Biogeochemistry of a temperate forest nitrogen gradient

Ecology, 92(7), 2011, pp. 1481–1491� 2011 by the Ecological Society of America

Biogeochemistry of a temperate forest nitrogen gradient

STEVEN S. PERAKIS1,3 AND EMILY R. SINKHORN2

1U.S. Geological Survey, Forest and Rangeland Ecosystem Science Center, Corvallis, Oregon 97331 USA2Department of Forest Ecosystems and Society, Oregon State University, Corvallis, Oregon 97331 USA

Abstract. Wide natural gradients of soil nitrogen (N) can be used to examine fundamentalrelationships between plant–soil–microbial N cycling and hydrologic N loss, and to test N-saturation theory as a general framework for understanding ecosystem N dynamics. Wecharacterized plant production, N uptake and return in litterfall, soil gross and net Nmineralization rates, and hydrologic N losses of nine Douglas-fir (Pseudotsuga menziesii )forests across a wide soil N gradient in the Oregon Coast Range (USA). Surface mineral soil N(0–10 cm) ranged nearly three-fold from 0.29% to 0.78% N, and in contrast to predictions ofN-saturation theory, was linearly related to 10-fold variation in net N mineralization, from 8to 82 kg N�ha�1�yr�1. Net N mineralization was unrelated to soil C:N, soil texture,precipitation, and temperature differences among sites. Net nitrification was negatively relatedto soil pH, and accounted for ,20% of net N mineralization at low-N sites, increasing to 85–100% of net N mineralization at intermediate- and high-N sites. The ratio of net : gross Nmineralization and nitrification increased along the gradient, indicating progressive saturationof microbial N demands at high soil N. Aboveground N uptake by plants increasedasymptotically with net N mineralization to a peak of ;35 kg N�ha�1�yr�1. Aboveground netprimary production per unit net N mineralization varied inversely with soil N, suggestingprogressive saturation of plant N demands at high soil N. Hydrologic N losses weredominated by dissolved organic N at low-N sites, with increased nitrate loss causing a shift todominance by nitrate at high-N sites, particularly where net nitrification exceeded plant Ndemands. With the exception of N mineralization patterns, our results broadly support theapplication of the N-saturation model developed from studies of anthropogenic N depositionto understand N cycling and saturation of plant and microbial sinks along natural soil Ngradients. This convergence of behavior in unpolluted and polluted forest N cycles suggeststhat where future reductions in deposition to polluted sites do occur, symptoms of Nsaturation are most likely to persist where soil N content remains elevated.

Key words: ammonium; dissolved organic nitrogen; Douglas-fir forests; equilibrium; gross and netnitrogen mineralization; nitrate leaching; nitrification; nutrient retention; Oregon Coast Range, USA;productivity; Pseudotsuga menziesii; temperate forest.

INTRODUCTION

Nitrogen (N) availability limits the productivity of

most temperate forests worldwide (LeBauer and Tre-

seder 2008). Soils are critical in shaping patterns of N

limitation, both by influencing stocks of growth-limiting

nutrients, and though microbial processes that decom-

pose organic matter to release simple organic and

inorganic forms of N that fuel plant growth (Schimel

and Bennett 2004). Plants also regulate their own access

to available soil N by shaping soil microbial activity

through carbon subsidies, and by employing a range of

nutrient-acquisition strategies (Chapman et al. 2006).

Plant, soil, and microbial processes can interact through

ecosystem feedbacks to further modify N dynamics and

availability within ecosystems. For example, there is

long-standing appreciation that plants growing on

infertile low-N sites use N efficiently and produce

biomass with low tissue-N concentrations that, when

senesced, decomposes and releases N slowly for subse-

quent plant growth (Gosz 1981, Vitousek 1982). As N

availability increases on more-fertile sites, feedbacks of

soil N supply, plant N uptake, and microbial N turnover

accelerate ecosystem N dynamics. Most studies demon-

strate such feedbacks through studies where species (and

associated nutrient cycling strategies) shift across sites,

yet single-species stands may also exhibit these charac-

teristics (Prescott et al. 2000). While such feedbacks do

not increase indefinitely, it remains unclear which

processes set limits to feedbacks in plant–soil–microbial

N cycling as N fertility increases.

Atmospheric N deposition has greatly increased N

inputs to many temperate forests worldwide and lead to

development of N-saturation theory, which provides a

framework for evaluating how N supply shapes forest N

dynamics (Agren and Bosatta 1988, Aber et al. 1989).

Nitrogen saturation is most often defined as occurring

when N supply exceeds ecosystem demands, resulting in

Manuscript received 7 September 2010; revised 17 February2011; accepted 23 February 2011. Corresponding Editor: R. W.Ruess.

3 E-mail: [email protected]

1481

increased N loss, eventually to where N losses match

inputs. As N losses progressively increase, net Naccumulation in plants and soils may still occur (Perakis

et al. 2005), but it is unknown if this is associated withfurther acceleration of N cycling in N-saturated stands.In addition, although there is great practical interest in

understanding the factors controlling the onset of Nsaturation in forests, it is also unclear whether N-

saturation theory is a generally robust way of describingforest N cycling along N supply gradients outside thecontext of chronic atmospheric N deposition (Agren and

Bosatta 1988). Can large accumulations of soil N alsoresult in soil N supply that exceeds ecosystem demands?

Few areas worldwide have yet received sufficient Ndeposition to appreciably alter soil N accumulation andlong-term cycling rates (Nave et al. 2009), and reduc-

tions in atmospheric N inputs can rapidly reduce keysymptoms of N saturation in forests (Bredemeier et al.

1998, Corre and Lamersdorf 2004). To the extent thatchronic N deposition increases long-term accumulationof soil N in the future, it is unknown whether plant–soil–

microbial feedback will maintain high rates of ecosystemN cycling and possibly N saturation, even if N inputs areeventually reduced.

Relationships among soil N accumulation, soil Nsupply, and plant productivity are important not only

for understanding long-term impacts of atmospheric Ndeposition and commercial forest N fertilization, but

also more fundamentally for testing conceptual modelsof N limitation. In a synthesis of studies encompassing50 temperate forests Reich et al. (1997) found that net

primary productivity increased linearly across an excep-tionally broad range of soil net N mineralization, withslight variation among forests attributable primarily to

soil type. However, their work did not explicitly considervariation in soil N across sites, and often soil N is a poor

predictor of net N mineralization where climate, species,and soil texture and/or type vary among sites (Nadel-hoffer et al. 1983, McCulley et al. 2009). As a result, we

lack clear understanding of whether variation inecosystem N accumulation alone translates to variation

in soil N supply and productivity. Prescott et al. (2000),

working in Douglas-fir forests across a fairly narrow

climate zone, found greater N return in litter and forest-

floor N turnover along a gradient of increasing soil-N

capital, yet surprisingly soil N mineralization (measured

in laboratory incubations) was not related to soil N

capital or plant production. Because field and laborato-

ry incubation of soils can yield divergent results (Hart et

al. 1994), there is still a need to understand whether wide

gradients in soil N accumulation drive in situ variation

in soil N supply and plant production, and whether this

relationship is linear or nonlinear over the range of

natural N accumulation observed in temperate forests.

We examined ecosystem N dynamics across a wide

gradient in surface soil N of nine planted Douglas-fir

forests of the Oregon Coast Range (USA). All of the

study sites occupied a narrow bioclimatic province with

low atmospheric N inputs, on soils developed from

sandstone parent materials, which allowed us to examine

how soil N influenced N dynamics and plant–soil–

microbial feedbacks to N cycling without confounding

soil, climate, or vegetation effects. Within this context,

we examined the following specific questions: How does

soil organic-N accumulation influence inorganic-N

supply? Do net and gross N cycling rates covary with

soil N? Does the balance of microbial N supply and

demand, and of inorganic-N turnover, vary with soil N?

