*Manuscript Click here to download Manuscript: Nitrogen Allocation Nov 28 Submission.doc

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2 3 4 5

Toward a mechanistic modeling of nitrogen limitation on vegetation dynamics

6 7

Chonggang Xu1*, Rosie Fisher2, Cathy J. Wilson1, Stan Wullschleger3, Michael Cai1, Nate G. McDowell1

8 9

1: Division of Earth and Environmental Sciences, Los Alamos National Laboratory, Los Alamos, NM; 2: National

10

Center for Atmospheric Research, Boulder, CO; 3: 1: Los Alamos National Laboratory, Los Alamos, NM; 2:

11

National Center for Atmospheric Research, Boulder, CO; 3: Oak Ridge National Laboratory, Oak Ridge, TN.

12

*Corresponding author: Chonggang Xu

13

J495, EES-14, Los Alamos National Lab

14

Los Alamos, NM 87545

15

Email: [email protected]

16

Phone: (505) 665-6136

17

Fax: (505) 665-3285

18

Key words: nitrogen allocation, Rubisco, nitrogen storage, CO2 fertilization, radiation

19

1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

Abstract: Nitrogen is an important regulator of vegetation dynamics, net primary production,

2

and terrestrial carbon cycles. Most ecosystem models utilize a prescribed relationship between

3

leaf nitrogen and photosynthetic capacity; however, this relationship varies with light,

4

temperature, CO2 and nitrogen availability, which can affect the reliability of vegetation

5

dynamics predictions under both current and future climate conditions. To account for this

6

known variability in nitrogen-photosynthesis relationships, in this study, we developed a

7

mechanistic model of nitrogen limitation on photosynthesis with nitrogen trade-offs among light

8

absorption, electron transport, carboxylation, respiration and storage. We tested the model

9

against data from experiments that manipulated CO2, temperature, radiation and nitrogen

10

availability, to determine whether it could capture the responses of leaf nitrogen and

11

photosynthetic capacity relationship under these conditions. When calibrated against

12

observations our model indicates that with nitrogen fertilization, coniferous trees allocate more

13

nitrogen for storage while deciduous trees allocate more nitrogen for growth. Elevated CO2

14

concentrations lead to lower carboxylation nitrogen allocation but higher storage nitrogen

15

allocation due to the increase in carboxylation nitrogen use efficiency, while low growing

16

temperatures cause higher carboxylation nitrogen allocation due to lower nitrogen requirements

17

for light absorption and for electron transport.

18

carboxylation under a low level of radiation results from increased nitrogen requirements for

19

light absorption and electron transport. As far as we know, the model developed in this paper is

20

the first model of complete nitrogen allocation that simultaneously considers nitrogen allocations

21

to light absorption, electron transport, carboxylation, respiration and storage, and the responses

22

of each to altered environmental conditions.

23

2

The reduction in nitrogen allocation for

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2

1. Introduction

3

Nitrogen limitation is an important regulator of vegetation growth and carbon cycles at local,

4

regional, and global scales [1,2,3,4,5]. This has been shown in temperate and tropical

5

ecosystems [6], but is especially critical in ecosystems at high latitudes [7,8]. Most ecosystem

6

models simulate the nitrogen effects on photosynthesis based on a prescribed relationship

7

between leaf nitrogen content and photosynthetic capacity (Vcmax) [9,10]; however, this

8

relationship may vary with different light, temperature, nitrogen availability and CO 2 conditions

9

[11,12,13]. Photosynthetic capacity is one of the most important parameters affecting simulated

10

carbon fluxes [14,15]. Using a constant relationship between leaf nitrogen content and

11

photosynthetic capacity can thus reduce the reliability of carbon balance prediction under current

12

and future climates. In order to improve the prediction accuracy of nitrogen limitation on

13

photosynthesis, it is important that we build models that account for key factors contributing to

14

this variability.

