1
2
3
R.Q. Thomas, S. Zaehle, P.H. Templer, and C.L. Goodale. Global patterns of nitrogen limitation:
confronting two global biogeochemical models with observations.
4
Supplemental Information
5
6
7
SI.1. CLM-CN model
The CLM-CN 4.0 model (Thornton et al. 2007, Thornton et al. 2009, Lawrence et al.
8
2011) is the land surface model within the Community Earth System Model (release 1.0; Gent et
9
al. 2011). The model uses fully prognostic terrestrial C and N cycles calculated on a 30-minute
10
time step. N and C are cycled through the following pools: 1) three litter pools, based on the
11
chemical composition of the inputs (e.g., labile, cellulose, lignin pools), with stoichiometry that
12
varies with the stoichiometry of litter inputs and different decomposition rates, 2) one coarse
13
woody debris pool with constant stoichiometry, 3) four soil organic matter pools with differing
14
decomposition rates and with C:N ratios that vary by pool but are constant over time, and 4) six
15
vegetation pools (leaves, live stem, dead stem, live coarse roots, dead coarse roots, and fine
16
roots) with time-invariant C:N ratios that differ by pool and among plant functional types. All
17
plant functional types within a grid cell share the same soil environment. Plant uptake of N
18
depends directly on the demand set by gross primary productivity (GPP) in the limits of the soil
19
mineral N available to plants. Microbial uptake of N is a function of decomposition rate, C use
20
efficiency, and the difference between the C:N ratio of the donor and receiving pool, based on
21
the Biome-BGC model (Thornton et al. 2002, Thornton and Rosenbloom 2005). When the
22
demand for N by both the microbes and vegetation exceeds the available inorganic N pool, both
23
GPP and decomposition rates are reduced in proportion to their N demand relative to total N
24
demand so that total N demand matches available N. All C and N uptake and competition occurs
1
25
at the 30-minute time step. Allocation of C and N among plant tissues is based on fixed
26
allocation ratios, with one exception: the ratio of stem allocation to leaf allocation increases with
27
NPP. Leaf area is determined through the balance between C allocation and turnover through
28
litterfall or fire. N inputs into the CLM-CN include N deposition and N fixation. N fixation is
29
calculated as a saturating relationship with annual NPP, based on Cleveland et al. (1999) and
30
varies over time and space. N outputs include hydrologic N leaching, N gas production
31
calculated as a fixed proportion of net mineralization, N gas production when N availability
32
exceeds plant and microbial demand, N volatilization by fire, and N removal during harvesting.
33
In this study we use the most recently publicly released version (v. 4.0; Lawrence et al. 2011)
34
that corresponds to the version used in published studies that report on C-N interactions
35
(Thornton et al. 2009, Bonan and Levis 2010) and CMIP 5 climate model intercomparison
36
(Taylor et al. 2012).
37
38
39
SI.2 O-CN model
The O-CN model (Zaehle and Friend 2010, Zaehle et al. 2010, Zaehle et al. 2011) has
40
been derived from the ORCHIDEE model (Krinner et al. 2005) and like CLM-CN has fully
41
prognostic C and N cycles calculated on a 30-minute time step. In the O-CN model, C and N
42
cycle through the following pools: 1) four litter pools (above- and below-ground metabolic and
43
structural), 2) two coarse woody debris pools, 3) four soil pools (surface, active, slow, and
44
passive), all with different turnover times, and 4) nine vegetation pools. The C-N ratio in the
45
vegetation depends on the balance of labile C to labile N, and all of the C:N ratios in the
46
vegetation pools, except the labile and reserve pools, vary in proportion to variations in foliar
47
C:N ratios. Foliar C:N ratios determine how GPP is influenced by N availability (i.e., increasing
2
48
GPP with increasing foliar N). The allocation of C and N to plant tissues varies in the O-CN as a
49
function of the labile C and N ratios, with the allocation to fine roots increasing with N stress.
50
The allocation to leaves depends on the pipe-model theory (Shinozaki et al. 1964, Zaehle et al.
51
2006), resulting in a theoretical maximum leaf area index (LAI) based on the light environment.