Does increasing N supply result in linear or nonlinear

response in plant production, N uptake, and N

recycling? Do forms of dissolved N in hydrologic losses

shift with soil N? We compare our findings across this

gradient to predictions from N-saturation theory (Agren

and Bosatta 1988, Aber et al. 1989, 1998) to evaluate the

generality of forest response to long-term N enrichment.

METHODS

Study sites

We selected a subset of nine study sites located in the

north-central Oregon Coast Range, USA, (Table 1)

from a broader range of sites shown previously to span a

wide range of soil and foliar N concentrations (Perakis

et al. 2006). We restricted sites to areas developed on

TABLE 1. Features of the nine study sites in the Oregon Coastal Range (USA).

Location Climate

Standage (yr)

Soil characteristic

Siteno.

Latitude(N)

Longitude(W)

Temp.(8C)

Ppt.(cm/yr)

Soil type(USDA)

Sand(%)

Silt(%)

Clay(%)

pH(H2O) C (%) N (%)

7 458800 1238390 11.3 180 31 Andic Dystrudept 33 20 48 5.83 6.04 0.2920 458480 1238240 9.5 231 22 Alic Hapludand 30 44 26 5.22 6.59 0.3176 448350 1238480 11.6 172 26 Andic Dystrudept 38 26 36 4.99 5.37 0.335 448380 1238480 11.4 173 29 Andic Dystrudept 53 23 25 5.50 6.25 0.3377 448420 1238400 11.2 158 28 Andic Dystrudept 29 31 40 5.30 5.49 0.3458 448430 1238490 11.2 175 23 Andic Dystrudept 40 28 33 4.79 6.65 0.3822 448360 1238550 10.8 155 29 Andic Dystrudept 46 25 29 5.61 8.24 0.5616 458100 1238550 11.2 169 26 Andic Dystrudept 30 31 39 4.61 9.22 0.5739 458440 1238530 10.7 200 27 Andic Dystrudept 33 29 39 4.13 16.43 0.78

Notes: Sites are arranged in order of increasing percentage N in the soil. Temperature (Temp.) and precipitation (Ppt.) areaverage annual data derived from PRISM (Daly et al. 1994). Stand age is for the year 2006. Soil variables were measured in surface(0–10 cm depth) mineral soil.

STEVEN S. PERAKIS AND EMILY R. SINKHORN1482 Ecology, Vol. 92, No. 7

sandstone parent materials with Andic soil properties,

with no evidence of significant geological N (Dalhgren et

al. 1994). The surface mineral soil (0–10 cm) of study

sites ranged from 0.29% to 0.78% N, which we used as a

predictor variable in many data analyses because of its

ease of measurement, and hence widespread use in

ecology, forestry, and soil-science research. At the time

of our study all stands were fully stocked plantations

dominated by Douglas-fir (Pseudotsuga menziesii ) (on

average, 93.4% of basal area .5 cm) ranging from 22 to

31 years of age, with minor contributions of western

hemlock (Tsuga heterophylla) (on average, 2.1% of basal

area), and sitka spruce (Picea sitchensis), cascara

buckthorn (Rhamnus purshiana), bitter cherry (Prunus

emarginata), red alder (Alnus rubra), and bigleaf maple

(Acer macrophyllum) (all ,2% basal area). Landowner

records indicate that all sites except number 39 included

a large component of N2-fixing red alder prior to

clearing, broadcast burning, and establishment of the

current stands. There is no history of fertilization at any

of the sites.

Vegetation and soil nitrogen

We used a pole pruner to clip foliage samples in

summer 2005 from the tip of the southern-most branch

in the fifth whorl from the top of three trees per site. In

winter 2005 we sampled stemwood by coring six trees

per site and bulking cores into two composite samples

per site, and also sampled branches using a pole pruner.

We collected litterfall monthly from September 2004 to

June 2006 using 10 traps (each 0.26 m2) per site and

bulked samples into three composites per site for

analysis. We also separated subsamples of Douglas-fir

needle litter from bulk litter for chemical analysis.

We used 0.02-ha permanent plots (Maguire et al.

2002) to inventory vegetation in 2002 and 2004, and

converted field measurements to stand biomass and

annual increment in foliage, branches, and stems using

an allometric growth model developed at these sites

(Weiskittel et al. 2010). We estimated aboveground net

primary production (ANPP) as the summed net

increment in foliage, branches, and stems, plus annual

litterfall. We estimated N pools in each tissue class by

multiplying biomass against the respective N concen-

tration. We estimated annual vegetation N uptake on

each site as the sum of N in average annual net

increment of foliage, branches, and stems, plus the N

content of litterfall. We calculated stand-level nutrient-

use efficiency (NUE) as ANPP per unit net N uptake by

vegetation. We also calculated resource supply efficiency

(RSE; after Binkley et al. 2004) as ANPP per unit soil

net N mineralization, and displayed its variation with

soil N instead of net N mineralization to avoid

autocorrelation.

We sampled soil for total N content at four locations

in each site, starting with forest floor in a 30 3 30 cm

frame, and followed by mineral soil in six depth

increments covering 0–100 cm (0–10 cm, 10–20 cm,

20–30 cm, 30–50 cm, 50–70 cm, 70–100 cm), using 303

30 cm pits down to 30 cm depth, and 4 cm diameter

cores for subsequent depths. We sieved mineral soil to 2

mm and dried a 10-g subsample for 48 h at 1058C for

moisture content. We dried ground foliage, branches,

stem tissue, litterfall, forest floor, and mineral soil at

658C for 48 h prior to fine grinding and analysis for total

C and N on a Costech ECS-4010 elemental combustion

analyzer (Costech Analytical, Valencia, California,

USA). We calculated soil nutrient content by multiply-

ing concentrations against gravel-corrected bulk density

in each horizon. We determined soil solution pH (2

water : 1 soil) using fresh soil from each depth increment

with an Accumet pH meter (Fisher Scientific, Hampton,

New Hampshire, USA). We analyzed subsamples of 0–

10 cm soil for texture by the hydrometer method.

Soil nitrogen transformations

We determined monthly net N mineralization and

nitrification rates in 0–10 cm mineral soil from

September 2004 to December 2005 using six replicate

intact cores (5 3 10 cm) per site, buried in polyethylene

bags (Hart et al. 1994). We processed both initial and

final samples on the same day as collection by sieving to

2 mm, extracting 7 g of soil for 1 h with 35 mL 2 mol/L

KCl, filtering through prerinsed Whatman number 40,

then freezing extracts until analysis. We analyzed

extractable NH4þ and NO3

� colorometrically by salic-

ylate and cadmium reduction, respectively, on a Lachat

QuikChem 8000 Flow-Injection Autoanalyzer (Lachat

Instruments, Milwaukee, Wisconsin, USA). We correct-

ed net N mineralization (NH4þþNO3

�) and nitrification

(NO3�) rates for extractable N in initial cores. We

calculated annual net N mineralization by multiplying

concentrations against 0–10 cm bulk density at each site

and summing averages of each month to annual totals

per site.