15

Nitrogen is a major constituent of proteins for biological processes (e.g. photosynthesis and

16

respiration) [16] and plants need to balance nitrogen investment in proteins for different

17

biological processes to optimize growth and/or survive under specific environmental conditions

18

[12,17,18,19,20]. Previous studies have pointed out that the altered nitrogen investment in

19

carboxylation enzymes (mainly ribulose-1,5-bisphosphate carboxylase oxygenase, Rubisco) and

20

in light-capturing proteins of thylakoid (responsible for light absorption and electron transport)

21

under different light conditions is one of the key factors contributing to the variability in the

22

relationship between leaf nitrogen and photosynthetic capacity [21]. In this paper, we propose

23

two additional types of nitrogen investment that can impact photosynthesis: storage nitrogen and 3

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

respiratory nitrogen. The storage nitrogen can persist in the form of inorganic nitrogen, amino

2

acid and proteins [19,22,23], which can be used to grow new plant organs and produce metabolic

3

enzymes. Along with stored carbohydrates, storage nitrogen can thus sustain plant growth and

4

survival under situations of plant tissue losses due to unpredicted disturbances (e.g., herbivory

5

attack and browsing) or reduced soil nitrogen availability due to competition [19,22,23,24].

6

Respiratory nitrogen is invested in enzymes of mitochondria to generate energies to support

7

growth and tissue maintenance. Importantly to this study, nitrogen allocated for storage and

8

respiration can impact photosynthesis rates because it reduces the total amount of nitrogen

9

available to existing photosynthetic tissues, while simultaneously enabling growth and survival.

10

Previous modeling studies to estimate nitrogen allocations for key photosynthetic enzymes

11

are encouraging [12,18,20,25,26]; however, few models have simultaneously considered

12

nitrogen allocations to storage, carboxylation, respiration and light harvesting. Furthermore,

13

previous models have mainly focused on the effects of light conditions on nitrogen allocations,

14

with few of them incorporate other important environmental factors such as temperature, CO2

15

and nitrogen fertilizations. In this study, we developed a complete nitrogen allocation model that

16

incorporates nitrogen trade-offs among major biological processes including light absorption,

17

electron transport, carboxylation, respiration and growth. The model should help us better

18

understand photosynthetic acclimation under future climate, and also provide a more

19

mechanistic prediction of nitrogen limitation upon photosynthesis.

20

2.

Methods

21

The developed model is based on three key assumptions. First, plants will balance nitrogen

22

allocation so that the photosynthesis is co-limited by light capture (including light absorption

23

and electron transport), carboxylation and respiration. Second, plants have different strategies

4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

related to the trade-off between plant persistence and growth, which will cause differences in the

2

amount of nitrogen allocated to storage and thus variability of the relationship between leaf

3

nitrogen and photosynthesis. Third, plants are able to dynamically adjust their nitrogen

4

allocations to different environmental conditions.

5

In our model, plant nitrogen is divided into four groups: structural nitrogen, photosynthetic

6

nitrogen, storage nitrogen and respiratory nitrogen (Figure 1). The structural nitrogen is mainly

7

used to build cell walls and DNA. The photosynthetic nitrogen is used to build three major

8

groups of enzymes/proteins: proteins for light absorption (mainly chlorophyll), proteins for

9

electron transport (mainly proteins in bioenergetics and proteins in phosystem I, II and light

10

harvesting complexes except for chlorophyll) and proteins for carboxylation (mainly Rubisco)

11

(Figure 1). A key assumption of our model is that plants will balance the nitrogen allocation for

12

these three groups of proteins to maximize the photosynthesis rate, based on the concept that

13

plants should seek to maximize photosynthetic carbon uptake for a given unit investment of

14

nitrogen. The respiratory nitrogen is invested in mitochondria to generate energy required for

15

growth and maintenance [27], which is dependent on the photosynthesis rate and the amount of

16

proteins to maintain. Under conditions of increased nitrogen demand for new tissue production

17

due to old tissue loss resulting from herbivory attack and browsing, or reduced nitrogen uptake

18

due to changes in environmental conditions such as reduction in soil nitrogen availability, it

19

might be possible that there is not enough nitrogen for new plant tissue production. To avoid this

20

situation in the model, we assume that the plant will store a certain amount of nitrogen so that

21

nitrogen is available for the production of additional photosynthetic, respiratory and defense

22

enzymes in old and new plant tissues [24], a process which can be important for plant growth