52
N uptake by vegetation is a function of the availability of inorganic N, root C, plant N demand,
53
and the abiotic environment. As labile N builds up relative to labile C, root uptake of N is
54
reduced. Unlike the CLM-CN, each plant functional type within a grid cell has a unique soil
55
environment in the O-CN. The soil organic dynamics are based on the Century approach (Parton
56
et al. 1993). The microbial N uptake depends on the decomposition rate, C:N ratio of the donor
57
and receiving pools, and inorganic N availability. Soil organic matter has a flexible
58
stoichiometry that depends on inorganic N concentrations in the soil, whereas litter stoichiometry
59
is dependent on the C:N ratio of the source material. When plant and microbial demand for N
60
exceeds inorganic N availability, N is first allocated to microbes, and plants receive the
61
remaining N in excess of microbial demand. N inputs into the O-CN include N deposition and N
62
fixation, with the latter a prescribed input that varies over space but not time and is based on the
63
relationship between evapotranspiration and N fixation (Cleveland et al. 1999). N export
64
includes N leaching, gaseous emissions, and harvesting. Gaseous emissions are based on the
65
LPJ-DyN simplification of the DNDC model (Xu-Ri and Prentice 2008) and depend on the
66
availability of nitrate, soil organic C, and soil water (Zaehle et al. 2011).
67
68
S.3 N fertilization response in grassland ecosystems
69
We used empirical observations of aboveground net primary production responses to N
70
additions in grasslands. The N addition rates ranged from 2.5 - 57.2 g N m-2 yr-1 over 1-6 years
3
71
in 39 the grassland studies obtained from the meta-analysis by LeBauer and Treseder (2008).
72
We used the same model-data comparison methods as described in section 2.2.1 of the main text.
73
The observed rates of ANPP increased on average by 60 ± 13% in the 39 grassland
74
studies (Supplemental Information Fig 1). The CLM-CN simulated a larger but not statistically
75
different response in ANPP response to N fertilization experiments in grasslands (94 ± 16%, p =
76
0.14). In contrast to the CLM-CN, the O-CN model simulated a significantly lower ANPP
77
response to N fertilization than observed in grassland (20 ± 10%; p = 0.02).
78
79
S3. References
80
Bonan GB, and Levis S (2010) Quantifying carbon-nitrogen feedbacks in the Community Land
81
82
83
84
85
86
Model (CLM4). Geophysical Research Letters 37:L07401.
Cleveland C, Townsend A, Schimel DS et al (1999) Global patterns of terrestrial biological
nitrogen (N2) fixation in natural ecosystems. Global Biogeochemical Cycles 13:623–645.
Gent PR, Danabasoglu G, Donner LJ et al. (2011) The Community Climate System Model
Version 4. Journal of Climate 24:4973–4991.
Krinner G, Viovy N, de Noblet-Ducoudre N et al. (2005) A dynamic global vegetation model for
87
studies of the coupled atmosphere-biosphere system. Global Biogeochemical Cycles
88
19:GB1015.
89
Lawrence DM, Oleson K, Flanner MG et al. (2011) Parameterization improvements and
90
functional and structural advances in Version 4 of the Community Land Model. Journal of
91
Advances in Modeling Earth Systems 3:M03001.
92
93
LeBauer DS and Treseder KK (2008) Nitrogen limitation of net primary productivity in
terrestrial ecosystems is globally distributed. Ecology 89:371–379.
4
94
Parton W, Scurlock J, Ojima D et al. (1993) Observations and modeling of biomass and soil
95
organic matter dynamics for the grassland biome worldwide. Global Biogeochemical Cycles
96
7:785–809.
97
98
Taylor KE, Stouffer RJ, and Meehl GA (2012) An Overview of CMIP5 and the Experiment
Design. Bulletin of the American Meteorological Society 93: 485–498.
99
Thornton PE and Rosenbloom NA (2005) Ecosystem model spin-up: Estimating steady state
100
conditions in a coupled terrestrial carbon and nitrogen cycle model. Ecological Modeling
101
189:25–48.
102
Thornton PE, Doney SC, Lindsay K et al (2009) Carbon-nitrogen interactions regulate climate-
103
carbon cycle feedbacks: results from an atmosphere-ocean general circulation model.
104
Biogeosciences 6:2099–2120.
105
Thornton PE, Lamarque J-F, Rosenbloom NA, and Mahowald NM (2007) Influence of carbon-
106
nitrogen cycle coupling on land model response to CO2 fertilization and climate variability.
107
Global Biogeochemical Cycles 21:GB4018.
108
Thornton PE, Law BE, Gholz H et al (2002) Modeling and measuring the effects of disturbance
109
history and climate on carbon and water budgets in evergreen needleleaf forests.
110
Agricultural and Forest Meteorology 113:185–222.
111
112
113
114
115
116
Shinozaki K, Yoda K, Hozumi K, and Kira T (1964) A quantitative analysis of the plant form the pipe model theory. Japanese Journal of Ecology 13:98–104.