We measured gross N mineralization and nitrification

on three dates (October 2004, February 2005, and June

2005) at four locations per site using 15N pool dilution

(Hart et al. 1994). Briefly, we injected individual cores

with 99.9 atom% 15N solutions (20 lg N/g dry soil) as

either 15NH4Cl or Na15NO3, collected initial samples 15

minutes after labeling, capped the remaining cores, and

placed them in polyethylene bags in the original holes

for 24 h. After incubation, we extracted 60 g soil from

each core for 1 h with 300 mL 0.5 mol/L K2SO4, then

filtered and froze extracts as above. We prepared

extractable NH4þ and NO3

� for 15N determination

using an eight-day Teflon tape diffusion (Stark and Hart

1996), followed by analysis at the Utah State University

Stable Isotope Laboratory. We calculated gross miner-

alization and nitrification rates according to standard

equations (Kirkham and Bartholomew 1954), and

estimated annual rates by averaging across the three

labeling dates. We calculated the mean residence times

(MRT) of soil inorganic N by dividing inorganic N

pools by gross N production rates (Davidson et al.

July 2011 1483FOREST NITROGEN BIOGEOCHEMISTRY

1992). To assess microbial N demands relative to supply,

while avoiding the use of gross N consumption rates that

can be stimulated by N substrate addition (Booth et al.

2005), we calculated the average annual fraction of gross

N production (from pool dilutions) that is released as

net inorganic N (from buried bags) as the ratio of

net : gross mineralization.

Soil hydrologic nitrogen fluxes

We characterized hydrologic N fluxes using six low-

tension lysimeters (Prenart Equipment ApS, Frederiks-

berg, Denmark) per site, three each at 20 cm (shallow)

and 100 cm (deep) depths. We installed lysimeters in

August 2004 using the method described in Perakis et al.

(2005) and sampled monthly from December 2004 to

June 2006 by application of 12 mm Hg vacuum for 48 h.

We collected replicate samples in 60-mL HDPE bottles,

preserved one sample with 0.2 mL of chloroform, and

kept samples cold prior to analysis for dissolved N

within 48 h. We analyzed samples for NH4þ and NO3

�

colorimetrically, as before, and analyzed for total

nitrogen (TN) by catalytic oxidation combustion using

a Shimadzu TOC-V CSH total organic-carbon analyzer

with a TNM-1 total nitrogen measuring unit (Shimadzu

Scientific Instruments, Columbia, Maryland, USA). We

calculated dissolved organic nitrogen (DON) as TN

minus NH4þ-N minus NO3

�-N.

We converted lysimeter N concentrations to monthly

N fluxes using a water balance calculated with PRISM

climate data and 3-PG modeling. We first obtained

monthly precipitation and temperature for each site

from PRISM output (Precipitation-elevation Regres-

sions on Independent Slopes Model; PRISM Group,

Oregon State University, Corvallis, Oregon, USA).

PRISM accounts for rain shadows, temperature inver-

sions, and coastal effects in the climate-mapping process

across a 4-km grid size, using a digital elevation model

to account for grid cell and weather station location

(Daly et al. 1994). We then used PRISM climate data to

drive 3-PG (Physiological Processes for Predicting

Growth), a forest process model that uses the Penman-

Monteith equation to estimate evapotranspiration

(Lansberg and Waring 1997). We constrained maximum

canopy conductance at 0.013 m/s by running a

sensitivity analysis with 3-PG to determine the conduc-

tance that produced the most reasonable maximum leaf-

area index and maximum annual increment for a site II

Douglas-fir forest in the Oregon Coast Range (Waring

et al. 2002, 2006). We estimated the volume of water

leaching past 100 cm as the difference between

precipitation and evapotranspiration, and estimated

transpiration in the upper 20 cm from the depth

distribution of fine roots from a nearby Coast Range

site (Lee et al. 2007).

Data analysis

We used SYSTAT 11 (SYSTAT Software 2004) to

conduct least-squares linear and nonlinear regression

and Pearson correlation to evaluate relationships

between response variables and indices of soil N

availability. We used log-transformed data when neces-

sary to meet assumptions of normality.

RESULTS

Surface (0–10 cm) mineral soil percentage nitrogen

(henceforth, soil %N) varied 2.7-fold across the nine

sites, from 0.29% to 0.78% N (Table 2). Total soil N of

forest floor and mineral soil to 1 m depth varied 2.6-fold

across sites, from 8609 to 22 379 kg N/ha (Table 2).

Surface mineral soil %N was closely correlated with

both surface soil %C (r¼0.93, P , 0.001, n¼9 sites) and

total soil N across sites (r¼ 0.88, P¼ 0.002, n¼ 9). The

amount of N stored in aboveground biomass (average

245 kg N/ha; range 162–296 kg N/ha, n ¼ 9) was small

compared to combined forest floor and mineral soil N

pools, with forest floor and mineral soil accounting for

an average of 98.3% of total ecosystem N across sites

(range 97.7–99.0%).

TABLE 2. Soil N concentrations, pools, net cycling, gross cycling, and mean residence times (MRT) of NH4þ and NO3

� at the ninesites.

SiteSoilN (%)

Soil Ncapital

(kg N/ha)

N cycling (kg N�ha�1�yr�1)

NH4þ

MRT (d)NO3

�

MRT (d)Net N

mineralizationNet

nitrificationGross N

mineralizationGross

nitrification

7 0.29 9486 8 2 91 91 2.9 0.120 0.31 8609 12 0 206 51 1.4 0.076 0.33 11 627 25 23 87 92 4.4 1.55 0.33 14 312 22 17 88 53 2.6 0.377 0.34 13 647 37 39 145 93 3.4 3.058 0.38 13 794 42 40 82 61 4.9 2.522 0.56 22 379 45 45 81 82 3.9 3.516 0.57 17 931 51 49 78 72 4.9 6.739 0.78 21 044 82 65 70 10 8.3 22.9

Notes: Sites are arranged in order of increasing soil %N. Soil characteristics were measured in 0–10 cm mineral soil, except soil Ncapital, which is the sum of total N in forest floor and mineral soil (with particulate size ,2 mm) to 1 m depth. Net Nmineralization and nitrification are annual sums of monthly rates. Gross N mineralization and nitrification were measured threetimes (October 2004, February 2005, and June 2005). Mean residence times are calculated by dividing extractable inorganic N intogross N production separately for ammonium and nitrate.


Annual net N mineralization in 0–10 cm soil ranged

from 7.8 to 81.9 kg N�ha�1�yr�1 (Table 2), and increased

linearly across the gradient in surface soil %N (r2¼ 0.87,

P , 0.001; Fig. 1A). Soil N mineralization was not

related to mean annual temperature (r2 , 0.01),

precipitation (r2 ¼ 0.02), soil C:N (r2 ¼ 0.04), or soil

texture (r2 , 0.01). Surface 0–10 cm soil C:N values

were low overall (average C:N¼18, range 14.6�21.5, n¼9) and the percentage of silt þ clay in surface 0–10 cm

soil varied from 48% to 71% across sites. Nitrification

was 1.2% and 28.6% of total net N mineralization at the

two lowest-N sites, and it approached 100% at high-N

sites (Fig. 1B). Neither temperature, precipitation, C:N,

nor texture effectively predicted nitrification rates across

sites (all r2 , 0.18), although nitrification was positively

related to surface soil %N (r2 ¼ 0.72, P ¼ 0.004) and

negatively correlated with soil pH (r2¼ 0.77, P¼ 0.002).

