2O emissionsand in plant C and N stocks caused by fertilizer. Within thisarea, urea was applied manually at 200kg N ha−1on 11 April2007 to four of eight 100 m2plots, while the remaining fourplots remained unfertilized. In addition, eight 2 m × 2 mplots, extended by 2 m wide buffers on each side, wereestablishedneartheECfluxtowerinAugust2006tomeasurechanges in soil CO2fluxes caused by fertilizer. These plotswere protected from aerial fertilization on 13 January 2007,and then four were fertilized manually with urea at 200 kg Nha−1on 31 January 2007. For both sets of plots, manualfertilizer application achieved the same fertilizer rate as thatfrom the aerial application, but with greater uniformityrequired for chamber flux measurements.3.3. Site Measurements[17] Ecosystem CO2and energy exchange have beenmeasured by EC continuously since 1998 at a flux tower site(BC‐DF49) within the fertilized area as described earlier[Jassal et al., 2007; Humphreys et al., 2006; Morgensternet al., 2004]. Soil respiration (Rs) has been measured con-tinuously since 2003 by one automated surface flux chambernear the flux tower. Rsmeasured with this chamber has beenshown to represent well the spatial average of Rsfor thestand [Jassal et al., 2007].[18] Soil N2O effluxes were measured with a vented staticchamber mounted on 21 cm diameter PVC collars installedin each 100 m2plot within the 17 ha experimental area. N2Oemissions from these plots were recorded every 2–3 weeksfrom fertilizer application on 11 April to the end of the year,and corroborated with emissions calculated from soil N2Oconcentration gradients [Jassal et al., 2008]. Soil CO2effluxes were measured every 2–4 weeks using portablevented chambers mounted on two collars (10 cm innerdiameter, 4 mm wall thickness) in each 2 m × 2 m plot nearthe flux tower, as described by Jassal et al. [2010]. Furtherdetails of site management and flux measurements are givenby Jassal et al. [2008].4. Model Experiment4.1. Model Initialization and Spinup[19] Before testing ecosys with the CO2and N2O fluxesrecorded at BC‐DF49, the model had to reproduce siteconditions by simulating site history. This was accom-plished by initializing ecosys with the biological propertiesof Douglas fir and a pioneer bush understory and with thephysical and chemical properties of the Quimper sandy loam(Table 1). Ecosys was also initialized with stocks of coarseand fine residue estimated to remain after a stand‐replacingfire presumed to have occurred in the model year 1780. Thisyear was selected to start the model runs so that theaboveground biomass generated by the model in 1919 wasconsistent with that estimated from wood volumes recordedduring a 1919 timber cruise. The Douglas fir and bush wereseeded after the presumed 1780 fire, grown from modelTable 1. Key Properties of the Quimper Sandy Loam Near Camp-bell RiveraQuimper Horizon LFH Ahe Bf Bm Bm BCcDepth to Bottom (m) 0.03 0.09 0.23 0.53 0.83 1.33Bulk Density (Mg m−3) 0.1 1.3 1.5 1.5 1.5 1.9Field Capacity (m3m−3) 0.21 0.21 0.21 0.19 0.13 0.10Wilting Point (m3m−3) 0.14 0.08 0.08 0.06 0.04 0.03Ksat(mm h−1) 210 10 10 100 100 1Sand (g kg−1) 0 540 660 760 720 740Silt (g kg−1) 0 380 270 220 250 240Clay (g kg−1) 0 80 70 20 30 20Coarse Frag. (m3m−3) 0 0.33 0.33 0.33 0.33 0.33CEC (cmol kg−1) 124 20 10 9 10 9pH 4.7 5.0 4.9 4.9 4.7 4.8Organic C (g kg−1) 385 74 62 17 11 9Total N (g Mg−1) 8100 2200 1900 850 750 720aFrom Humphreys [2004] and Soil Landscapes of Canada v. 3.1.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G040094of17years 1780 to 1919 under hourly weather data recorded atBC‐DF49 during 2004, then from model years 1920 to 1997under daily weather data constructed from meteorologicalrecords [Régnière and St‐Amant, 2007], and then from 1998to 2006 under hourly weather data recorded at BC‐DF49during 1998–2006. The modeled Douglas fir stand wasburnt in 1939, salvage logged and burnt again in 1943, re-seeded in 1949, and fertilized with 20 g N m−2as broadcasturea in 1994 so that the age and disturbance history of themodeled stand at the end of 2006 corresponded to that atBC‐DF49 as derived from field records.4.2. Boundary Conditions[20] During the model run, atmospheric CO2concentra-tion (Ca) was prescribed to rise exponentially from 280 to385 mmol mol−1, and concentrations of NH4+and NO3−inprecipitation used to simulate wet N deposition were pre-scribed to rise from historical values based on Holland et al.[1999] to current values based on Meteorological Service ofCanada [2004]. Atmospheric concentration of NH3used tosimulate dry N deposition was maintained at 0.002 mmolmol−1. A background mortality rate of 0.75% per year wasapplied to Douglas fir during model runs, simulating naturalself‐thinning.4.3. Model Testing[21] The model run was then continued during 2007 underhourly weather data recorded at BC‐DF49, with and without20 g N m−2of urea broadcast on 13 January to simulatechanges in CO2exchange, or on 11 April to simulatechanges in N2O emissions. The ability of the model tosimulate changes in forest productivity with N fertilizer wasevaluated by testing modeled CO2fluxes against EC CO2flux measurements under comparable environmental con-ditions before and after fertilization. The ability of the modelto simulate changes in N2O emissions with N fertilizer wasevaluated by testing modeled N2O fluxes against surfacechamber measurements recorded on the fertilized versusunfertilized plots near BC‐DF49. To examine longer‐termeffects of the 2007 fertilization, these model runs wereextended by a further 10 years until the likely harvest date ofthe stand, under weather recorded during 2005, a meteoro-logically near‐average year. Values of all model parametersin this study were unchanged from those in earlier studies[e.g., Grant, 2004; Grant et al., 2005, 2006a, 2006c,2007b, 2007c].5. Results5.1. Uptake and Assimilation of Fertilizer N[22] Fertilization on 13 January 2007 hastened root uptakeand root‐shoot transfer of N in the model, raising averageneedle N content of Douglas fir by 14% from 0.022 g N gC−1in the unfertilized stand to 0.025 g N g C−1in the fer-tilized stand by 6 December 2007, about 11 months afterfertilization (Table 2). This increase was smaller than onefrom 0.023 to 0.032 g N g C−1(assuming needles are 50% C)measured at BC‐DF49 on this date by Jassal et al. [2008],butsimilartoonefrom 0.022to0.027gNgC−1measuredbyNason et al. [1990] 30 weeks after applying 20 g N m−2asurea to a nearby Douglas fir stand of similar age.[23] In the model, this rise was driven by hydrolysis of theurea fertilizer to NH3(equation (A1a)) and equilibrationwith NH4+(equation (A1b)), which drove nitrification toNO2−(equation (A3a)), and NO3−(equation (A3b)) [Grant,1994], and reduction of NO2−to N2O (A3d) [Grant, 1995].The NH4+and NO3−products of hydrolysis and nitrificationwere also subject to competitive uptake by microbialpopulations (A2b and A2c) [Grant et al., 1993a, 1993b] andby root and mycorrhizal populations (equations (A6a)–(A6d)) [Grant, 1998]. Root and mycorrhizal uptake pro-ducts raised root and mycorrhizal nonstructural N con-centrations (equation (A7c)) and thereby transfers to shootnonstructural N (equations (A7a) and (A7b)), raising shootnonstructural N (equation (A7d)) and hence the N/C ratio ofleaf structural growth (equation (A7e)) that determined thefoliar N concentrations.5.2. Fertilization and Net CO2Exchange[24] Greater foliar N content in the model raised photo-synthetic capacity (equations (A7i) and (A7j)), rubiscoactivation (equation (A7k)), and hence CO2fixation rates(equations (A7f), (A7g), and (A7h)) [Grant et al., 2001] inthe fertilized stand. Greater CO2fixation rates caused morerapid CO2influxes to be modeled and measured during2007 than under comparable weather in 2001 (e.g., DOY163–169 in Figure 2b versus 2d), the year meteorologicallymost similar to 2007 within the instrumental period prior tofertilization as apparent from meteorological data summa-rized in Table 3. However, CO2effluxes modeled or gap‐filled during 2007 were comparable to those during 2001,causing net C uptake to rise with fertilization.