Zaehle S, Ciais P, Friend AD, and Prieur V (2011) Carbon benefits of anthropogenic reactive
nitrogen offset by nitrous oxide emissions. Nature Geoscience 4:601–605.
Zaehle S and Friend AD (2010) Carbon and nitrogen cycle dynamics in the O-CN land surface
model: 1. Model description, site-scale evaluation, and sensitivity to parameter estimates.
5
117
118
Global Biogeochemical Cycles 24:GB1005.
Zaehle S, Friend AD, Friedlingstein P, Dentener F, Peylin P, and Schulz M (2010) Carbon and
119
nitrogen cycle dynamics in the O-CN land surface model: 2. Role of the nitrogen cycle in the
120
historical terrestrial carbon balance. Global Biogeochemical Cycles 24:GB1006.
121
122
123
124
Zaehle S, Sitch S, Prentice IC et al. (2006) The importance of age-related decline in forest NPP
for modeling regional carbon balances. Ecological Applications 16:1555–1574.
Xu-Ri and Prentice IC (2008) Terrestrial nitrogen cycle simulation with a dynamic global
vegetation model. Global Change Biology 14: 1745–1764
125
126
S4 References for forest nitrogen fertilization experiments
127
128
Adamek M Corre MD, and Hölscher D (2009) Early effect of elevated nitrogen
129
input on above-ground net primary production of a lower montane rain forest, Panama.
130
Journal of Tropical Ecology 25:637-647.
131
Campo J and Vazquez-Yanes C (2004) Effects of nutrient limitation on aboveground carbon
132
dynamics during tropical dry forest regeneration in Yucatan, Mexico. Ecosystems 7:311–
133
319.
134
Cusack DF, Silver WL, Torn MS, and McDowell WH (2010) Effects of nitrogen additions on
135
above- and belowground carbon dynamics in two tropical forests. Biogeochemistry 104:203–
136
225.
137
Davidson EA, de Carvalho CJR, Vieira I et al. (2004) Nitrogen and phosphorus limitation of
138
biomass growth in a tropical secondary forest. Ecological Applications 14:S150–S163.
139
DeWalle DR, Kochenderfer JN, Adams MB et al. (2006) Vegetation and Acidification. Pages
140
137–188 in Adams MB, DeWalle DR, and Hom JL, editors. The Fernow Watershed
6
141
142
Acidification Study. Springer, London.
Emmett BA, Brittain S, Hughes S et al. (2004) Nitrogen additions (NaNO3 and NH4NO3) at Aber
143
forest, Wales: I. Response of throughfall and soil water chemistry. Forest Ecology and
144
Management 71:45–59.
145
Elvir J, Wiersma G, White A, and Fernandez IJ (2003) Effects of chronic ammonium sulfate
146
treatment on basal area increment in red spruce and sugar maple at the Bear Brook
147
Watershed in Maine. Canadian Journal of Forest Research 33:862–869.
148
Finzi AC (2009) Decades of atmospheric deposition have not resulted in widespread phosphorus
149
limitation or saturation of tree demand for nitrogen in southern New England.
150
Biogeochemistry 92:217–229.
151
Gaige E, Davidson EA, Hollinger DY, Fernandez IJ, Sievering H, White A, and Halteman W
152
(2007) Changes in canopy processes following whole-forest canopy nitrogen fertilization of a
153
mature spruce-hemlock forest. Ecosystems 10:1133–1147.
154
Gundersen P (1998) Effects of enhanced nitrogen deposition in a spruce forest at Klosterhede,
155
Denmark, examined by moderate NH4NO3 addition. Forest Ecology and Management
156
101:251–268.
157
Hogberg P, Fan H, Quist M, Binkley D, and Tamm C (2006) Tree growth and soil acidification
158
in response to 30 years of experimental nitrogen loading on boreal forest. Global Change
159
Biology 12:489–499.
160
Houle D and Moore J-D (2008) Soil solution, foliar concentrations and tree growth response to
161
3-year of ammonium-nitrate addition in two boreal forests of Quebec, Canada. Forest
162
Ecology and Management 255:2049–2060.
163
Hyvonen R, Persson T, Andersson, Olsson SB, Agren GI, and Linder S (2008) Impact of long-
7
164
term nitrogen addition on carbon stocks in trees and soils in northern Europe.
165
Biogeochemistry 89:121–137.
166
Kulmatiski A, Vogt KA, Vogt DJ et al. (2007) Nitrogen and calcium additions increase forest
167
growth in northeastern USA spruce-fir forests. Canadian Journal of Forest Research
168
37:1574–1585.