Gross N mineralization and gross nitrification did not

vary systematically across the surface soil N gradient or

with net N cycling rates (all r2 , 0.3). Gross N

consumption was significantly greater than gross pro-

duction for NH4þ (P ¼ 0.001), but not for NO3

� (P ¼0.52), and gross nitrification was not related to gross

NH4þ consumption (P ¼ 0.86) (data not shown). The

ratio of net : gross N mineralization increased linearly

with surface soil %N (r2¼ 0.93, P , 0.001, n¼ 9), and it

approached 1 at high soil N (Fig. 2A). Similarly, the

ratio net : gross nitrification increased linearly with soil

%N (r2 ¼ 0.68, P , 0.01, n ¼ 9), although the value

anomalously exceeded 1 at the most N-rich site (site

number 39; Fig. 2A). This site also displayed the highest

net nitrification (64.6 kg N�ha�1�yr�1) and lowest gross

nitrification (9.7 kg N�ha�1�yr�1) of all sites. Excludingthis site from linear regression altered the slope, but the

relationship remained significant (r2¼ 0.58, P¼0.03, n¼8). Because soil %C and %N were positively related, the

ratios of net : gross mineralization and nitrification were

also positively related to soil %C (r2¼0.83 and r2¼0.90,

respectively, both P � 0.001, n ¼ 9; data not shown).

The mean residence time (MRT) of NH4þ and NO3

�

both increased linearly across the range of soil N

examined (Fig. 2B), with NO3� exhibiting lower MRT

(i.e., faster turnover) than NH4þ at low soil N (Table 2),

then shifting to NO3� exhibiting higher MRT (i.e.,

slower turnover) than NH4þ at high soil N (Fig. 2B).

Nitrogen concentration in live foliage ranged from

1.2% to 1.8% (Table 3) and was related logarithmically

to annual net N mineralization (r2 ¼ 0.72, P ¼ 0.003).

Nitrogen concentration of foliar litterfall ranged from

0.8% N to 1.4% N and was linearly related to foliar N

(r2 ¼ 0.73, P ¼ 0.003). Foliar N resorption efficiency

ranged from 19% to 45% (data not shown) and declined

linearly with foliar %N (r2 ¼ 0.82, P , 0.001) andFIG. 1. Net N cycling in relation to surface soil (0–10 cm)

%N for (A) annual net N mineralization and (B) nitrification ofnet mineralized N. In panel (A) the regression line is y¼ 127.3x� 19.0.

FIG. 2. Soil inorganic-N turnover in relation to surface soil(0–10 cm) %N for (A) the ratio of net : gross NH4

þ and NO3�

production and (B) mean residence times (MRT), in days, ofNH4

þ and NO3�. Linear regressions are (A) NH4

þ ¼ 1.99x �4.33, NO3

� ¼ 10.43x � 3.45; and (B) NH4þ ¼ 9.80x � 0.16,

NO3�¼ 39.09x� 12.40. Note that excluding the high outlier for

NO3� (site number 39) maintains significant regressions for

both panel (A) (r2¼0.56, P , 0.01, n¼8 sites) and panel (B) (r2

¼ 0.74, P , 0.01, n ¼ 8); these regression lines are not shown.


logarithmically with annual net N mineralization (r2 ¼0.78, P ¼ 0.002). Whole canopy foliage N translocation

ranged from 7.3 to 17.8 kg N�ha�1�yr�1 and also declined

logarithmically with net N mineralization (r2¼0.66, P ,

0.01). Aboveground plant N uptake ranged from 20 to

34 kg N�ha�1�yr�1 and increased in roughly linear

fashion with net N mineralization up to 50 kg

N�ha�1�yr�1, then leveled off at the site of greatest net

N mineralization (Fig. 3A). NUE (nitrogen-use efficien-

cy) of aboveground biomass did not vary systematically

with soil N or net N mineralization (r2 , 0.04), and had

a median value of 487 kg/kg across sites (range 356–942

kg/kg, n ¼ 9). Removal of site number 5 with relatively

high NUE did not clarify relationships with soil N or N

mineralization. Aboveground net primary production

(ANPP) per unit soil net N mineralization (i.e., resource

supply efficiency) declined rapidly at low to intermediate

soil N, then leveled off at higher soil N (Fig. 3B).

Total dissolved-N leaching loss at 100 cm depth

ranged from ,1 kg N�ha�1�yr�1 at low-N sites up to 29

kg N�ha�1�yr�1 at more N-rich sites. Dissolved organic

nitrogen (DON) dominated N loss in deep lysimeters at

low-N sites (Table 4), was unrelated to soil %N (r2 ¼0.01), and varied relatively little among sites (average

0.60 kg N�ha�1�yr�1, range 0.39–0.92 kg N�ha�1�yr�1, n¼ 9). Hydrologic N loss shifted to dominance by NO3

�

at intermediate-N and high-N sites, particularly when

net nitrification rates exceeded ;40 kg kg N�ha�1�yr�1,with NO3

� loss averaging 5.17 kg N�ha�1�yr�1 across allsites (range 0.05–28.33 kg N�ha�1�yr�1, n ¼ 9; Fig. 4A).

The percentage of N loss as NO3� varied from 6% to

97% due to large variation in NO3� loss and was linearly

related to nitrification for both 20 cm (r2 ¼ 0.74, P ¼0.003) and 100 cm (r2¼0.82, P , 0.001) lysimeters (Fig.

4B). Modeled water yield at 100 cm soil depth varied

twofold across sites (range 721–1486 mm/yr), yet

variation in NO3� leaching loss was driven almost

entirely (r2 ¼ 0.997) by fourfold variation in volume-

weighted mean NO3� concentrations across sites (range

1.5–1552 lg N/L; data not shown). N fluxes in 20 cm

and 100 cm deep lysimeters were significantly related

near 1:1 for NO3� (slope¼0.91, r2¼0.91, P , 0.001, n¼

9), but not for DON (r2¼ 0.06), which exhibited greater

concentrations at 20 cm than 100 cm depth. Fluxes of

NH4þ were low across all sites (average 0.17 kg N�

ha�1�yr�1, range 0.13–0.23 kg N�ha�1�yr�1, n ¼ 9).

TABLE 3. Vegetation biomass, productivity, and N relations at the nine study sites.

Site

Abovegroundbiomass(Mg/ha)

ANPP(Mg�ha�1�

yr�1)

Litterfall(Mg�ha�1�

yr�1)FoliarN (%)

LitterfallN (%)

LitterfallN return(kg N�

ha�1�yr�1)

N uptake(kg N�

ha�1�yr�1)NUEkg/kg

ANPP/net Nmineral-ization(Mg/kg)

7 59 8.9 2.34 1.31 1.04 24.3 25.0 356 1.1420 98 9.1 2.30 1.21 0.83 19.0 19.7 461 0.7976 126 14.9 2.76 1.60 1.09 30.0 30.9 483 0.605 163 21.5 2.08 1.42 1.01 21.0 22.8 942 0.9777 169 19.3 2.86 1.49 1.15 33.0 34.2 563 0.5358 121 17.0 2.67 1.46 1.15 30.7 31.5 540 0.4022 67 18.2 2.80 1.60 1.18 33.0 33.9 536 0.4016 122 14.3 2.13 1.75 1.43 30.3 33.7 425 0.2839 160 13.6 2.31 1.70 1.15 26.6 28.0 487 0.17

Notes: Sites are arranged in order of increasing soil %N (see Table 1). Aboveground net primary productivity (ANPP) wascalculated as the net annual aboveground increment in stems, branches, and foliage mass plus litterfall. Foliar %N was based onlive needles. Litterfall concentrations and fluxes were based on whole litter. N uptake is the total aboveground annual increment.NUE (nutrient-use efficiency) was calculated as ANPP per unit N uptake by vegetation.