[25] Rapid net C uptake modeled and measured in 2007(Figure 2b) allowed daily net ecosystem productivity (NEP)to rise with air temperature (Ta) and day length to maximumvalues of 5–6gCm−2d−1sustained during May and June(Figure 3b). Daily NEP subsequently declined with Taandday length after the summer solstice but remained positive(C sink) until mid‐October. Slower net C uptake modeledand measured in 2001 (Figure 2d) caused maximum dailyNEP of 4–5gCm−2d−1to be sustained only to mid‐May,after which NEP declined to near zero (C neutral) by mid‐September, about 1 month earlier than in 2007 (Figures 3dversus 3b). Lower maxima and earlier declines in dailyNEP were also modeled and measured in 2006, the yearTable 2. Changes in Stocks and Transfers of C and N BetweenFertilized and Unfertilized Stands of Douglas Fir Modeled byEcosys and Estimated From Biometric Measurements at BC‐DF49 on 6 December 2007 After Fertilization on 11 April 2007(N2O) or 13 January 2007 (Others)UnitFertilized UnfertilizedModeled Measured Modeled MeasuredFoliar Na(g N g C−1) 0.025 0.032 0.022 0.023Litterfall (g C m−2yr−1) 55 57 ± 24 86 62 ± 5bN2Oemissiona(g N m−2yr−1) 0.753 1.04 0.017 −0.014Foliar Ca(g C m−2) 833 0.280c745 0.223cWood C (g C m−2) 414 1.15 ± 0.17d216 0.78 ± 0.31daMeasured data from Jassal et al. [2008].bLitterfall in unfertilized plots from J. A. Trofymow (personalcommunication).cGrams C per 100 needles assuming dry matter is 50% C.dAnnual increment in bole diameter (mm) from Jassal et al. [2009].GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G040095of17before fertilization (Figure 4), as well as in other earlieryears at this site [Grant et al., 2009], indicating a consistentseasonal pattern in NEP before fertilization.[26] These earlier seasonal declines in NEP prior to fer-tilization in 2007 were modeled from earlier seasonal de-clines in foliar nonstructural N content (equation (A7d)) andhence in foliar structural N content (equation (A7e)) andrubisco activation (equation (A7k)) as described in section5.1. Seasonal declines in foliar nonstructural N contentoccurred when rates of nonstructural N assimilation drivenfrom biomass growth (equation (A7e)) exceeded rates ofnonstructural N transfer (equations (A7a) and (A7b)) fromroots driven from mineral N uptake by roots and mycor-rhizae (equations (A6a)–(A6d)). These modeled declineswere delayed in 2007 by additional mineral N uptake offertilizer products. Net C uptake in all 3 years was adverselyaffected by several brief warming events (Figures 3a and3c), as measured and modeled in other years at this site[Grant et al., 2007c, 2009].[27] To test consistency of model performance before andafter fertilization, hourly modeled CO2fluxes were re-gressed on hourly averaged EC CO2fluxes measured underconditions of adequate turbulence during each year from1998 to 2007 (i.e., excluding gap‐filled values). Parametersand correlation coefficients from this regression in 2007were very similar to those from 1998 to 2006 before fer-tilizer application (b = 1.0, R2= 0.8, RMSD = 3.4 mmol m−2s−1in Table 3), indicating that the model response to addedN did not introduce a bias into modeled CO2fluxes withrespect to measured values.Figure 2. (a, c) Incoming shortwave radiation and air temperature and (b, d) CO2fluxes measured (solidsymbols), gap filled (open symbols), and modeled (lines) during DOY 163–169 (12–18 June) in 2007after fertilization (Figures 2a and 2b) and in 2001 before fertilization (Figures 2c and 2d) over a Douglasfir stand fertilized on 13 January 2007. Positive fluxes represent CO2uptake; negative fluxes representCO2emission.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G040096of175.3. Fertilization and Net Primary Productivity[28] Greater CO2fixation rates modeled during 2007 withN fertilization (Figures 2 and 3) caused a 14% gain inannual gross primary productivity (GPP) from that modeledin 2007 without fertilization, consistent with a 10% gainestimated from EC measurements by Jassal et al. [2009](Table 4). However, fertilizer N had very different effectson Raand net primary productivity (NPP) of shoots versusroots. Shoot Ramodeled during 2007 rose 21% with fertil-ization, driven by greater Rgand Rmrequired for greaterbiomass growth and maintenance [Grant et al., 2007c,equations (C12)–(C16)]. Root Ramodeled during 2007 alsorose slightly with fertilization during the first 3 months afterapplication but decreased thereafter so that cumulative rootRaat the end of the year was slightly lower. Shoot NPP inthe model rose 43% with fertilization by the end of 2007,while root NPP declined by 33% (Table 4).[29] In the model, these differences in shoot versus rootresponses to fertilization arose from reduced below‐groundallocation of the nonstructural C product of GPP caused bymore rapid consumption of nonstructural C in shoot growthand Ra(equations (A7d) and (A7e)) with more rapid Nuptake from fertilization. This consumption lowered con-centration gradients of nonstructural C that drove transfer toroots (equation (A7a)), thereby implementing the functionalequilibrium by which shoot and root growth are thought tobe governed.5.4. Fertilization and Heterotrophic Respiration[30] Fertilization lowered heterotrophic respiration (Rh)by6% from that modeled without fertilization during 2007(Table 4). This reduction arose from reduced shoot‐root Ctransfer with fertilization which caused lower nonstructuralC concentrations in roots and mycorrhizae and hence lessexudation of nonstructural C to soil [Grant, 1993]. Fur-thermore, by raising GPP in the model, fertilization raisedproduction of nonstructural C, thereby reducing remobili-zation of foliar and root structural C needed to meet Rmrequirements and hence reducing shoot and root litterfalldriven by remobilization [Grant et al., 2007c, equations(C14)–(C18)]. However, measurements in the fertilizedversus unfertilized field plots at BC‐DF49 indicated nosignificant reduction in foliar litterfall with fertilization(Table 2). Less litterfall and exudates modeled with fertil-ization in 2007 provided less substrate for decomposition[Grant et al., 2007c, equation (A3)] and hence for Rhandmicrobial growth [Grant et al., 2007c, equations (A13) and(A28)].[31] Lower Rhpartially offset higher root Ramodeledduring the first three months after fertilizer application, sothat soil respiration (Rs= Rh+ below‐ground Ra) rose onlyslightly during this period. Thereafter lower Rhand lowerroot Rareduced Rsso that cumulative Rsmodeled at the endof 2007 was about 5% lower with fertilization (Table 4).Greater Rswas also measured with surface flux chambersduring the first 3–4 months after application in the fertilizedversus unfertilized field plots near the BC‐DF49 fluxtower, while similar Rswas measured in both treatmentsthereafter, so that cumulative Rsmeasured at the end of2007 was estimated to have been raised by about 6% withfertilization [Jassal et al., 2010] (Table 4). Both modeledand measured results thus indicated only small changes inRswith fertilization.5.5. Fertilization, Net Ecosystem Productivity, andGrowth[32] Greater gains in GPP versus ecosystem respiration(Re= Ra+ Rh) caused a 39% gain in net ecosystem pro-ductivity (NEP) to be modeled with fertilization, similar to a37% gain estimated from EC measurements by Jassal et al.[2009] (Table 4). Somewhat larger modeled versus mea-sured NEP may partly be attributed to subsurface leachingof dissolved inorganic C (DIC), which was assumed in themodel to have been lost from the site, but some of whichmay in fact have volatilized within the fetch area and sohave been detected at the EC tower.