23

and survival [17,19,28]. The amount of storage nitrogen is defined in the model by the duration

5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

of time that the nitrogen storage can support the current rate of carbon assimilation if nitrogen

2

uptake were to cease altogether. We denote this period of time as Dns (days) and the size of the

3

nitrogen store is affected both by the rate of use, and by life history strategy as expressed

4

through Dns . We do not explicitly estimate nitrogen investment for defense due to the lack of

5

data; however, the exclusion of defense enzymes will not affect the validity of the model

6

because the model will be fitted to the observed Vcmax dataset to estimate the nitrogen storage

7

duration, which will implicitly incorporate nitrogen allocated for defense enzymes in the storage

8

nitrogen allocations.

9

In the model, we define the total nitrogen for photosynthesis, storage and respiration as

10

functional nitrogen, which is the whole plant nitrogen pool minus the amount of nitrogen used

11

for structural purposes. See Figure 2 for details of nitrogen classifications in this paper and Table

12

1 for lists of definitions for main model parameters. The functional nitrogen content can be

13

expressed as leaf-area based ( FNCa , g N/m2 leaf) or leaf-mass based ( FNCm , g N/g leaf) and is

14

used as an indicator of plant nitrogen availability. The functional nitrogen is hierarchically

15

allocated for five major processes (Figure 2). First, functional nitrogen is allocated between

16

growth and storage based on a plant’s strategies of growth and persistence. Second, the growth

17

nitrogen is partitioned into photosynthetic and respiratory nitrogen. Third, the photosynthetic

18

nitrogen is allocated between light-harvesting and carboxylation. Finally, light-harvesting

19

nitrogen is allocated between light absorption and electron transport.

20

In summary, we impose a series of assumptions on the model to generate the ideal nitrogen

21

distributions. These are i) storage is allocated to meet requirements based on the nitrogen storage

22

duration; ii) respiratory nitrogen is equal to the demand implied by growth; iii) light absorption,

23

electron transport and carboxylation are co-limiting to maximize photosynthesis. These 6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

assumptions are mostly based on the presumption of optimality [29] and together form a testable

2

hypothesis concerning the function of plan nitrogen allocation under varying environmental

3

conditions.

4

2.1. Growth and storage nitrogen allocation

5

The proportion of storage nitrogen (1- PN g ) in the functional nitrogen pool is determined by

6

a nitrogen storage duration parameter Dns , the net carbon assimilation rate ( An ) and the target

7

sink tissue nitrogen content ( NC sink ) as follows,

8

Dns An NC

sink

 (1  PN g ) FNCa .

(1)

9

The left side of the equations is the demand of storage nitrogen, while the right side of the

10

equation is the supply of storage nitrogen. See Table 1 for definitions of main model parameters.

11

In this equation, the net carbon assimilation rate ( An ) is calculated by subtracting respiration

12

from gross carbon assimilate rate,

13

An  Cv [(1  C gr ) NUE p PN p PN g FNCa  Rm )]

(2)

14

where the gross carbon assimilate rate (i.e., NUE p PN p PN g FNCa ) equals the size of the

15

photosynthetic nitrogen pool ( PN p PN g FNCa , see figure 2) multiplied by the photosynthetic

16

nitrogen use efficiency ( NUE p , umol CO2/g photosynthetic N/day, see Text S1 for details). Net

17

carbon assimilation is the gross carbon assimilate rate minus the growth respiration fraction

18

( Cgr ) and the maintenance respiration (Rm, umol CO2/m2/day). We assume that 25% of gross

19

carbon assimilation is used for growth respiration [30,31] and the maintenance respiration is

20

dependent on functional nitrogen content as follows [32],

21

Rm  MRb FNCa .

(3)

7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

where MRd is the maintenance respiration demand per gram of nitrogen (umol CO2/m2/day).