FIG. 3. Plant N use along the gradient as (A) net annual Nuptake vs. soil net N mineralization, and (B) aboveground netprimary production (ANPP) per unit net N mineralization vs.surface soil (0–10 cm) %N. Regression lines are (A) y ¼�0.006x2 þ 0.648xþ 16.706, and (B) y ¼ 0.080(x� 0.235)�1.


DISCUSSION

Soil N mineralization is a key N cycling process that

influences the supply of N for forest productivity,

microbial nitrification, and the development of N

saturation in temperate forests. Variation in net N

mineralization among temperate forest stands is com-

monly attributed to effects of climate, soil type (usually

via texture), and tree species (usually via C:N) (Pastor et

al. 1984, Reich et al. 1997). Our study of nine Douglas-

fir forests in the Oregon Coast Range (USA) controlled

for potential tree-species effects, and our sites encom-

passed narrow ranges in temperature, precipitation, soil

texture, and soil C:N that were unrelated to variation in

net N mineralization (all r2 , 0.04). In the absence of

these important and potentially confounding factors, we

found that net N mineralization across our sites

increased linearly with surface soil %N (r2 ¼ 0.87; Fig.

1A). This wide range of soil N is diagnostic of strong

spatial variation in biogeochemical cycling across the

Oregon Coast Range, and is ultimately shaped by multi-

century interactions of wildfire disturbance and subse-

quent colonization by N2-fixing red alder that influence

long-term ecosystem N balances (Perakis et al. 2006;

Perakis et al., in press). The wide range of N

mineralization that we observed across a relatively small

geographic area is also noteworthy because Pacific

Northwest conifer forests are often characterized as N-

poor (Sollins et al. 1980, Fenn et al. 1998). Indeed, the

highest rates of annual net N mineralization that we

observed (82 kg N�ha�1�yr�1 ) in these unpolluted

coniferous forests are more commonly associated with

deciduous and/or highly N-polluted forests (e.g., Gil-

liam et al. 2001).

The question of what (if anything) sets limits to

positive feedbacks in plant–soil–microbial N cycling is

both of fundamental interest, and of increasing impor-

tance as humans accelerate N availability in many

ecosystems worldwide. The 10-fold variation we ob-

served in soil net N mineralization greatly exceeds the 3-

fold variation observed in soil N, and suggests that

internal cycling processes at our sites greatly amplify soil

N supply relative to substrate availability. However,

increases in soil N along our gradient did not contribute

indefinitely to positive feedbacks in plant–soil–microbial

N cycling, due to saturation of biotic N sinks as N

availability increased. Even though net N mineralization

increased linearly with soil N (Fig. 1A), we found that

ANPP (aboveground net primary production), plant N

uptake, and N return did not increase across the full

range of N mineralization (Table 2, Fig. 3A). Prescott et

al. (2000) reported a linear increase in litterfall N inputs

across a similarly constructed soil N gradient of

Douglas-fir forests in Oregon and Washington, but we

note that their N gradient spanned only half as high as

ours (up to ;12 vs. 22 Mg N/ha), across the range where

our data also show a linear increase in plant N uptake.

The progressive saturation of plant N demands at our

high-N sites is consistent with fertilization studies

showing that one-third of coastal Oregon Douglas-fir

forests are not N-limited (Peterson and Hazard 1990).

This lack of continued N uptake and recycling at high-N

sites sets a limit to feedbacks on ecosystem N cycling

that allow soil inorganic N to accumulate in excess and

become susceptible to NO3� leaching loss.

Microbial uptake of N is critical in regulating both

short- and long-term N retention in N-poor forest soils.

Gradient studies of experimental and anthropogenic N

TABLE 4. Fluxes of water and N, and %NO3� of total N flux, from lysimeters at 20 cm and 100 cm

depths at the nine sites.

SiteWater(cm)

Nitrogen fluxes (kg N�ha�1�yr�1)NO3

�

(% of total N)NH4þ NO3

� DON

Lysimeters at 20 cm depth

7 123 0.23 0.05 1.07 320 177 0.19 0.53 1.52 2476 117 0.21 0.15 3.19 45 119 0.17 0.03 1.23 277 107 0.19 2.40 1.24 6358 120 0.22 12.50 3.83 7622 101 0.18 1.26 0.99 5216 114 0.21 21.20 2.13 9039 150 0.25 9.76 0.90 89

Lysimeters at 100 cm depth

7 94 0.17 0.07 0.50 1020 148 0.20 0.05 0.66 576 88 0.23 0.23 0.39 275 90 0.13 0.05 0.69 677 78 0.15 1.50 0.92 5858 92 0.13 9.66 0.66 9222 72 0.13 0.92 0.46 6116 85 0.16 28.33 0.49 9839 122 0.21 5.76 0.67 87

Note: Sites are arranged in order of increasing soil %N (see Table 1.)


addition in temperate forests suggest that microbial

regulation of N cycling declines at high N supply (Perakis

et al. 2005, Corre et al. 2007), and our results extend those

findings to natural soil N gradients. Gross N mineraliza-

tion did not exhibit a clear trend across our soil N

gradient, yet net N mineralization steadily increased,

highlighting a progressive saturation of microbial N

uptake at our most N-rich sites. Because microbial N

uptake is fueled by available C, and plants supply C to

soils, both plant and soil factors can be important in

regulating saturation of microbial N demands. Across

ecosystems worldwide, microbial N uptake is related to

soil C and C:N (Booth et al. 2005), but soil C:N varied

little across our sites. Total soil C did increase with soil N

at our sites, but high N under Douglas-fir can reduce soil

C availability to heterotrophic microbes (Swanston et al.

2004) and increase soil bacterial : fungal ratios (Boyle et al.

2008), both of which would reduce microbial N immobi-

lization. Microbial processing may also differ for C and N

(Schimel and Weintraub 2003) and promote high N

availability with declining C turnover, as occurs during

decomposition of Douglas-fir litter, which shows slower

mass loss yet faster N release when litter has high initial N

concentrations (Matkins 2009). High soil N availability

under Douglas-fir also decreases C allocation to fine-root

turnover (Vogt et al. 1987), and in other conifers reduces

C transfer to mycorrhizae (Hogberg et al. 2010). In this

way, the decline we observed in ANPP per unit net N

mineralization with increasing soil N (Fig. 4B) may signal

a general reduction in plant C inputs relative to soil N

supply at N-rich sites. This range of factors may explain

why soil net N mineralization continues to accelerate even

as plant N uptake levels off at our most N-rich sites,

leading to excess inorganic-N availability. Our finding

that soil N mineralization exceeds plant N uptake

contrasts with theoretical predictions of steady-state N

saturation (Agren and Bosatta 1988), and raises the

possibility that ecosystem N dynamics are far from

equilibrium at our most N-rich sites. Our results do,

however, broadly support the related theoretical predic-

tion that soil microbial N uptake saturates before plant

demands, so that saturation of plant N demands

ultimately controls the onset of system-level N saturation.

Nitrogen leaching losses across our sites shifted from

dominance by DON at low-N sites to NO3� at high-N

sites, primarily due to increases in NO3�. The domi-

nance of DON at low-N sites is characteristic of other

N-poor unpolluted temperate forests (Sollins et al. 1980,

Perakis and Hedin 2002). In contrast to previous studies,

however, DON loss rates were not related to site N

status, C:N, nor to NO3� leaching (Compton et al. 2003,

Brookshire et al. 2007). It is possible that high within-

site variability in our lysimeter data may obscure

patterns that emerge at the scale of entire watersheds.