[33] By the end of 2007, fertilization caused a substantialrise in shoot C and a decline in root C in the model (Table 4),driven by a rise in shoot NPP, a decline in root NPP anddeclines in both shoot and root litterfall. However, fertil-ization reduced gains in soil organic C (SOC) during the firstyear after application because it reduced litterfall more thanit did Rs.Table 3. Mean Annual Temperatures, Annual Precipitation, Intercepts, Slopes, Coefficients of Determination, Root Mean Square of Dif-ferences, and Number of Accepted Eddy Covariance fluxes From Regressions of Hourly Modeled CO2Fluxes Versus Hourly AveragedEC CO2Fluxes Measured at BC‐DF49 From 1998 to 2007aYear MAT (°C) Precip (mm) a (mmol m−2s−1) bR2RSMD(mmol m−2s−1) n1998 9.10 1844 0.24 1.10 0.75 3.3 51871999 7.64 1913 0.40 1.09 0.77 3.1 49822000 8.21 1127 0.43 1.02 0.76 3.5 49542001 8.09 1166 0.36 1.01 0.77 3.3 49902002 8.48 1222 0.54 1.07 0.75 3.4 50262003 8.48 1621 0.33 1.04 0.77 3.4 52772004 8.77 1404 0.63 0.99 0.75 3.8 43622005 8.31 1467 0.91 0.96 0.77 3.8 43882006 8.40 1809 0.13 1.02 0.76 3.4 52332007b7.71 1482 0.40 1.04 0.80 3.4 5235aMAT, mean annual temperatures; Precip, annual precipitation; a, intercepts; b, slopes; R2, coefficients of determination; RMSD, root mean square ofdifferences; EC, number of accepted eddy covariance fluxes.bUrea applied at 20 g N m−2on 13 January 2007.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G040097of175.6. Fertilization and N2O Emissions[34] Emissions of N2O modeled in the fertilized stand andmeasured in the fertilized field plots were temporally vari-able but continuous during most of 2007, while emissionsmodeled in the unfertilized stand and measured in theunfertilized field plots were negligible (Figure 5). Totalemissions of N2O attributed to the fertilizer from applicationon 11 April 2007 to the end of the year were 0.74 g N m−2from differences between fertilized and unfertilized stands inthe model and 1.05 g N m−2derived from differences inmeasurements of surface fluxes and subsurface concentra-tion gradients in the fertilized versus unfertilized field plotsby Jassal et al. [2008] (Table 2). These emissions accountedfor 3.7% and 5.2% of the added fertilizer N in the model andthe field plots, respectively.[35] In the model, emissions were driven by urea hydro-lysis (equations (A1a) and (A1b)) which accelerated NH4+oxidation to NO2−(equation (A3a)) and NO3−(equation(A3b)) [Grant, 1994], and hence NO3−reduction to NO2−(equation (A4a)) and NO2−reductiontoN2O (equations(A3d) and (A4b)) [Grant, 1995; Grant et al., 2006b].These reductions were driven by demand from oxidationreactions for electron acceptors unmet by O2. However, thisdemand was strongly limited by aerated conditions in thesurface residue to which the urea was added. Rises in NO2−concentrations from accelerated NH4+oxidation caused risesFigure 3. (a, c) Air temperature (line), precipitation (bars), and (b, d) daily net ecosystem productivity(NEP) calculated from gap‐filled EC measurements (symbols) or modeled (line) during 2007 after fertil-ization (Figures 3a and 3b) and during 2001 before fertilization (Figures 3c and 3d) over a Douglas firstand fertilized on 13 January 2007. Positive NEP represents net CO2uptake; negative NEP representsnet CO2emission. Solid or open symbols represent daily EC values derived from gap‐filling less or morethan 24 half‐hourly values, respectively.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G040098of17in HNO2in the model (equation (A5a)) under the low pH ofDouglas fir residue (4.75 in the work of Welke and Hope[2005]). HNO2drove NO2−reduction through chemodeni-trification (equation (A5b)), a fraction of which generatedN2O (equation (A5c)). Large HNO2concentrations alsoslowed NO2−oxidation (equation (A3c)), sustaining theHNO2concentrations driving chemodenitrification.5.7. Fertilization and Changes in Ecosystem C[36] Increased GPP from fertilization caused aggregatedgap‐filled EC fluxes to rise by about 150 g C m−2yr−1during the first year after application, similar to the modeledrise (Figure 6a). In the model, NEP of the fertilized standwas predicted to decline after 2007, rapidly for the first fewyears and then more slowly thereafter. NEP of the unfer-tilized stand was also predicted to decline with forest age[Grant et al., 2007c] but more slowly than that of the fer-tilized stand, so that the two values of NEP gradually con-verged. The larger modeled versus EC‐derived values maybe partly attributed to the accounting of DIC losses asdescribed earlier (Table 4).[37] The modeled gains in NEP from fertilization werefound mostly in wood C and partly in foliage C (Figure 6b).By 6 December 2007, about 11 months after fertilization,foliar C in the model rose 12% from 745 g C m−2in theunfertilized stand to 833 g C m−2in the fertilized stand,comparable to a rise in 100‐needle C of 23% from 0.223 to0.280 g measured on this date in the unfertilized versusfertilized field plots by Jassal et al. [2008] (Table 2). By thissame date wood C growth in the model rose 92% from 216to 414 g C m−2yr−1, which was greater than a rise of 47% intree ring growth measured in the fertilized versus unfertil-ized plots in 2007 (Table 2). Gains in wood C growthdeclined during the years after fertilization, so that cumu-lative gains in wood C rose more slowly with time, whilegains in foliage C were reversed after several years andeventually lost (Figure 6b).[38] Changes in tree C allocation and hence in shootversus root NPP with fertilization caused modeled gains inwood and foliage C to be partly offset by losses in soil, rootand nonstructural C during the first 3–5 years after fertil-ization (Figure 6c). These losses were caused by more rapiduse of nonstructural C in foliage and wood growth(equations (A7d) and (A7e)) which reduced transfers ofnonstructural C to roots and mycorrhizae (equation (A7a))as described earlier. Declines in soil C modeled more than2 years after fertilization were partly offset by rises in surfacelitter C (Figure 6c) caused by more rapid aboveground lit-terfall from larger foliar and wood phytomass. Declines inbelow‐ground C were eventually reversed, so that soil, root,foliage and nonstructural C gradually returned to valuessimilar to those modeled in the unfertilized stand by 11 yearsafter fertilization.5.8. Fertilization and Changes in Ecosystem N[39] Changes in ecosystem C stocks modeled after fertil-ization were associated with those in ecosystem N stocks. Atthe end of 2007, the added fertilizer N in the model wasfound mostly in nonstructural and foliage N through rootuptake (equations (A6a)–(A6d)) and root‐shoot transfer(equation (A7b)) (Figure 6d). These gains were partly offsetby a decline in root N (Figure 6e) associated with that inroot C (Figure 6c). Some of the added fertilizer was retainedin soil and surface litter (Figure 6e) through litterfall [Grantet al., 2007c, equation (C18)], mineralization (equation(A2a)), and immobilization (equations (A2b) and (A2c)),or lost to groundwater mostly as NO3−through nitrification(equations (A3a) and (A3b)) and leaching (Figure 6e), oremittedtotheatmosphereasN2OandN2(Figure6e)fromnitrifi-cation (equation (A3d)), denitrification (equation (A4b)), andFigure 4. (a) Air temperature (lines), precipitation (bars), and (b) daily net ecosystem productivity(NEP) calculated from gap‐filled EC measurements (symbols) or modeled (line) during 2006 before fer-tilization over a Douglas fir stand fertilized on 13 January 2007. Positive NEP represents net CO2uptake;negative NEP represents net CO2emission. Solid or open symbols represent daily EC values derived fromgap‐filling less or more than 24 half‐hourly values, respectively.