2

See Text S1 for details of estimation of MRd . The target new tissue concentration ( NC sink ) is

3

calculated based on the sum of functional and structure nitrogen content,

4

NC

sink

 (l  FNCmtar  SNCm )

(4)

5

where l is the proportion of biomass is allocated for leaf; FNCmtar is the target functional

6

nitrogen content (g N/g leaf biomass); and SNCm (=0.001) is the structural nitrogen content (g

7

N/g leaf biomass) [33]. In the field, it will be difficult to directly measure functional nitrogen

8

content. Thus, FNCmtar is estimated from the measured mean leaf nitrogen content (g N/g leaf

9

biomass, MLNCm ) as follows

10

FNCmtar  k MLNCm ,

(5)

11

where the coefficient k is the ratio of total plant functional nitrogen to the amount of nitrogen

12

allocated to leaf. We assume that the majority of functional nitrogen is allocated for leaf and

13

empirically set k to be 1.1 in this paper. Replacing eqs. (2) and (4) into eq. (1), we have

14

Dns Cv [(1  C gr ) NUE p PN p PN g FNCa  Rm )](l  FNCmtar  SNCm )  (1  PN g ) FNCa . (6)

15

We solve eq. (6) to derive PN g given values of PN p and FNCa .

16

2.2. Photosynthetic and respiratory nitrogen allocation

17

Second, the growth nitrogen is partitioned into photosynthetic and respiratory organelles.

18

We assume that 25% of photosynthesis production is used for growth respiration [30] and

19

maintenance respiration is dependent on functional nitrogen content (see eq. (3)). To optimize 8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

nitrogen use efficiency, we equalize the nitrogen allocated to respiratory organelles to the rate of

2

respiration implied by the growth and maintenance terms,

CvCgr NUE p PN p PN g FNCa  Rm

3

NUEr

 (1  PN p ) PN g FNCa

(7)

4

where the numerator on the leaf side of equation specifies the carbon used in growth and

5

maintenance respiration. This is divided by the nitrogen use efficiency of respiratory enzymes

6

( NUEr umol CO2/m2/day, see text S1 for details) to give the N requirements for respiration. The

7

right hand side represents the allocation of growth N to respiratory organelles (1- PN p ) as

8

opposed to photosynthesis PN p .

9

2.3. Light absorption and electron transport nitrogen allocation

10

The electron transportation rate is dependent on light absorption efficiency, photosynthetic

11

active radiation and maximum electron transportation rate [26,34]. It is estimated using the

12

Smith's equation as follows [26,34]

J  J max

13

14

15

0.292 Q J max 2  (0.292 Q)2

(8)

with



[Chl ] 0.076  [Chl ]

(9)

16

where J max is the maximum electron transportation rate and

17

radiation (umol photon/m2/s).  is the light absorption efficiency and is dependent on

9

Q is the photosynthetic active

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

chlorophyll content ( [Chl ], mmol Chl/m2). J max and [Chl ]depends on the amount of nitrogen

2

allocated for electron transport ( 1  PNchl ) and light absorption( PN chl ), respectively. Specifically,

J max  NUEJ (1  PNchl ) PNl PNCa

3

(10)

4

The term PNl PNCa specifies the light-harvesting nitrogen content (g N/m2). NUEJ is the nitrogen

5

use efficiency for maximum electron transport (umol electron/g N). See text S1 for details of

6

NUEJ estimation. The chlorophyll content can be determined by the proportion of nitrogen

7

allocated for light absorption as follows,

[Chl ]  17.8PNchl PNl PNCa

8

(11)

9

where PN chl is the proportion of nitrogen allocated for light absorption within the light-

10

harvesting nitrogen pool (See Figure 2 for details). The coefficient 17.8 is the nitrogen binding

11

coefficient for chlorophyll (mmol Chl/g N), given that one mole chlorophyll contains four mole

12

nitrogen [21]. Replacing eqs. (9), (10), and (11) into eq.(8), we have

13

J

5.2 NUEJ (1  PNchl ) PNchl PNl PNCaQ [ NUEJ (1  PNchl )(0.076  17.8PNchl PNl PNCa )]2  (5.2PNchl Q)2

(12)

14

The nitrogen allocation for chlorophyll PN chl is thus estimated by maximizing the electron

15

transport rate in the above equation given values of PNl and PNCa .