Regardless, the switch we observed to NO3� dominance

at more N-rich sites is characteristic of an accelerated N

cycle (Johnson and Lindberg 1992). At low soil N the

mean residence time of NO3� in soil was less than NH4

þ,

implying tighter cycling of NO3� than NH4

þ at low-N

sites, but this pattern reversed at high soil N, suggesting

declining biotic control of NO3�. Yet, NO3

� loss

remained relatively low (below 0.5 kg NO3�-

N�ha�1�yr�1) until nitrification rates exceeded ;38 kg

N�ha�1�yr�1. This threshold increase in hydrologic NO3�

loss coincides with the maximum measured plant N

uptake of ;34 kg N�ha�1�yr�1, and reinforces the idea

that substantial NO3� loss occurs only when plant N

demands are exceeded.

Using published information on N inputs and

measured N losses, we can compare ecosystem N

input–output budgets for low-N vs. high-N sites across

our gradient. Total N inputs to our sites are ;3 kg

N�ha�1�yr�1, which is the sum of inorganic wet N

deposition (0.65 kg N�ha�1�yr�1; National Acid Depo-

sition Program: data for site OR02, 1980–2004, available

online),4 doubling wet deposition to account for fog

(Issac 1946, Bormann et al. 1989), adding a 30%contribution of organic N to total deposition (Neff et

al. 2002), plus modest asymbiotic N2-fixation (1 kg

FIG. 4. Annual nitrate leaching in relation to net nitrifica-tion rates for (A) nitrate-N fluxes in shallow and deeplysimeters and (B) percentage of total hydrologic N leachingas nitrate. Lysimeters at 20 cm depth are represented with opensymbols and dashed lines, and lysimeters at 100 cm depth arerepresented with solid symbols and solid lines. Regression linesshown are: (A) shallow lysimeters (y ¼ 0.063e0.088x) and deeplysimeters (y ¼ 0.039e0.096x); and (B) shallow lysimeters (y ¼1.46x� 0.63) and deep lysimeters (y ¼ 1.58xþ 0.20).

4 hhttp://nadp.sws.uiuc.edu/nadpdata/ntnsites.aspi


N�ha�1�yr�1, Heath et al. 1988). (Epiphytic N2 fixation is

negligible until forests reach ;250 years of age; McCune

1993). Total measured N losses from our four low-N

sites where DON dominates are ;1.3 kg N�ha�1�yr�1,which is the sum of hydrologic N loss (0.84 kg

N�ha�1�yr�1; Table 4) and soil N2O þ NO gas fluxes

(0.5 kg N�ha�1�yr�1; H. E. Erickson and S. S. Perakis,

unpublished manuscript). Thus, the overall N budget of

our low-N sites displays only a small imbalance between

known N inputs (3 kg N�ha�1�yr�1) and losses (1.3 kg

N�ha�1�yr�1), so that even small quantities of unmea-

sured N2 gas losses could balance the N budget (e.g.,

Houlton et al. 2006). Alternatively, these low-N sites

could slowly be accumulating N, based on their

relatively low soil N capital (8608–13 647 kg N/ha)

compared to our high-N sites (up to 22 379 kg N/ha),

and by the high capacity for N retention in low-N

Douglas-fir forests (Flint et al. 2008). The N budget of

our high-N sites, by contrast, displays large annual net

N loss, with inputs (;3 kg N�ha�1�yr�1) less than

outputs (10–28 kg N�ha�1�yr�1), where any unmeasured

N2 losses would only intensify this imbalance. Thus,

while N budgets are in approximate balance at our low-

N sites, they are in strong disequilibrium at our high-N

sites.

Nitrogen balances in forests are often close to

equilibrium, or slowly aggrading, with relatively small

N inputs and losses compared to high rates of internal

recycling. With the exception of very recently disturbed

forests (Vitousek and Reiners 1975), reports of signif-

icant net N loss from temperate forests are rare

(Gunderson et al. 1998, Parfitt et al. 2002). It is thought

that constraints on biological N2-fixation and low N

inputs, combined with persistent losses of availability-

independent forms of N that are not under direct biotic

control (e.g., DON and trace N gases), maintain N

scarcity and thus favor N retention in most temperate

forests (Hedin et al. 1995, Vitousek et al. 1998).

However, the wide range of ecosystem N retention that

we observed suggests that availability-independent

pathways of N loss are not always sufficient to maintain

N limitation in temperate forests. Given the high nitrate

mobility we observed (indicated by near 1:1 correlation

of hydrologic NO3� fluxes in 20 cm vs. 100 cm

lysimeters), it is possible that net N losses will continue

from our high-N sites until N capital and availability

decline to match internal sinks. Yet, even a millennium

of elevated NO3� loss at our most N-rich site

(cumulative loss of ;5800 kg N/ha) could be easily

replaced if the site was disturbed and colonized by

symbiotic N2-fixing red alder, which is capable of adding

5000–15 000 kg N/ha (100–200 kg�ha�1�yr�1; Binkley et

al. 1994) over the lifetime of a 50–75 year-old stand.

Long-term N and d15N mass balance modeling at our

sites (Perakis et al., in press) suggests that episodic N

inputs from red alder are necessary to account for the

high N capital of Oregon Coast Range forests, and may

also contribute to nonequilibrium internal N dynamics

and whole-system N imbalances. The high soil N accrual

capacity of these forests may also be related to their

Andic properties (Batjes 1996), but Andic soils alone are

not a sufficient condition for high N accumulation, as

Andic forest soils elsewhere in Oregon display only

modest N capital similar to most other temperate forests

worldwide (e.g., the H. J. Andrews site in Oregon, USA,

has a value of 5000 kg N/ha [Sollins et al. 1980]

compared to 2500–14 000 kg N/ha for temperate forests

elsewhere [Cole and Rapp 1981]). The disequilibrium in

N balances of our most N-rich sites may be analogous to

high N losses observed from some tropical forests

(Hedin et al. 2009). It remains unknown, however,

whether Pacific Northwest forests growing on N-rich

soils can sustain high N losses over the long-term, given

the competing influences of N uptake in growing

biomass and detritus vs. increased N inputs from

epiphytic N2-fixation (;5 kg N�ha�1�yr�1; Sollins et al.

1980) as forests age.

In conclusion, our results illustrate that some unpol-

luted temperate forests can naturally accumulate suffi-

cient soil N to foster excess N availability and

disequilibrium in ecosystem N dynamics, ultimately

leading to elevated NO3� loss characteristic of N

saturation. Our data are also consistent with the idea

that microbial N uptake saturates before plant uptake,

and that N supply in excess of plant N demands

ultimately controls the onset of system-level N satura-

tion (e.g., Agren and Bosatta 1988, Aber et al. 1998).

Our finding that net N mineralization increased linearly

beyond the point of plant and ecosystem demands for N

differs from N-saturation theory which predicts declin-

ing mineralization at high N, and suggests limits to the

application of theories derived from N-polluted ecosys-

tems for understanding N dynamics of naturally N-rich

forests. To date, atmospheric N deposition has not

generally increased soil N accumulation to levels

associated with N inputs from legacies of N2-fixing tree

species (Nave et al. 2009), and experimental reductions

in atmospheric N inputs usually reverse symptoms of N

saturation within several years (Bredemeier et al. 1998,

Corre and Lamersdorf 2004). If further N inputs

continue to promote N accumulation and availability

in temperate forests, however, our results raise the

possibility that N saturation may persist even when

inputs are reduced.