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G040099of17chemodenitrification (equation (A5c)) (Figure 5). After 2007,there was a gradual translocation of added N out of non-structural stocks into foliage stocks and after 2010 out ofnonstructural and foliage stocks into soil, wood, and rootstocks (Figure 6d versus 6e). Losses of added N throughleaching and gaseous emissions in the model were negligibleafter 2007 (Figure 6e), as was found from measurements ofemissions at BC‐DF49 in 2008 [Jassal et al., 2010].6. Discussion6.1. Fertilizer N and Aboveground Productivity[40] Productivity at the BC‐DF49 site was likely limitedby N, as evidenced by foliar N concentrations of only0.022 g N g C−1modeled in the unfertilized stand and0.023 g N g C−1measured in the unfertilized field plots(Table 2). These concentrations were much lower than 0.029gNgC−1below which growth responses to N fertilization inDouglas fir were found by Hopmans and Chappell [1994].The N concentration of the soil LFH layer at BC‐DF49Table 4. Annual Carbon Fluxes and Changes in Key C StocksFrom Fertilized and Unfertilized Stands of Douglas Fir ModeledBy Ecosys and Estimated From Eddy Covariance Measurementsat BC‐DF49 in 2007Fertilized UnfertilizedModeled Measured Modeled MeasuredFluxesa(g C m−2yr−1)GPP 2459 1944 2161 1760bRaShoot 1012 838Rootc439 455Total 1451 1293NPPShoot 801 560Rootc207 308Total 1008 868LitterFoliage 55 86Other AGd211 190Rootc,e264 318Total 530 594Rh417 443Rs856 921 897 862bRe1868 1397 1736 1362bNEP (atm) 591 547 425 398bDIC, DOCf53 57NEP (total) 538 368Stock Changes (g C m−2)DABCcLiving 476 225Dead 59 59DBGChRoota−83 −7Storage 25 −1DSOC 58 89DSoil CO233Total 538 368aAbbreviations: Ra, autotrophic respiration; Rh, heterotrophic respiration;Rs, ecosystem respiration; DIC, dissolved inorganic carbon; DOC,dissolved organic carbon; ABC, aboveground plant carbon; BGC, below‐ground plant carbon; SOC, soil organic carbon.bFrom an empirical model fitting measured C fluxes to climate variablesfrom 1998 to 2006 and used to estimate C fluxes from climate variables in2007 by Jassal et al. [2009, 2010].cAll root terms include mycorrhizae.dIncludes twig, reproductive material, coarse woody litter.eIncludes litterfall plus exudation.fDissolved inorganic and organic C (mostly aqueous CO2, carbonates)exported in runoff and drainage water.gAboveground C, including foliage, twigs, boles, reproductive material,and nonstructural C.hBelow‐ground C, including coarse and fine roots, mycorrhizae, andnonstructural C.Figure 5. N2O emissions measured (symbols) and modeled(lines) during 2007 from Douglas fir stands that were notfertilized or fertilized with 20 g N m−2as urea on 11 April2007 (DOY 101).Figure 6. (a) Net ecosystem productivity (NEP) and cumu-lative changes in (b) aboveground and (c) below‐ground Cstocks, and (d) aboveground and (e) below‐ground N stocksand fluxes modeled (lines) and measured (symbols in Figure6a) over 11 years in a Douglas fir stand fertilized on 11 Jan-uary 2007 versus an unfertilized Douglas fir stand.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G0400910 of 17(0.021g N gC−1from Table 1) was alsomuch lower than oneof 0.036 g N g C−1below which growth of Douglas fir wasfound by Edmonds and Hsiang [1987] to become N limited.These foliar and soil N contents indicated that a strongresponse of CO2uptake to N fertilizer should be expected atBC‐DF49, as modeled and measured in Figures 2 and 3.[41] The modeled response of CO2uptake to N fertilizergenerated a gain in modeled wood C of 752 g C m−2by8 years after the 20 g N m−2urea application (Figure 6b),which was consistent with gains measured under similar soiland climates in earlier field studies. This gain was largerthan an average gain of ∼550 g C m−2calculated fromvolume growth measured in field studies 8 years afterapplying 22.4 g N m−2urea to young (<30 years) Douglasfir stands in the nearby states of Oregon and Washington[Hopmans and Chappell, 1994; Stegemoeller and Chappell,1990]. However, most of the stands in these studies hadfoliar N contents before fertilization that were larger thanthat at BC‐DF49, so that a smaller response of growth tofertilizer might be expected. The time course of the modeledgain in wood C attributed to the fertilizer application in thisstudy was similar to, or smaller than, ones averaging2430 g C m−2over 16 years measured by Adams et al.[2005] after four applications of 22.4 g N m−2on soils inWashington State similar to that at BC‐DF49.6.2. Fertilizer N and Below‐Ground Productivity[42] Gains in foliar and wood C with fertilization in themodel were partly offset for several years by declines in rootand soil C (Table 4) which were in turn only partly offset bya rise in surface litter C (Figure 6c). These declines werecaused by reductions in below‐ground C allocation andhence in NPP and litterfall of roots and mycorrhizae asdescribed earlier, so that fertilization caused net reductionsin modeled root and microbial growth and hence in Ra, Rh,and Rs(Table 4). At this stage of model development, directadverse effects of high soil N on Rhare not simulatedbecause underlying mechanisms are not well understood[Berg and McClaugherty, 2003]. Therefore reductions in Rhwith fertilization may have been underestimated whenmodeled only from reduced root C inputs.[43] Reductions in Rswith fertilization in the model (5%in Table 4) were consistent with, but usually smaller than,reductions of Rsmeasured after fertilization of tropical (18%in the work of Giardina et al. [2004]), temperate (17% inthe work of Burton et al. [2004]; ∼25% in the work ofHaynes and Gower [1995]), and boreal (40% in the work ofOlsson et al. [2005]) forests. In some studies, Rshas beenfound to rise slightly with fertilization for a brief period afterapplication and to decline thereafter [Burton et al., 2004], aswas modeled in this study. As in this modeling study, de-clines in Rsmeasured in these experiments were attributed toreductions in below‐ground C allocation and hence in rootand mycorrhizal growth and litterfall [Haynes and Gower,1995] and in microbial growth [Burton et al., 2004],which were only partly offset by increases in abovegroundlitterfall [Giardina et al., 2004]. Some studies have indi-cated gains in soil C following fertilization, although withlimited certainty [e.g., Adams et al., 2005]. These resultsindicate that gains in C sequestration attributed to fertiliza-tion must be estimated from total changes in aboveground,surface, and below‐ground C stocks, rather than fromchanges in wood C alone.6.3. Retention of Fertilizer N[44] The rapid early gain in NEP from fertilizer N in themodel (Figure 6a) was driven by rapid uptake of fertilizer N,much of which appeared as nonstructural and foliar N intrees within the first year after application (Figure 6d). Thismodel result is consistent with those from many field studiesshowing that fertilizer N in forests is largely taken up duringthe first growing season after application [Mead et al.,2008]. Six months after application, the fertilized stand inthe model gained 9.5 g N m−2in aboveground phytomass(foliage, wood, and nonstructural) and 7.4 g N m−2in soil(mostly as nitrate) above values in the unfertilized stand(Figures 6d and 6e). These gains in the model were slightlylarger than ones of 5.8 and 6.0 g N m−2in trees and soil,respectively, measured by Flint et al. [2008] 6 months aftera spring application of 22.4 g N m−2as urea on a similarDouglas fir stand in nearby Washington State. However, thegains measured in their study may have been underestimatedbecause the gain in tree N did not account for gain in treebiomass and the gain in soil N was measured only to 0.4 mdepth and excluded the coarse organic fraction. These earlygains in N measured in a field study conducted under con-ditions comparable to those at BC‐DF49 corroborate therapid uptake of fertilizer N in the model. The large nitrateaccumulation modeled during 2007 (Figure 6e) arose fromrapid nitrification of the added fertilizer, consistent withrapid nitrification measured in fertilized Douglas fir soils[Matson et al., 1992].