169
Magill AH, Aber JD, Currie W et al. (2004) Ecosystem response to 15 years of chronic nitrogen
170
additions at the Harvard Forest LTER, Massachusetts, USA. Forest Ecology and
171
Management 196:7–28.
172
Magill AH, Downs MR, Nadelhoffer KJ, Hallett RA, and Aber JD (1996) Forest ecosystem
173
response to four years of chronic nitrate and sulfate additions at Bear Brooks Watershed,
174
Maine, USA. Forest Ecology and Management 84:29–37.
175
McNulty SG, Boggs J, Aber JD, Rustad L, and Magill AH (2005) Red spruce ecosystem level
176
changes following 14 years of chronic N fertilization. Forest Ecology and Management
177
219:279–291.
178
Mirmanto E, Proctor J, Green J, Nagy L, and Suriantata (1999) Effects of nitrogen and
179
phosphorus fertilization in a lowland evergreen rainforest. Philosophical Transactions Of
180
The Royal Society Of London Series B-Biological Sciences 354:1825–1829.
181
Moldan F, Kjonaas OJ, Stuanes AO, and Wright RF (2006) Increased nitrogen in runoff and soil
182
following 13 years of experimentally increased nitrogen deposition to a coniferous-forested
183
catchment at Gardsjon, Sweden. Environmental Pollution 144:610–620.
184
185
186
Nilsen P and Abrahamsen G (2003) Scots pine and Norway spruce stands responses to annual N,
P and Mg fertilization. Forest Ecology and Management 174:221–232.
Pregitzer KS, Burton AJ, Zak DR, and Talhelm AF (2008) Simulated chronic nitrogendeposition
8
187
188
189
increases carbon storage in Northern Temperate forests. Global Change Biology 14:142–153.
Tanner E, Kapos V, and Franco W (1992) Nitrogen and phosphorus fertilization effects on
Venezuelan montane forest truck growth and litterfall. Ecology 73:78–86.
190
Schleppi P, Muller N, Feyen H, Papritz A, Bucher J, and Fliihler H (2004) Nitrogen budgets of
191
two small experimental forested catchments at Alptal, Switzerland. Forest Ecology and
192
Management 101:177–185.
193
Sikstrom U (2002) Effects of liming and fertilization (N, PK) on stem growth, crown
194
transparency, and needle element concentrations of Picea abies stands in southwestern
195
Sweden. Canadian Journal of Forest Research 32:1717–1727.
196
Wallace ZP, Lovett GM, Hart JE, and Machona B (2007) Effects of nitrogen saturation on tree
197
growth and death in a mixed-oak forest. Forest Ecology and Management 243:210–218.
9
Supplemental Information Table 1. Forest nitrogen fertilization experiments used in modeldata comparison. Simulation level refers to the fertilization inputs used in the model
simulations. ANPP, Aboveground net primary productivity
Citation
Forest Type Latitude Longitude Duration Fertilization Simulation
(years) Level
Level
(g N m-2 yr-1) (gN m-2 yr1
)
Magill et al 2004
Broadleaf
43.4 N 72.2 W 16
5
4
(Hardwood +5)
temperate
Magill et al 2004
Broadleaf
43.4 N 72.2 W 16
15
10
(Hardwood +15)
temperate
Pregitizer et al 2008 Broadleaf
46.7 N 88.9 W 10
3
4
(Site A)
temperate
Pregitizer et al 2008 Broadleaf
45.6 N 84.9 W 10
3
4
(Site B)
temperate
Pregitizer et al 2008 Broadleaf
44.4 N 85.8 W 10
3
4
(Site C)
temperate
Pregitizer et al 2008 Broadleaf
43.7 N 86.2 W 10
3
4
(Site D)
temperate
Magill et al 1996
Broadleaf
44.9 N 68.1 W 3
2.5
2
(+ 2.5)
temperate
Magill et al 1996
Broadleaf
44.9 N 68.1 W 3
5.6
4
(+5.6)
temperate
Finzi 2009
Broadleaf
42 N
73.3 W 2
15
10
(Sugar maple – ash) temperate
Finzi 2009
Broadleaf
42 N
73.3 W 2
15
10