ACKNOWLEDGMENTS

We thank Chris Catricala and Melissa McCartney for fieldand laboratory assistance, Kermit Cromack and HeatherErickson for discussions, and Dan Binkley, David Myrold,and two anonymous reviewers for helpful comments on themanuscript. This research was supported by NSF DEB-0346837. Any use of trade names is for descriptive purposesonly and does not imply endorsement by the U.S. Government.

LITERATURE CITED

Aber, J., W. McDowell, K. Nadelhoffer, A. Magill, G. M.Berntson, M. Kamakea, S. McNulty, W. Currie, L. Rustad,and I. Fernandez. 1998. Nitrogen saturation in temperate


forest ecosystems: hypotheses revisited. BioScience 48:921–934.

Aber, J., K. Nadelhoffer, P. A. Steudler, and J. M. Melillo.1989. Nitrogen saturation in northern forest ecosystems.BioScience 39:378–86.

Agren, G. I., and E. Bosatta. 1988. Nitrogen saturation ofterrestrial ecosystems. Environmental Pollution 54:185–197.

Batjes, N. H. 1996. Total carbon and nitrogen in the soils of theworld. European Journal of Soil Science 47:151–163.

Binkley, D., K. Cromack, Jr. and D. D. Baker. 1994. Nitrogenfixation by red alder: biology, rates, and controls. Pages 57–72 in D. E. Hibbs, D. S. DeBell, and R. F. Tarrant, editors.The biology and management of red alder. Oregon StateUniversity Press, Corvallis, Oregon, USA.

Binkley, D., J. L. Stape, and M. G. Ryan. 2004. Thinking aboutefficiency of resource use in forests. Forest Ecology andManagement 193:5–17.

Booth, M. S., J. M. Stark, and E. Rastetter. 2005. Controls onnitrogen cycling in terrestrial ecosystems: a synthetic analysisof literature data. Ecological Monographs 75:139–157.

Bormann, B. T., R. F. Tarrant, M. H. McClellan, and T.Savage. 1989. Chemistry of rainwater and cloudwater atremote sites in Alaska and Oregon. Journal of EnvironmentalQuality 18:149–152.

Boyle, S. A., R. R. Yarwood, P. J. Bottomley, and D. D.Myrold. 2008. Bacterial and fungal contributions to soilnitrogen cycling under Douglas-fir and red alder at two sitesin Oregon. Soil Biology and Biochemistry 40:443–451.

Bredemeier, M., K. Blank, Y. J. Xu, A. Tietema, A. W.Boxman, B. Emmett, F. Moldan, P. Gundersen, P. Schleppi,and R. F. Wright. 1998. Input–output budgets at theNITREX sites. Forest Ecology and Management 101:57–64.

Brookshire, E. N. J., H. M. Valett, S. A. Thomas, and J. R.Webster. 2007. Atmospheric N deposition increases organicN losses from temperate forests. Ecosystems 10:252–262.

Chapman, S. K., J. A. Langley, S. C. Hart, and G. W. Koch.2006. Plants actively control nitrogen cycling: uncorking themicrobial bottleneck. New Phytologist 169:27–34.

Cole, D. W., and M. Rapp. 1981. Elemental cycling in forestecosystems. Pages 341–409 in D. E. Reichle, editor. Dynamicproperties of forest ecosystems. Cambridge University Press,London, UK.

Compton, J. E., M. R. Church, S. T. Larned, and W. E.Hogsett. 2003. Nitrogen export from forested watersheds inthe Oregon Coast Range: the role of N2-fixing red alder.Ecosystems 6:773–785.

Corre, M. D., R. Brumme, E. Veldkamp, and F. O. Beese. 2007.Changes in nitrogen cycling and retention processes in soilsunder spruce forests along a nitrogen enrichment gradient inGermany. Global Change Biology 13:1509–1527.

Corre, M. D., and N. P. Lamersdorf. 2004. Reversal of nitrogensaturation after long-term reduction of nitrogen deposition:changes in the soil nitrogen cycle. Ecology 85:3090–3104.

Dahlgren, R. A. 1994. Soil acidification and nitrogen saturationfrom weathering of ammonium-bearing rock. Nature368:838–841.

Daly, C., R. P. Neilson, and D. L. Phillips. 1994. A statistical-topographical model for mapping climatological precipita-tion over mountainous terrain. Journal of Applied Meteo-rology 33:140–158.

Davidson, E. A., S. C. Hart, and M. K. Firestone. 1992.Internal cycling of nitrate in soils of a mature coniferousforest. Ecology 73:1148–1156.

Fenn, M. E., M. A. Poth, J. D. Aber, J. S. Baron, B. T.Bormann, D. W. Johnson, A. D. Lemly, S. G. McNulty,D. F. Ryan, and R. Stottlemyer. 1998. Nitrogen excess inNorth American ecosystems: predisposing factors, ecosystemresponses, and management strategies. Ecological Applica-tions 8:706–733.

Flint, C. M., R. B. Harrison, B. D. Strahm, and A. B. Adams.2008. Nitrogen leaching from Douglas-fir forests after urea

fertilization. Journal of Environmental Quality 37:1781–1788.

Gilliam, F. S., B. M. Yurish, and M. B. Adams. 2001. Temporaland spatial variation of nitrogen transformations in nitrogen-saturated soils of a central Appalachian hardwood forest.Canadian Journal of Forest Research 31:1768–1785.

Gosz, J. R. 1981. Nitrogen cycling in coniferous ecosystems.Pages 405–426 in F. E. Clark, and T. Rosswall, editors.Terrestrial nitrogen cycles. Ecological Bulletin (Stockholm) 33.

Gunderson, P., B. A. Emmett, O. J. Kjonaas, C. J. Koopmans,and A. Tietema. 1998. Impact of nitrogen deposition onnitrogen cycling in forests: a synthesis of the NITREX data.Forest Ecology and Management 101:37–55.

Hart, S. C., J. M. Stark, E. A. Davidson, and M. K. Firestone.1994. Nitrogen mineralization, immobilization, and nitrifica-tion. Pages 985–1018 in R. W. Weaver, J. S. Angle, and P. J.Bottomley editors. Method of soil analysis. Part 2. Micro-biological and biochemical properties. Soil Science Society ofAmerica, Madison, Wisconsin, USA.

Heath, B., P. Sollins, D. A. Perry, and K. Cromack, Jr. 1988.Asymbiotic nitrogen fixation in litter from Pacific Northwestforests. Canadian Journal of Forest Research 18:72–78.

Hedin, L. O., J. J. Armesto, and A. H. Johnson. 1995. Patternsof nutrient loss from unpolluted, old-growth temperateforests: evaluation of biogeochemical theory. Ecology76:493–509.

Hedin, L. O., E. N. J. Brookshire, D. N. L. Menge, and A. R.Barron. 2009. The nitrogen paradox in tropical forestsecosystems. Annual Review of Ecology, Evolution andSystematics 40:613–35.

Hogberg, M. N., M. J. I. Briones, S. G. Keel, D. B. Metcalfe, C.Campbell, A. J. Midwood, B. Thornton, V. Hurry, S. Linder,T. Nasholm, and P. Hogberg. 2010. Quantification of effectsof season and nitrogen supply on tree below-ground carbontransfer to ectomycorrhizal fungi and other soil organisms ina boreal pine forest. New Phytologist 187(2):485–93.

Houlton, B. Z., D. M. Sigman, and L. O. Hedin. 2006. Isotopicevidence for large gaseous nitrogen losses from tropicalrainforests. Proceedings of the National Academy ofSciences, USA 103:8745–8750.