[45] After several years, N gains modeled in the fertilizedtrees were distributed away from nonstructural and foliarstocks into root, wood, and soil stocks (Figures 6d and 6e).The time course of the initial rise and subsequent decline ofmodeled gains in foliar N was consistent with that of foliarfertilizer‐derived N measured in field experiments withDouglasfir[e.g.,Meadet al.,2008].After10years,fertilizedtrees in the model gained 3.2 and 6.3 g N m−2in foliar andnonfoliar (wood, nonstructural) phytomass, and the fertilizedsoil gained6.0gNm−2asorganic matter, abovevalues intheunfertilized stand.These gains in the model were comparableto ones of 3.8 and 7.0 g N m−2in foliar and nonfoliar (wood)phytomass and 8.0 g N m−2in soil measured by Mead et al.[2008] 10 tears after applying 20 g N m−2as urea to nearbyDouglas fir stands of a similar age to that at BD‐DF49. Thelong‐term gain in soil + litter N with respect to soil + litter Cin the model (Figures 6e versus 6c) caused gradual declinesin soil + litter C/N ratios following fertilization as has beenmeasured in Douglas fir stands by Prietzel et al. [2004].These declines indicated long‐term retention of fertilizer Nin soils with large initial C/N ratios, enabling long‐termincreases in net N mineralization [Prietzel et al., 2004] thatsustained long‐term gains in ecosystem productivity(Figures 6a and 6b) [Adams et al., 2005].6.4. Losses of Fertilizer N[46] Some of the fertilizer N was not retained by the foreststand, but was lost from leaching and gaseous emissions. Ofthe 20 g N m−2fertilizer N added in the model on 13 January,2.1 g N m−2was leached and 0.8 g N m−2emitted to theatmosphere as N2O and N2during the first year after appli-GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G0400911 of 17cation (Figure 5c). However, no fertilizer N was emittedthereafter, as was also found from N2O flux measurementstaken during 2008 at the BC‐DF49 site [Jassal et al., 2010].Most of the leaching loss modeled in 2007 occurred underheavier rainfall after mid‐October 2007 when plant growthand hence uptake had slowed (Figure 3). This loss was largerthan one of 0.4 g N m−2measured by Flint et al. [2008], mostof which also occurred after mid‐October, following a springapplication of 22.4 g N m−2as urea to a Douglas fir stand innearby Washington State under climate conditions similar tothose at BC‐DF49. Substantial downward water movementoccurs through the soil profile with heavier rainfall andcooler weather after mid‐October in the Pacific Northwest.These leaching losses would thus represent most of theresidual mineral N not taken up during the previous summergrowing season, indicating that most fertilizer N was takenup soon after application at both sites, but perhaps more so inthe study of Flint et al. [2008].[47] The fraction of fertilizer N emitted as N2O, oremission coefficient EF, during the first year after the ureaapplication on 11 April in the model (3.74%) and in the fieldstudy (5.2%) (Figure 5) were larger than ones measured inother studies of fertilizer application in forests. Sitaula et al.[1995b] calculated an annual EF of 1.7% in boreal pine byextrapolating 1 month of measurements. Matson et al.[1992] calculated an annual EF of 0.35% in Douglas firfrom a few measurements during one growing season. Inboth these experiments, N was applied as ammoniumnitrate, only one half of which would contribute to N2Oemissions through nitrification. However, Brumme andBeese [1992] measured a similar annual EF of 1.6% in anacidic beech soil from five measurements per day withautomated chambers following application of ammoniumsulphate. The EFs modeled and measured in this study werealso larger than those commonly measured or modeled inarable soils [e.g., Grant et al., 2006b].[48] The large N2O emissions modeled after fertilizationin this study were attributed mostly to chemodenitrificationof NO2−from nitrification (70%), rather than to biologicalreduction of NO2−from nitrification and denitrification(30%). These emissions were driven by the low pH ofDouglas fir litter (4.75) to which the fertilizer was applied(equation (A5a)). However, the parameterization of che-modenitrification in the model, while based on literaturevalues derived from laboratory incubations (equations (A5b)and (A5c)), needs further testing in field studies. Nonethe-less the lower pH of forest soils is known to favor pro-duction of N2O over NO3−during nitrification [Martikainenand de Boer, 1993; Sitaula et al., 1995a]. Consequentlyraising forest soil pH from 4.5 to 6.5 by liming has beenfound to reduce N2O emissions substantially [Brumme andBeese, 1992]. N2O emissions from coniferous forest soilsmay therefore be more sensitive to N additions from fertil-izer or deposition than are those from arable soils withgenerally higher pH [Sitaula et al., 1995b]. The possibilityof larger N2O EFs for fertilizer applications on forest soilswith lower pH needs to be examined in further studies,given the rising use of fertilizer in forest production.6.5. Net CO2‐Equivalent Exchange From Fertilizer N[49] Fertilization of mature Douglas fir stands wouldtypically occur 10–12 years before harvesting (although itmay be done earlier) so that net GHG exchange attributed tofertilizer application might be assessed over this period.After 11 years, a total C sequestration of 1045 g C m−2inthe model was attributed to the 20 g N m−2of fertilizerapplied on 13 January 2007 (Figures 6b and 6c). Of thisamount, only 48 g C m−2was sequestered in below‐groundpools (root, litter, soil) that would remain on site after har-vest (Figure 6c). An additional 71 g C m−2was sequesteredas foliage, some of which could be lost at harvest,depending on logging practices. Most of these pools wouldbe respired after harvest, and only a fraction, depending onclay content, would be stabilized in long‐term humus pools.The remaining 926 g C m−2in the model was sequestered aswood, the duration of which would depend on uses of thewood product.[50] This sequestration had a direct cost of 0.60 g N m−2or 78 g C m−2equivalent from N2O emission and indirectcosts of 17 g C m−2from fertilizer transport and application[Jassal et al., 2008] and a further 2.7 g C m−2equivalentthrough N2O emission from fertilizer N leached to ground orsurface water. These costs offset C sequestration in allbelow‐ground pools and most of that in foliage for the 11years after application, so that any net gain in sequestrationfrom this fertilizer application would depend entirely on theaccounting of C from harvested wood products.Appendix AA1. Urea Hydrolysis[51]COðNH2Þ2lþ H2O ! CO2lþ 2NH3lðA1aÞ½NHþ4lC138$½NH3slC138þ½HþlC138 pKa¼ 9:25 ðA1bÞA2. Mineralization‐Immobilization[52]UNH4i;l;m;j¼ðMCi;l;m;jCNjC0 MNi;l;m;jÞðmineralizationÞ UNH4< 0ðA2aÞUNH4i;l;m;j¼ min MCi;l;m;jCNjC0 MNi;l;m;jC16C17n;U0NH4Ai;l;m;jNHþ4i;l;m;jhiC0 NHþ4mnC2C3C16C17.NHþ4i;l;m;jhiC0 NHþ4mnC2C3þ KNH4C16C17oUNH4> 0 ðA2bÞUNO3i;l;m;j¼ minnMCi;l;m;jCNjC0 MNi;l;m;jþ UNH4i;l;m;jC0C1C0C1;U0NO3Ai;l;m;jNOC03i;l;m;jhiC0 NOC03mnC2C3C16C17.NOC03i;l;m;jhiC0 NOC03mnC2C3þ KNO3C16C17oUNO3> 0 ðA2cÞGRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G0400912 of 17A3. Nitrification[53]XNH3i;l;n¼ X0NH3MCi;l;n;aNH3sl½C138= NH3sl½C138þKNH3ðÞfgC1 CO2sl½C138= CO2sl½C138þKCO2ðÞfgftslfO2lðA3aÞXNO2i;l;o¼ X0NO2MCi;l;o;aNOC02lhi.NOC02lhiþ KNO2oC16C17noC1 CO2sl½C138= CO2sl½C138þKCO2ðÞfgftslfO2lfilðA3bÞfil¼ 1:0= 1:0 þ NH3sl½C138=KiNH3ðÞ1:0 þ HNO2l½C138=KiHNO2ðÞðA3cÞRNO2i;l;n¼ R0NO2i;l;nNOC02lhi.NOC02lhiþ KNO2nC16C17noC1 CO2sl½C138= CO2sl½C138þKCO2ðÞfgðA3dÞA4. Biological Denitrification[54]RNO3i;l;d¼ R0NO3i;l;dNOC03lhi.NOC03lhiþ KNO3C16C17ðA4aÞRNO2i;l;d¼ R0NO3i;l;dC0 RNO3i;l;dC16C17NOC02C2C3.NOC02C2C3þ KNO2dC0C1ðA4bÞRN2Oi;l;d¼ 2 R0NO3i;l;dC0 RNO3i;l;dC0 RNO2i;l;dC16C17C1 N2Ol½C138= N2Ol½C138þKN2OðÞ ðA4cÞA5. Chemodenitrification[55]HNO2l½C138$NOC02lhiþ HþlC2C3pKa¼ 3:3 ðA5aÞQNO2l¼ Q0NO2HNO2l½C138C18lftslðA5bÞQN2Ol¼ fN2OQNO2lðA5cÞA6. Root and Mycorrhizal Uptake[56]UNH4s;l;r¼ Uws;l;rNHþ4lC2C3þ 2C25Ls;l;rDeNH4NHþ4lC2C3C0 NHþ4s;l;rhiC16C17n.ln ds;l;r=rs;l;rC0C1oftslðA6aÞ¼ U0NH4As;l;rNHþ4s;l;rhiC0 NHþ4mnC2C3C16C17.NHþ4s;l;rhiC0 NHþ4mnC2C3þ KNH4C16C17ftslðA6bÞUNO3s;l;r¼ Uws;l;rNOC03lC2C3þ 