(Oak-beech-hemlock) temperate
Wallace et al. 2008 Broadleaf
41.8 N 73.8 W 9
72
10
temperate
Kulmatiski et al 2007 Needleleaf 43.8 N 74.9 W 6
10
10
(Big Moose)
evergreen
Kulmatiski et al 2007 Needleleaf 43.92 N 71.7 W 6
10
10
(Hubbard Brook)
evergreen
Högberg et al 2006 Needleleaf 64.4 N 19.8 E
30
3.4
4
(+3.4)
evergreen
Högberg et al 2006 Needleleaf 64.4 N 19.8 E
30
6.4
4
(+6.4)
evergreen
Högberg et al 2006 Needleleaf 64.4 N 19.8 E
30
10.8
10
(+10.8)
evergreen
Hyvonen et al. 2008 Needleleaf 64.3 N 19.75 E 19
3.6
4
(Norrliden U1)
evergreen
Hyvonen et al. 2008 Needleleaf 60.92 N 16.02 E 32
3.4
4
(Stråsan)
evergreen
Hyvonen et al. 2008 Needleleaf 57.13 N 14.75 E 14
6.4
4
(Asa)
evergreen
Hyvonen et al. 2008 Needleleaf 57.10 N 15.48 E 16
4.4
4
(Åseda)
evergreen
Hyvonen et al. 2008 Needleleaf 64.1 N 19.45 E 15
8
10
(Flakaliden)
evergreen
10
Type of
measurement
ANPP
ANPP
ANPP
ANPP
ANPP
ANPP
ANPP
ANPP
ANPP
ANPP
Basal Area
Increment
ANPP
ANPP
Volume increment
Volume increment
Volume increment
Carbon increment
Carbon increment
Carbon increment
Carbon increment
Carbon increment
Hyvonen et al. 2008
(Skogaby)
Nilsen and
Abrahamsen 2003
(Pine + 3)
Nilsen and
Abrahamsen 2003
(Pine + 9)
Nilsen and
Abrahamsen 2003
(Spruce +3)
Nilsen and
Abrahamsen 2003
(Spruce +9)
Moldan et al. 2006
Needleleaf
evergreen
Needleleaf
evergreen
56.6 N
13.2 E
14
10
10
Carbon increment
58.9 N
8.57 E
4
3
4
Volume increment
Needleleaf
evergreen
58.9 N
8.57 E
4
9
10
Volume increment
Needleleaf
evergreen
61.15 N 8.60 E
9
3
4
Volume increment
Needleleaf
evergreen
61.15 N 8.60 E
9
9
10
Volume increment
59.07 N 12.05 E
13
3.5
4
Volume increment
56.8 N
12.9 E
5
2
2
Volume increment
56.8 N
12.8 E
4
2
2
Volume increment
56.3 N
13.1 E
4
2
2
Volume increment
56.2 N
13.1 E
4
2
2
Volume increment
47.3 N
71.2 W
3
1.8
2
47. 3 N
71.2 W
3
6.0
4
49.2 N
73.7 W
3
0.9
0.5
49.2 N
73.7 W
3
3
4
21.1 N
89.3 W
3
22
30
21.1 N
89.3 W
3
22
30
8.75 N
82.3 W
2
12.5
10
Basal Area
Increment
Basal Area
Increment
Basal Area
Increment
Basal Area
Increment
Basal Area
Increment
Basal Area
Increment
ANPP
3.0 S
47.5 W
2
10
10
9.1 N
79.8 W
9
12.5
10
Needleleaf
evergreen
Sikstrom 2002
Needleleaf
(Site 244)
evergreen
Sikstrom 2002
Needleleaf
(Site 245)
evergreen
Sikstrom 2002
Needleleaf
(Site 246)
evergreen
Sikstrom 2002
Needleleaf
(Site 247)
evergreen
Houle and Moore
Needleleaf
2008 (Fir + 1.8)
evergreen
Houle and Moore
Needleleaf
2008 (Fir + 6.0)
evergreen
Houle and Moore
Needleleaf
2008 (Spruce + 0.9) evergreen
Houlle and Moore
Needleleaf
2008 (Spruce + 3)
evergreen
Campo and Vazquez- Tropical
Yanes 2004 (old)
evergreen
Campo and Vazquez- Tropical
Yanes 2004 (young) evergreen
Adamek et al. 2009 Tropical
evergreen
Davidson et al. 2004 Tropical
evergreen
Wright et al. 2011
Tropical
evergreen
11
Aboveground C
increment
ANPP
Supplemental Information Figure 1: Aboveground net primary productivity response to nitrogen
fertilization grasslands (n = 35). The observations are from the N fertilization experiments and
the model response is from the grid cells that contain the same field experiments. The model
fertilization matched the duration and magnitude of N fertilization in the field experiment.
12
a
b
c
d
e
f
Supplemental Information Figure 2. 5-yr mean NPP responses in the CLM-CN (a.c,e) and O-CN
(b,d,f) for 5 gN m-2 yr-1 (a,b), 10 g N m-2 yr-1 (c,d) and 30 gN m-2 yr-1 (e,f) global nitrogen
fertilization simulations.
13