16

2.4. Light harvesting and carboxylation nitrogen allocation

17

Finally, based on the electron transport rates calculated above, photosynthetic nitrogen is

18

allocated between light-harvesting and carboxylation by equalizing the Rubisco-limited 10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

carboxylation rate (Wc) and electron-limited carboxylation rate (Wj). Following the Farquhar

2

model [35], Rubisco-limited carboxylation rate, Wc , is estimated as follows,

Wc  RcVcmax

3

(13)

4

where Vcmax is the maximum rate of carboxylation (umol CO2/m2/s) and Rc is the CO2

5

concentration adjustment factor. See Text S2 for details of Rc calculation. The value of Vcmax is

6

determined by the nitrogen allocated for carboxylation as follows,

Vcmax  NUEc (1  PNl ) PNCa ,

7

(14)

8

where NUEc is the nitrogen use efficiency for maximum rate of carboxylation (umol CO2/m2/s/g

9

N). See Text S1 for details of NUEc calculation. Meanwhile, the electron-transport-limited

10

carboxylation rate can be estimated based on the potential electron transport rate [36], Wj  Rj J ,

11

(15)

12

where R j is the CO2 concentration adjustment factor. See Text S2 for details of R j calculation.

13

To optimize the nitrogen use efficiency, we equalize Wc and W j . This leads to the following

14

equation,

15

5.2 NUEJ (1  PNchl ) PNchl PNl QR j [ NUEJ (1  PNchl )(0.076  17.8PNchl PNl PNCa )]2  (5.2 PNchl Q) 2

 Rc NUEc (1  PNl ) . (16)

16

The nitrogen allocation for light harvesting is thus estimated by solving for PNl in the above

17

equation given values of PN chl and PNCa .

11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

2.5. Model fitting

2

A key unknown parameter in the model is the nitrogen storage duration ( Dns in eq. (6)),

3

which determines the storage nitrogen allocation. This nitrogen storage duration parameter

4

determines the strategies for nitrogen trade-off between growth and persistence. Long storage

5

duration indicates that plants allocate more nitrogen toward persistence, while short storage

6

duration indicates that plants allocate more nitrogen toward growth. It is possible that the

7

nitrogen storage duration may change with nitrogen availability and to account for this

8

possibility, the nitrogen storage duration is simulated to be linearly dependent on functional

9

nitrogen content as follows,

10

Dns  a  b( FNCm  FNCmtar ) ,

(17)

11

where the parameter a is the reference number of day for nitrogen storage duration, b is the

12

change rate of nitrogen storage duration (days) with increase of nitrogen availability, and

13

FNCm is the current functional nitrogen content. If the slope b is zero, then nitrogen storage

14

duration does not change with functional nitrogen content. If the slope b is positive, then the

15

nitrogen storage duration increases with increased functional nitrogen content. Otherwise, the

16

nitrogen storage duration decreases with increased functional nitrogen content.

17

For specific nitrogen storage duration, we can iteratively solve eqs. (6), (7), (12) and (16) to

18

estimate the hierarchical nitrogen allocations for growth ( PN g ), photosynthesis ( PN p ), light

19

absorption ( PN chl ) and light harvesting ( PNl ). Using the hierarchical nitrogen allocation

20

coefficients, we are able to estimate the proportions of storage nitrogen ( PNFs ), respiratory

21

nitrogen ( PNFr ), carboxylation nitrogen ( PNFc ), light absorption nitrogen ( PNFl ) and electron 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

transport nitrogen ( PNFe ) within the functional nitrogen pool as follows (see Figure 2 for a

2

better understanding),

PNFs = 1  PN g PNFr = (1  PN p ) PN g 3

PNFc = (1  PN l )(1  PN p ) PN g

(18)

PNFl = PN chl (1  PN l )(1  PN p ) PN g PNFe = (1  PN chl )(1  PN l )(1  PN p ) PN g

.

4

To fit our model to the observed values of Vcmax at different levels of leaf nitrogen content,

5

we need to first estimate the leaf nitrogen content based on the functional nitrogen content. In

6

this paper, we empirically allocate 80% of storage nitrogen and 50% of respiration nitrogen to be

7

in leaf (see eq. (S2.1)). With estimated leaf nitrogen content, we can also estimate the

8

proportions of structural nitrogen, storage nitrogen, respiratory nitrogen, carboxylation nitrogen,

9

light absorption nitrogen and electron transport nitrogen within the leaf nitrogen pool. See eq.