2C25Ls;l;rDeNO3NOC03lC2C3C0 NOC03s;l;rhiC16C17n.ln ds;l;r=rs;l;rC0C1oftslðA6cÞ¼ U0NO3As;l;rNOC03s;l;rhiC0 NOC03mnC2C3C16C17.NOC03s;l;rhiC0 NOC03mnC2C3þ KNO3C16C17ftslðA6dÞA7. Plant Assimilation[57]C27Cs;l;r=MCs;l;rþ TCs;l;r¼ C27Cs;b=MCs;bC0 TCs;l;rðA7aÞC27Ns;l;r=C27Cs;l;rC0 TNs;l;r¼ C27Ns;b=C27Cs;bþ TNs;l;rðA7bÞC27Ns;l;rðtÞ¼ C27Ns;l;rðtC01Þþ UNH4s;l;rþ UNO3s;l;rC0 TNs;l;rC0 GNs;l;rðA7cÞC27Ns;bðtÞ¼ C27Ns;bðtC01ÞþXlTNs;l;rC0 GNs;bðA7dÞGNs;l;r¼ C14MNs;b=C14MCs;b¼ C14MCs;b=C14tf min C27Ns;b.C27Ns;bþ C27Cs;b=KiC27NC0C1;nC16C1 C27Ps;b.C27Ps;bþ C27Cs;b=KiC27PC0C1gÞ ðA7eÞGPPs¼XbXlXkXxXyXzmin Vbs;b;l;k;x;y;z;Vjs;b;l;k;x;y;znoðA7fÞVbs;b;l;k;x;y;z¼ Vbmaxs;b;kðCcs;b;l;k;x;y;zC0Gs;b;kÞ=ðCcs;b;l;k;x;y;zþ KCO2sÞfysftbsfNs;bðA7gÞVjs;b;l;k;x;y;z¼ "Is;l;x;y;zþ Vjmaxs;b;kC0 "Ii;l;x;y;zþ Vjmaxs;b;kC16C172C18C18C0 4C11"Is;l;x;y;zVjmaxs;b;kC190:5C19,ð2C11ÞYs;b;l;k;x;y;zfysftjsfNs;bðA7hÞVbmaxs;b;k¼ V0bsNrubiscosMNs;b;k=As;b;kðA7iÞVjmaxs;b;k¼ V0jsNchlorophyllsMNs;b;k=As;b;kðA7jÞfNs;b¼ min C27Ns;b.C27Ns;bþ C27Cs;b=KiC27NC0C1;C27Ps;b.C27Ps;bþ C27Cs;b=KiC27PC0C1noðA7kÞNotationA microbial or root surface area, equal to 4pr2N(mic.) and pr2L (roots) (equations (A2a)–(A2c),(A6a)–(A6d), and (A7a)–(A7k)) (m2m−2).GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G0400913 of 17a shape parameter for response of Vjto I, equalto 0.7.Ccchloroplast [CO2] in equilibrium with Ci(equation (A7g)) (mM).[CO2S]CO2concentration in soil solution (equations(A3a), (A3b), and (A3d)) (g C m−3).CNjmaximum ratio of N to C in M, equal to 0.22and 0.13 for labile and resistant j (equations(A2a), (A2b), and (A2c)) (g N g C−1).DeNH4effective dispersivity‐diffusivity of NH4+dur-ing root uptake (equation (A6a)) (m2h−1).ds,zhalf distance between adjacent roots assumedequal to uptake path length, equal to (pLs,l,r/Dz)−1/2(equations (A6a) and (A6c)) (m)" quantum yield, equal to 0.40 (equation (A7h))(mmol e−mmol quanta−1).fifunction for inhibition of XNO2by NH3andHNO2(equations (A3b) and (A3c)).fNnutrient inhibition of Vb, (equations (A7g),(A7h), and (A7k))fN2Ofraction of QNO2released as QN2O, equal to 0.1[Mørkved et al., 2007] (equation (A5c)).fO2ratio of O2supply to demand at ambient [O2](equations (A3a) and (A3b)).fynonstomatal water effect on Vb, Vj[Grantand Flanagan, 2007] (equations (A7g) and(A7h)).ftsArrhenius temperature function for soil pro-cesses (equations (A3a)–(A3d), (A5a)–(A5c),and (A6a)–(A6d)).ftbArrhenius temperature function for Vb[Bernacchi et al., 2003] (equation (A7g)).ftjArrhenius temperature function for Vj[Bernacchi et al., 2003] (equation (A7h)).GNsNconsumed in growth (equations (A7c),(A7d), and (A7e)) (g m−2h−1).G CO2compensation point (equation (A7g))(mM).[H+] hydrogen ion concentration (equations (A1c)and (A5a)) (mol L−1).I irradiance (equation (A7h)) (mmol m−2s−1).KCO2Michaelis‐Menten constant for CO2reduction,equal to 0.15 (equations (A3a), (A3b), and(A3d)) (g C m−3).KCO2Michaelis‐Menten constant for CO2fixation(equation (A7g)) (mM).KiHNO2[HNO2] at 1/2 inhibition of XNO2, equal to25.0 [Blackburne et al., 2007] (equation(A3c) (g N m−3).KiNH3[NH3] at 1/2 inhibition of XNO2, equal to10.0 [Blackburne et al., 2007] (equation (A3c)(g N m−3).KisNinhibition constant for sCi,jversus sNi,j, equalto 100 [Grant, 1998] (equations (A7e) and(A7k)) (g C g N−1).KisPinhibition constant for sCi,jversus sPi,j,equalto 1000 [Grant, 1998] (equations (A7e) and(A7k)) (g C g P−1).KNH3Michaelis‐Menten constant for XNH3, equal to2.0 × 10−4[Stark and Firestone, 1996](equation (A3a)) (g N m−3).KNH4Michaelis–Menten constant for INH4or UNH4,equal to 0.40 [Barber and Silberbush, 1984](equations (A2b) and (A6b)) (g N m−3)KNO3Michaelis–Menten constant for INO3or UNO3,equal to 0.35 [Barber and Silberbush, 1984](equations (A2c) and (A6c)) (g N m−3)KNO2dMichaelis‐Menten constant for RNO2by deni-trifiers, equal to 2.5 [Yoshinari et al., 1977](equation (A4b)) (g N m−3).KNO2nMichaelis‐Menten constant for RNO2by nitrifi-ers, equal to 2.5(equation (A3d)) (g N m−3).KNO2oMichaelis‐Menten constant for XNO2,equalto0.7 [Blackburne et al., 2007] (equation (A3b))(g N m−3).KNO3Michaelis‐Menten constant for RNO3, equal to2.5 [Yoshinari et al., 1977] (equation (A4a))(g N m−3).KN2OMichaelis‐Menten for RN2O, equal to 2.5[Yoshinarietal.,1977](equation(A4c))(gNm−3).kTrate constant for TN(equation (A7b)) (h−1).L rootlength(equations(A6a)and(A6c))(mm−2).MCmicrobial, root or shoot biomass C (equations(A2a)–(A2c), (A1a), (A1b), (A3), (A7a), and(A7e)) (g C m−2).MNmicrobial, root or shoot biomass N (equations(A2a)–(A2c), (A1a)–(A1b), (A3a)–(A3d), and(A7e)) (g N m−2).N microbial number (m−2).Nchlorophyllfraction of leaf N in chlorophyll, equal to0.025 (equation (A7j)).Nrubiscofraction of leaf N in rubisco, equal to 0.125(equation (A7i)).[NH4+mn] [NH4+] below which UNH4= 0, equal to 0.0125[Barber and Silberbush, 1984] (equations(A2b) and (A6b)) (g N m−3).[NO3−mn][NO3−]belowwhichUNO3= 0, equal to 0.03[Barber and Silberbush, 1984] (equations(A2c) and (A6c)) (g N m−3).QNO2NO2−reductionbychemodenitrification(equations (A5b) and (A5c)) (g N m−2h−1).Q′NO2rate constant for decomposition of HNO2at25°C, equal to 0.075 [Cleemput and Samater,1996] (equation (A5b)) (h−1).QN2ON2O production by chemodenitrification(equation (A5c)) (g N m−2h−1).C18 soil water content (equation (A5b)) (m3m−3).RNO2NO2−reduction under ambient [NO2−](equations (A3d) and (A4b)) (g N m−2h−1).R′NO2RNO2under nonlimiting [NO2−] (equation(A3d)) (g N m−2h−1).RNO3NO3−reduction under ambient [NO3−](equations(A4a),(A4b),and(A4c))(gNm−2h−1).R′NO3RNO3under nonlimiting [NO3−] (equations(A4a), (A4b), and (A4c)) (g N m−2h−1).RN2ON2O reduction under ambient [N2O] (equation(A4c)) (g N m−2h−1).r radius of root or mycorrhizae, equal to 1.0 ×10−3,5.0×10−6(equations(A6a)and(A6c))(m).sCnonstructural C (equations (A7a), (A7b),(A7e), and (A7k)) (g C m−2).sNnonstructural N (equations (A7b), (A7c),(A7d), (A7e), and (A7k)) (g N m−2).GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G0400914 of 17TCtransfer of sCbetween shoot and root(equation (A7a)) (g N m−2h−1).TNtransfer of sNbetween root and shoot(equations (A7b), (A7c), and (A7d)) (g Nm−2h−1).UNH4NH4+uptake by microbial, root or mycorrhizalsurfaces(equations(A2a)–(A2c),(A6a)–(A6d),and (A7c)) (g N m−2h−1).U′NH4maximum UNH4at 25°C and nonlimiting NH4+,equal to 5.0 × 10−3[Barber and Silberbush,1984](equations(A2b)and(A6b))(gNm−2h−1).UNO3NO3−uptake by microbial, root or mycorrhizalsurfaces(equations(A2a)–(A2c),(A6a)–(A6d),and (A7c)) (g N m−2h−1).U′NO3maximum UNO3at 25°C and nonlimiting NO3−,equal to 5.0 × 10−3[Barber and Silberbush,1984](equations(A2c)and(A6c))(gNm−2h−1).Uwwater uptake by roots or mycorrhizae(equations (A6a) and (A6c)) (m3m−2h−1).VbCO2‐limited leaf carboxylation (equation(A7g)) (mmol m−2s−1).Vb′ specific rubisco carboxylation at 25°C, equalto 45 [Farquhar et al., 1980] (equation(A7i)) (mmol g−1rubisco s−1).VbmaxVbat nonlimiting CO2, y;ci, Tcand N,P(equations (A7g) and (A7i)) (mmol m−2s−1).Vjirradiance‐limited carboxylation (equation(A7h)) (mmol m−2s−1).Vj′ specific chlorophyll e−transfer at 25°C, equalto 450 [Farquhar et al., 1980] (equation(A7j)) (mmol g−1chlorophyll s−1).VjmaxVjat nonlimiting CO2, yci, Tcand N,P(equation (A7h)) (mmol m−2s−1).X′NH3specific NH3oxidation at 25°C and nonlimit-ing [O2], equal to 0.625 [Belser and Schmidt,1980] (equation (A3a)) (g N g−1Mh−1).XNH3NH3oxidation under ambient Tsand [O2](equation (A3a)) (g N m−2h−1).X′NO2specific NO2−oxidation under at 25°C and non-limiting [O2s], equal to 0.75 [Blackburne et al.,2007] (equation (A3b)) (g C m−2h−1).XNO2NO2−oxidation under ambient temperature and[O2s] (equation (A3b)) (g C m−2h−1).Subscripta active component of Mi,mb branch of plant species s.d heterotrophic denitrifier population (subsetof h).h heterotrophic community (subset of m).i substrate‐microbe complex.j kinetic components of M.k node of branch b.l soil, surface litter or canopy layer.m all microbial communities.n autotrophic ammonia oxidizer population (sub-set of m).o autotrophic nitrite oxidizer population (subsetof m).r root or mycorrhizae population.s plant species.x leaf azimuth.y leaf inclination.z leaf exposure (sunlit or shaded).