Isaac, L. A. 1946. Fog drip and rain interception in coastalforests. Pages 15–16 in USDA Forest Service Research Notenumber 34. Pacific Northwest Forest and Range ExperimentStation, Portland, Oregon, USA.

Johnson, D. W., and S. E. Lindberg. 1992. AtmosphericDeposition and Forest Nutrient Cycling. Ecological Series91, Springer-Verlag, New York, New York, USA.

Kirkham, D., and W. V. Bartholomew. 1954. Equations forfollowing nutrient transformations in soil, utilizing tracerdata. Soil Science Society of America Proceedings 18:33–34.

Lansberg, J. J., and R. H. Waring. 1997. A generalized model offorest productivity using simplified concepts of radiation-useefficiency, carbon balance and partitioning. Forest Ecologyand Management 95:209.

LeBauer, D. S., and K. K. Treseder. 2008. Nitrogen limitationof net primary productivity in terrestrial ecosystems isglobally distributed. Ecology 89:371–379.

Lee, E. H., D. T. Tingey, P. A. Beedlow, M. G. Johnson, andC. A. Burdick. 2007. Relating fine root biomass to soil andclimate conditions in the Pacific Northwest. Forest Ecologyand Management 242:195–208.

Maguire, D. A., A. Kanaskie, W. Voelker, R. Johnson, and G.Johnson. 2002. Growth of young Douglas-fir plantationsacross a gradient in Swiss Needle Cast severity. WesternJournal of Applied Forestry 17:86–95.

Matkins, J. 2009. Decomposition and nitrogen dynamics of redalder and Douglas-fir leaf litter in Oregon Coast Rangeriparian forests. Thesis. Oregon State University, Corvallis,Oregon, USA.

McCulley, R. L., I. C. Burke, and W. K. Lauenroth. 2009.Conservation of nitrogen increases with precipitation across


a major grassland gradient in the Central Great Plains ofNorth America. Oecologia 159:571–581.

McCune, B. 1993. Gradients in epiphyte biomass in threePseudotsuga–Tsuga forests of different ages in westernOregon and Washington. Bryologist 96:405–411.

Nadelhoffer, K. J., J. D. Aber, and J. M. Melillo. 1983. Leaf-litter production and soil organic matter dynamics along anitrogen availability gradient in southern Wisconsin (USA).Canadian Journal of Forest Research 13:12–21.

Nave, L. E., E. D. Vance, C. W. Swanston, and P. S. Curtis.2009. Impacts of elevated N inputs on north temperate forestsoil C storage, C/N, and net N-mineralization. Geoderma153:231–240.

Neff, J. C., E. A. Holland, F. J. Dentener, W. H. McDowell,and K. M. Russell. 2002. The origin, composition and ratesof organic nitrogen deposition: a missing piece of thenitrogen cycle? Biogeochemistry 57/58:99–136.

Parfitt, R. L., G. J. Salt, and L. F. Hill. 2002. Clear-cuttingreduces nitrate leaching in a pine plantation of high naturalN status. Forest Ecology and Management 170:43–53.

Pastor, J., J. D. Aber, C. A. McClaugherty, and J. M. Melillo.1984. Aboveground production and N and P cycling along anitrogen mineralization gradient on Blackhawk Island,Wisconsin. Ecology 65:256–268.

Perakis, S. S., J. E. Compton, and L. O. Hedin. 2005. Nitrogenretention across a gradient of 15N additions to an unpollutedtemperate forest soil in Chile. Ecology 86:96–105.

Perakis, S. S., and L. O. Hedin. 2002. Nitrogen loss fromunpolluted South American forests mainly via dissolvedorganic compounds. Nature 415:416–419.

Perakis, S. S., D. A. Maguire, T. D. Bullen, K. Cromack, R. H.Waring, and J. R. Boyle. 2006. Coupled nitrogen and calciumcycles in the forests of the Oregon Coast Range. Ecosystems9:63–74.

Perakis, S. S., E. R. Sinkhorn, and J. E. Compton. In press.d15N constraints on long-term nitrogen balances in temperateforests. Oecologia.

Peterson, C. E., and J. W. Hazard. 1990. Regional variation ingrowth response of coastal Douglas-fir to nitrogen fertilizerin the Pacific Northwest. Forest Science 36:625–640.

Prescott, C. E., H. N. Chappell, and L. Vesterdal. 2000.Nitrogen turnover in forest floors of coastal Douglas-fir atsites differing in soil nitrogen capital. Ecology 81:1878–1886.

Reich, P. B., D. F. Grigal, J. D. Aber, and S. T. Gower. 1997.Nitrogen mineralization and productivity in 50 hardwoodand conifer stands on diverse soils. Ecology 78:335–347.

Schimel, J. P., and J. Bennett. 2004. Nitrogen mineralization:challenges of a changing paradigm. Ecology 85:591–602.

Schimel, J. P., and M. N. Weintraub. 2003. The implications ofexoenzyme activity on microbial carbon and nitrogenlimitation in soil: a theoretical model. Soil Biology andBiochemistry 35:549–563.

Sollins, P., C. C. Grier, F. M. McCorison, K. J. Cromack, R.Fogel, and R. L. Fredriksen. 1980. The internal elementcycles of an old-growth Douglas-fir ecosystem in westernOregon. Ecological Monographs 50:261–285.

Stark, J. M., and S. C. Hart. 1996. Diffusion technique forpreparing salt solutions, Kjeldahl digest, and persulfatedigests for nitrogen-15 analysis. Soil Science Society ofAmerica Journal 60:1846–1855.

Swanston, C. W., P. S. Homann, B. A. Caldwell, D. D. Myrold,L. Ganio, and P. Sollins. 2004. Long-term effects of elevatednitrogen on forest soil organic matter stability. Biogeochem-istry 70:227–250.

SYSTAT Software. 2004. SYSTAT version 11. SYSTATSoftware, Richmond, California, USA.

Vitousek, P. M. 1982. Nutrient cycling and nutrient useefficiency. American Naturalist 119:553–572.

Vitousek, P. M., L. O. Hedin, P. A. Matson, J. H. Fownes, andJ. C. Neff. 1998. Within-system element cycles, input-outputbudgets, and nutrient limitation. Pages 432–451 in M. L.Pace and P. M. Groffman, editors. Successes, limitations, andfrontiers in ecosystem science. Springer-Verlag, New York,New York, USA.

Vitousek, P. M., and W. A. Reiners. 1975. Ecosystemsuccession and nutrient retention: a hypothesis. BioScience25:376–81.

Vogt, K. A., D. J. Vogt, E. E. Moore, B. A. Fataga, M. R.Redlin, and R. L. Edmonds. 1987. Conifer and angiospermfine-root biomass in relation to stand age and site produc-tivity in Douglas-fir forests. Journal of Ecology 75:857–870.

Waring, R. H., and N. McDowell. 2002. Use of a physiologicalprocess model with forestry yield tables to set limits onannual carbon balances. Tree Physiology 22:179–188.

Waring, R. H., K. S. Milner, W. M. Jolly, L. Phillips, and D.McWethy. 2006. Assessment of site index and forest growthcapacity across the Pacific and Inland Northwest U.S.A. witha MODIS satellite-derived vegetation index. Forest Ecologyand Management 228:285–291.

Weiskittel, A. R., D. A. Maguire, R. A. Monserud, and G. P.Johnson. 2010. A hybrid model for intensively managedDouglas-fir plantations in the Pacific Northwest, USA.European Journal of Forest Research 129:325–338.


Documents

Biogeochemistry of a temperate forest nitrogen gradient