10

(S2.2) for details. With estimated carboxylation nitrogen, we are also able to estimate Vcmax with

11

eq.(14) and compare that to observed Vcmax. We tune parameters a and b in eq. (17) so that

12

modeled Vcmax is in a good agreement with observation. See Text S2 for details of model fitting.

13

Photosynthetic constants of J max and Vc max are temperature dependant. There are different

14

forms of temperature dependence function reported in the literature. This paper uses the

15

temperature dependant-functions from CLM4.0 [37]. See Text S3 for details of the specifications

16

for temperature dependence functions.

17

2.6. Data and experimental designs

13

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

To obtain a better understanding of the nitrogen allocation strategies under different

2

environmental conditions, we applied our model to four different experimental treatments: 1)

3

nitrogen fertilization for a coniferous tree species (Douglas fir, Pseudotsuga menziesii) and for a

4

deciduous tree species (Populus euroamericana) [13]; 2) CO2 fertilization for loblolly pine

5

(Pinus taeda)

6

asiatica) [39]; and 4) reduced radiation for the Japanese plantain [39]. These studies provide a

7

wide range of environmental conditions to allow testing of the impacts of resource changes on

8

nitrogen allocation, and they each provide the critical data for model fitting purposes, which are

9

i) Vcmax at different levels of leaf nitrogen content, ii) leaf mass per area (LMA, directly or

10

indirectly through other studies), iii) photosynthetic active radiation (PAR) and iv) growing

11

temperature. See Table 2 for the main model inputs. We assume that the plant allocation

12

strategy will be weighted to optimize assimilation during the times of the day when potential

13

assimilation is highest. Thus, we use the mean radiation during 8:00AM-4:00PM for test cases 1

14

and 2 and the four hour high radiation for test cases 3 and 4 (see Figure S2 and see Q in eqs. (12)

15

and (16)) to calculate the nitrogen allocations among light absorption, electron transport and

16

carboxylation. In this paper, we do not utilize meta-analysis studies [40,41] that reported pooled

17

relationship between leaf nitrogen content and Vcmax , because the pooling may obscure the

18

details of individual species’ nitrogen allocation strategies.

[38]; 3) reduced growing temperature for the Japanese plantain (Plantago

19

We tune the nitrogen storage duration by varying parameters a and b in eq. (17) so that the

20

Vcmax determined by carboxylation nitrogen allocation with eq. (14) in our model fits the

21

observed Vcmax at different leaf nitrogen concentrations. With the tuned nitrogen storage

22

duration, we derive a complete inventory of the nitrogen investments for light absorption,

23

electron transport, carboxylation, storage and respiration. Estimated nitrogen allocation 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

coefficients with the tuned model can help us gain insights into the effects of nitrogen on

2

photosynthesis with different environmental conditions. Furthermore, in order to test if the

3

model is able to predict the Vcmax at different environmental conditions, for test cases 2-4, we

4

tune the model for the control condition and use the tuned model to predict Vcmax for the

5

treatment condition. See Table 2 for control and treatment conditions for each test case.

6

Since the reported values of Vcmax in different studies can be estimated based on different

7

values of Michealis constants for CO2 and O2 (i.e., K c and K o in eq. (S3.2) in text S3) and

8

different temperature dependence functions, we specifically standardize the Vcmax using the

9

values of K c and K o and temperature dependence functions reported by Collatz et al [42] (see

10

Text S5 for details). For test cases 3-4, Hikosaka [39] measured the Vcmax at both 15 oC and 30

11

o

12

estimate the Vcmax at 25oC by scaling from that measured at the plant’s growing temperature (15

13

or 30 oC).

14

3.