[58] Acknowledgments. The Canada Carbon Project is funded bythe Canadian Foundation for Climate and Atmospheric Sciences (CFCAS).Computational facilities for ecosys were provided by the Westgrid high‐performance computing infrastructure. We greatly appreciate the litterfalldata provided by A. J. (Tony) Trofymow, Pacific Forest Center CFS,Victoria.ReferencesAber, J. D., K. J. Nadelhoffer, P. Steudler, and J. M. Mellilo (1989), Nitro-gen saturation of northern forest ecosystems, BioScience, 39,378–386,doi:10.2307/1311067.Adams, A. B., R. B. Harrison, R. S. Sletten, B. D. Strahm, E. C. Turnblom,and C. M. Jensen (2005), Nitrogen‐fertilization impacts on carbonsequestration and flux in managed coastal Douglas fir stands of thePacific Northwest, For. Ecol. Manage., 220,313–325, doi:10.1016/j.foreco.2005.08.018.Barber, S. A., and M. Silberbush (1984), Plant root morphology and nutri-ent uptake, in Roots, Nutrient and Water Influx, and Plant Growth, ASASpec. Publ., vol. 49, edited by S. A. Barber and D. R. Bouldin, pp. 65–87, Soil Sci. Soc. of Am., Madison, Wis.Bernacchi, C. J., C. Pimentel, and S. P. Long (2003), In vivo temperatureresponse functions of parameters required to model RuBP‐limited photo-synthesis, Plant Cell Environ., 26, 1419–1430, doi:10.1046/j.0016-8025.2003.01050.x.Belser, L. W., and E. L. Schmidt (1980), Growth and oxidation kinetics ofthe three genera of ammonia oxidizers, FEMS Microbiol. Lett., 7, 213–216, doi:10.1111/j.1574-6941.1980.tb01628.x.Berg, B., and C. McClaugherty (2003), Plant Litter: Decomposition,Humus Formation, Carbon Sequestration, Springer, Berlin.Blackburne, R., V. M. Vadivelu, Z. Yuan, and J. Keller (2007), Kineticcharacterisation of an enriched Nitrospira culture with comparison toNitrobacter, Water Res., 41,3033–3042, doi:10.1016/j.watres.2007.01.043.Bowden, R. D., J. M. Melillo, P. D. Steudler, and J. D. Aber (1991), Effectsof nitrogen additions on annual nitrous oxide fluxes from temperate for-est soils in the northeastern United States, J. Geophys. Res., 96, 9321–9328, doi:10.1029/91JD00151.Brumme, R., and F. Beese (1992), Effects of liming and nitrogen fertiliza-tion on emissions of CO2and N2O from a temporate forest, J. Geophys.Res., 97(D12), 12,851–12,858, doi:10.1029/92JD01217.Burton, A. J., K. S. Pregitzer, J. N. Crawford, G. P. Zogg, and D. R. Zak(2004), Simulated chronic NO3‐ deposition reduces soil respiration innorthern hardwood forests, Global Change Biol., 10,1080–1091,doi:10.1111/j.1365-2486.2004.00737.x.Cleemput, O., and A. H. Samater (1996), Nitrite in soils: Accumulation androle in the formation of gaseous N compounds, Fert. Res., 45,81–89,doi:10.1007/BF00749884.Edmonds, R. L., and T. Hsiang (1987), Forest floor and soil influence onresponse of Douglas fir to urea, Soil Sci. Soc. Am. J., 51, 1332–1337,doi:10.2136/sssaj1987.03615995005100050043x.Ericsson, T., L. Rytter, and E. Vapaavuori (1996), Physiology of carbonallocation in trees, Biomass Bioenergy, 11,115–127, doi:10.1016/0961-9534(96)00032-3.Farquhar, G. D., S. von Caemmerer, and J. A. Berry (1980), A biochemicalmodel of photosynthetic CO2assimilation in leaves of C3species, Planta,149,78–90, doi:10.1007/BF00386231.Flint, C., A. Adams, B. Strahm, and R. Harrison (2008), Nitrogen Leachingfrom Douglas fir Forests after Urea Fertilization, J. Environ. Qual., 37,1781–1788, doi:10.2134/jeq2007.0367.Giardina, C. P., D. Binkley, M. G. Ryan, J. H. Fownes, and R. S. Senock(2004), Belowground carbon cycling in a humid tropical forest decreaseswith fertilization, Oecologia, 139, 545–550, doi:10.1007/s00442-004-1552-0.Gower, S. T., K. A. Vogt, and C. C. Grier (1992), Carbon dynamics ofRocky Mountain Douglas fir: Influence of water and nutrient availability,Ecol. Monogr., 62,43–65, doi:10.2307/2937170.Grant, R. F. (1993), Rhizodeposition by crop plants and its relationship tomicrobial activity and nitrogen distribution, Model. Geo‐Biol. Processes,2, 193–209.Grant, R. F. (1994), Simulation of ecological controls on nitrification, SoilBiol. Biochem., 26, 305–315, doi:10.1016/0038-0717(94)90279-8.Grant, R. F. (1995), Mathematical modelling of nitrous oxide evolutionduring nitrification, Soil Biol. Biochem., 27, 1117–1125, doi:10.1016/0038-0717(95)00038-G.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G0400915 of 17Grant, R. F. (1998), Simulation in ecosys of root growth response to con-trasting soil water and nitrogen, Ecol. Modell., 107,237–264,doi:10.1016/S0304-3800(97)00221-4.Grant, R. F. (2004), Modelling topographic effects on net ecosystem pro-ductivity of boreal black spruce forests, Tree Physiol., 24,1–18.Grant, R. F., and E. Pattey (2003), Modelling variability in N2O emissionsfrom fertilized agricultural fields, Soil Biol. Biochem., 35,225–243,doi:10.1016/S0038-0717(02)00256-0.Grant, R. F., and L. B. Flanagan (2007), Modeling stomatal and nonstoma-tal effects of water deficits on CO2fixation in a semiarid grassland, J.Geophys. Res., 112, G03011, doi:10.1029/2006JG000302.Grant, R. F., N. G. Juma, and W. B. McGill (1993a), Simulation of carbonand nitrogen transformations in soils: I. Mineralization, Soil Biol. Bio-chem., 25, 1317–1329, doi:10.1016/0038-0717(93)90046-E.Grant, R. F., N. G. Juma, and W. B. McGill (1993b), Simulation of carbonand nitrogen transformations in soils. II. Microbial biomass and meta-bolic products, Soil Biol. Biochem., 25, 1331–1338, doi:10.1016/0038-0717(93)90047-F.Grant, R. F., et al. (2001), CO2effects on mass and energy exchange ofwheat under different N fertilization: Model theory and testing with aFree Air CO2Enrichment (FACE) experiment, Agron. J., 93, 638–649,doi:10.2134/agronj2001.933638x.Grant, R. F., A. Arain, V. Arora, A. Barr, T. A. Black, J. Chen, S. Wang,F. Yuan, and Y. Zhang (2005), Intercomparison of techniques to modelhigh temperature effects on CO2and energy exchange in temperate andboreal coniferous forests, Ecol. Modell., 188, 217–252, doi:10.1016/j.ecolmodel.2005.01.060.Grant, R. F., T. A. Black, D. Gaumont‐Guay, N. Kljun, A. G. Barr,K. Morgenstern, and Z. Nesic (2006a), Net ecosystem productivity ofboreal aspen forests under drought and climate change: Mathematicalmodeling with ecosys, Agric. For. Meteorol., 140, 152–170,doi:10.1016/j.agrformet.2006.01.012.Grant, R. F., E. M. Pattey, T. W. Goddard, L. M. Kryzanowski, andH. Puurveen (2006b), Modeling the effects of fertilizer application rateon nitrous oxide emissions from agricultural fields, Soil Sci. Soc. Am.J., 70, 235–248, doi:10.2136/sssaj2005.0104.Grant, R. F., et al. (2006c), Intercomparison of techniques to model waterstress effects on CO2and energy exchange in temperate and boreal decid-uous forests, Ecol. Modell., 196,289–312, doi:10.1016/j.ecolmodel.2006.02.035.Grant, R. F., T. J. Arkebauer, A. Dobermann, K. G. Hubbard, T. T.Schimelfenig, A. E. Suyker, S. B. Verma, and D. T. Walters (2007a), Netbiome productivity of irrigated and rainfed maize–soybean rotations:Modeling versus measurements, Agron. J., 99, 1404–1423, doi:10.2134/agronj2006.0308.Grant, R. F., et al. (2007b), Net ecosystem productivity of boreal jack pinestands regenerating from clearcutting under current and future climates,Global Change Biol., 13, 1423–1440, doi:10.1111/j.1365-2486.2007.01363.x.Grant, R. F., T. A. Black, E. R. Humphreys, and K. Morgenstern (2007c),Changes in net ecosystem productivity with forest age following clear-cutting of a coastal Douglas fir forest: Testing a mathematical model witheddy covariance measurements along a forest chronosequence, Tree Phy-siol., 27, 115–131.Grant,R.F.,A.G.Barr,T.A.Black,H.A.Margolis,A.L.Dunn, J. Metsaranta, and S. Wang (2009), Interannual variation in netecosystem productivity of Canadian forests as affected by regionalweather patterns—AFluxnet‐Canada synthesis, Agric. For. Meteorol.,149, 2022–2039, doi:10.1016/j.agrformet.2009.07.010.Haynes, B. E., and S. T. Gower (1995), Belowground carbon allocation inunfertilized and fertilized red pine plantations in northern Wisconsin,Tree Physiol., 15, 317–325.Holland, E. A., F. J. Dentener, B. H. Braswell, and J. M. Sulzman (1999),Contemporary and pre‐industrial global reactive nitrogen budgets, Bio-geochemistry, 46,7–43, doi:10.1007/BF01007572.Hopmans, P., and H. N. Chappell (1994), Growth response of young,thinned Douglas fir stands to nitrogen fertilizer in relation to soil proper-ties and tree nutrition, Can. J. For. Res., 24(8), 1684–1688, doi:10.1139/x94-217.Humphreys, E. R. (2004), Net ecosystem production of three coastalDouglas fir stands at different stages of development after harvesting,Ph.D. thesis, 122 pp., Univ. of B. C., Vancouver, Canada.Humphreys, E. R., T. A. Black, K. Morgenstern, T. Cai, G. B. Drewitt,Z. Nesic, and J. A. Trofymow (2006), Carbon dioxide fluxes in coastalDouglasfirstandsatdifferentstagesofdevelopmentafterclearcutharvest-ing, Agric. For. Meteorol., 140,6–22, doi:10.1016/j.agrformet.2006.03.018.Iivonen, S., R. Rikala, and E. Vapaavuori (2001), Seasonal root growth ofScots pine seedlings in relation to shoot phenology, carbohydrate statusand nutrient supply, Can. J. For. Res., 31, 1569–1578, doi:10.1139/cjfr-31-9-1569.Iivonen, S., S. Kaakinen, A. Jolkkonen, E. Vapaavuori, and S. Linder(2006), Influence of long‐term nutrient optimization on biomass, carbon,and nitrogen acquisition and allocation in Norway spruce, Can. J. For.Res., 36, 1563–1571, doi:10.1139/X06-035.Jassal,R.S.,T.A.Black,T.Cai,K.Morgenstern,Z.Li,D.Gaumont‐Guay, and Z. Nesic (2007), Components of ecosystem respiration andan estimate of net primary productivity of an intermediate‐aged Douglasfir stand, Agric. For. Meteorol., 144,44–57, doi:10.1016/j.agrfor-met.2007.01.011.Jassal, R. S., T. A. Black, B. Z. Chen, R. Roy, Z. Nesic, D. L. Spittlehouse,and J. A. Trofymow (2008), N2O emissions and carbon sequestration in anitrogen‐fertilized Douglas fir stand, J. Geophys. Res., 113, G04013,doi:10.1029/2008JG000764.Jassal, R. S., T. A. Black, T. Cai, G. Ethier, S. Pepin, C. Bruemmer,Z. Nesic, D. L. Spittlehouse, and J. A. Trofymow (2009), Impact of nitro-gen fertilization on carbon and water balances in a chronosequence ofthree Douglas fir stands in the Pacific Northwest, Agric. For. Meteorol.,doi:10.1016/j.agrformet.2009.10.005, in press.Jassal, R. S., T. A. Black, J. A. Trofymow, R. Roy, and Z. Nesic (2010),Soil CO2and N2O flux dynamics in a nitrogen‐fertilized Pacific North-west Douglas‐fir stand, Geoderma, 157,118–125, doi:10.1016/j.geoderma.2010.04.002.Johnson, D. W., and P. S. Curtis (2001), Effects of forest management onsoil C and N storage: Meta analysis, For. Ecol. Manage., 140, 227–238,doi:10.1016/S0378-1127(00)00282-6.Kaakinen, S., A. Jolkkonen, S. Iivonen, and E. Vapaavuori (2004), Growth,allocation and tissue chemistry of Picea abies seedlings affected bynutrient supply during the second growing season, Tree Physiol., 24,707–719.Lal, R. B., D. E. Kissel, M. L. Cabrera, and A. P. Schwab (1993), Kineticsof urea hydrolysis in wheat residue, Soil Biol. Biochem., 25, 1033–1036,doi:10.1016/0038-0717(93)90151-Z.Martikainen, P. J., and W. de Boer (1993), Nitrous oxide production andnitrification in acidic soil from a Dutch coniferous forest, Soil Biol. Bio-chem., 25, 343–347, doi:10.1016/0038-0717(93)90133-V.Matson, P. A., S. T. Gower, C. Volkmann, C. Billow, and C. C. Grier(1992), Soil nitrogen cycling and nitrous oxide flux in a Rocky MountainDouglas fir forest: Effects of fertilization, irrigation and carbon addition,Biogeochemistry, 18, 101–117, doi:10.1007/BF00002705.Mead, D. J., S. X. Chang, and C. M. Preston (2008), Recovery of15N‐urea10 years after application to a Douglas fir pole stand in coastal BritishColumbia, For. Ecol. Manage., 256,694–701, doi:10.1016/j.foreco.2008.05.022.Meteorological Service of Canada (2004), Canadian Acid DepositionScience Assessment, Environ. Can., Ottawa.Morgenstern, K., T. A. Black, E. R. Humphreys, T. J. Griffis, T. Cai, G. B.Drewitt, D. Gaumont‐Guay, and Z. Nesic (2004), Sensitivity and uncer-tainty of the carbon balance of a Pacific Northwest Douglas fir forest dur-ing an El Niño/La Niña cycle, Agric. For. Meteorol., 123,201–219,doi:10.1016/j.agrformet.2003.12.003.Mørkved, P. T., P. Dörsch, and L. R. Bakken (2007), The N2Oproductratio of nitrification and its dependence on long‐term changes in soilpH, Soil Biol. Biochem., 39, 2048–2057, doi:10.1016/j.soilbio.2007.03.006.Moyo, C. C., D. E. Kissel, and M. L. Cabrera (1989), Temperature effectson soil urease activity, Soil Biol. Biochem., 21,935–938, doi:10.1016/0038-0717(89)90083-7.Nason, G. E., D. J. Pluth, R. T. Hardin, and W. B. McGill (1990), Dynam-ics of foliar N in Douglas fir after spring and fall application of ammo-nium nitrate and urea, Can. J. For. Res., 20,1515–1523, doi:10.1139/x90-201.Olsson, P., S. Linder, R. Giesler, and P. Högberg (2005), Fertilization ofboreal forest reduces both autotrophic and heterotrophic soil respiration,Global Change Biol., 11, 1745–1753, doi:10.1111/j.1365-2486.2005.001033.x.Prietzel, J., G. L. Wagoner, and R. B. Harrison (2004), Long‐term effectsof repeated urea fertilization in Douglas fir stands on forest floor nitrogenpools and nitrogen mineralization, For. Ecol. Manage., 193,413–426,doi:10.1016/j.foreco.2004.02.006.Régnière, J., and R. St‐Amant (2007), Stochastic simulation of daily airtemperature and precipitation from monthly normals in North Americanorth of Mexico, Int. J. Biometeorol., 51,415–430, doi:10.1007/s00484-006-0078-z.Sitaula, B. K., L. R. Bakken, and G. Abrahamsen (1995a), N‐fertilizationand soil acidification effects on N2O and CO2emission from temperatepine forest soil, Soil Biol. Biochem., 27, 1401–1408, doi:10.1016/0038-0717(95)00078-S.GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G0400916 of 17Sitaula, B. K., L. R. Bakken, and G. Abrahamsen (1995b), Nutrient balancein Scots pine (pinus sylvestris l.) forest. 3. Fluxes of N2O from lysimeteras influenced by nitrogen input, Water Air Soil Pollut., 85, 1155–1159,doi:10.1007/BF00477137.Stark, J. M., and M. K. Firestone (1996), Kinetic characteristics of ammo-nium‐oxidizer communities in a California oak woodland‐annual grass-land, Soil Biol. Biochem., 28, 1307–1317, doi:10.1016/S0038-0717(96)00133-2.Stegemoeller, K. A., and H. N. Chappell (1990), Growth response of un-thinned and thinned Douglas‐fir stands to single and multiple applica-tions of nitrogen, Can. J. For. Res., 20, 343–349, doi:10.1139/x90-050.Valinger, E. (1993), Effects of thinning and nitrogen fertilization on growthof Scots pine trees: Total annual biomass increment, needle efficiency,and above‐ground allocation of biomass increment, Can. J. For. Res.,23, 1639–1644, doi:10.1139/x93-204.Vlek, P. L. G., and M. F. Carter (1983), The effect of soil environment andfertilizer modifications on the rate of urea hydrolysis, Soil Sci., 136,56–63, doi:10.1097/00010694-198307000-00008.Welke, S. E., and G. D. Hope (2005), Influences of stand composition andage on forest floor processes and chemistry in pure and mixed stands ofDouglas fir and paper birch in interior British Columbia, For. Ecol.Manage., 219,29–42, doi:10.1016/j.foreco.2005.08.040.Yoshinari, T., R. Hynes, and R. Knowles (1977), Acetylene inhibition ofnitrous oxide reduction and measurement of denitrification and nitrogenfixation in soil, Soil Biol. Biochem., 9, 177–183, doi:10.1016/0038-0717(77)90072-4.T. A. Black, C. Bruemmer, and R. S. Jassal, Biometeorology and SoilPhysics Group, Faculty of Land and Food Systems, University of BritishColumbia, Vancouver, BC, V6T 1Z4, Canada.R. F. Grant, Department of Renewable Resources, University of Alberta,4‐30 Earth Science Bldg., Edmonton, AB T6G 2E3, Canada. (robert.grant@ales.ualberta.ca)GRANT ET AL.: FERTILIZATION AND NEP OF DOUGLAS FIR G04009G0400917 of 17


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items