C. To reduce the effect of measurement temperature on active status of Rubisco [43], we

Results

15

3.1. Nitrogen fertilization effects on leaf nitrogen allocation

16

Nitrogen fertilization increased leaf nitrogen concentrations in both species, but Vcmax only

17

increased substantially in the poplar leaves, while in the douglas fir leaves, it remained almost

18

constant as leaf nitrogen increased (Figure 3 a). This implies that the extra nitrogen was

19

allocated to some pools other than carboxylation in the douglas fir leaves. When the nitrogen

20

allocation model is fitted to this dataset, the results suggest that, compared to poplar tree, the

21

douglas fir has longer nitrogen storage duration (Figure 3 b) and higher storage nitrogen

22

allocation (Figure 3 c). It appears that douglas fir keeps the increased functional nitrogen in 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

storage, leading to longer nitrogen storage duration. Meanwhile, the poplar tree adjusts its

2

nitrogen allocation with increased nitrogen availability such that it keeps the proportion of

3

storage nitrogen to functional nitrogen relatively stable (Figure 3 c). Because we assume that

4

major photosynthetic and respiratory processes are co-limiting, a higher carboxylation rate for

5

the poplar trees will also need a higher rate of light absorption, electron transport and respiration

6

for maintenance and production. Thus, compared to douglas fir, the poplar trees also allocate

7

more nitrogen for light absorption (Figure 3 e), electron transport (Figure 3 f) and respiration

8

(Figure 3 g).

9

3.2. CO2 fertilization effects on leaf nitrogen allocations

10

The Vcmax data for loblolly pine with 8-9 years of CO2 fertilization shows that elevated CO2

11

down-regulates the Vcmax with the same level of leaf nitrogen content compared to that under

12

ambient CO2 concentrations (Figure 4 a). Our nitrogen allocation model is first tuned to Vcmax

13

data under ambient CO2 concentrations and is then used to predict Vcmax under the elevated CO2

14

concentrations. The predicted values of Vcmax under elevated CO2 is in a reasonable agreement

15

with the observed values of Vcmax (Figure 4 a). In the model, elevated CO2 induces a higher

16

photosynthesis rate for the same amount of Rubisco, by inhibiting photorespiration and

17

increasing carboxylation rate [35]. This increases the demand for nitrogen storage (eq. (1)),

18

which leads to higher storage nitrogen allocation (Figure 4 c and lower carboxylation nitrogen

19

allocation (Figure 4 d) and lower Vcmax (Figure 4 a) under elevated CO2 concentrations.

20

Meanwhile, because electron transport and carboxylation are co-limiting in our model, with

21

increased enzyme activities under elevated CO 2 concentrations, the nitrogen allocated for

22

electron transport and light absorption should increase relative to nitrogen allocated for

23

carboxylation under elevated CO2 concentrations (Figure S4 a, b); however their allocations 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

decreases at the whole-leaf level (Figure 4 e, f). This is because the increase in storage nitrogen

2

allocation can result in less nitrogen available for light absorption and electron transport.

3

3.3. Temperature effects on leaf nitrogen allocations

4

The Vcmax data for Japanese plantain with 1-2 months of reduced growing temperature show

5

that decreased growing temperature up-regulates the Vcmax for the same leaf nitrogen content

6

(Figure 5 a). Our nitrogen allocation model is first fitted to Vcmax dataset at the high growing

7

temperature (30oC) and is then used to predict nitrogen allocations for the low growing

8

temperature (15oC). Vcmax under reduced temperature predicted by our model is in a good

9

agreement with the observed values of Vcmax (Figure 5 a, the dashed line and open circles). Our

10

model predicts that lower growing temperature increases nitrogen allocation for carboxylation

11

(Figure 5 d) due to the reduction in nitrogen allocation for light absorption (Figure 5 e) and

12

electron transport (Figure 5 f). The reduction in nitrogen allocation for light absorption is

13

because enzyme-dependant processes including electron transport, carboxylation and respiration

14

will generally decrease with reduced growing temperature, while the leaf light absorption by

15

chlorophyll is barely affected by temperature [44]. Therefore, at a low growing temperature, less

16

nitrogen is needed to be allocated to chlorophyll to support the lower levels of electron transport

17

and carboxylation. Meanwhile, the reduction in electron transport nitrogen allocation with the

18

low growing temperature is because J max is assumed in the model to be less sensitive to

19

temperature compared to with Vcmax [45,46] (see Text S4 for details). The lower growing

20

temperature also reduces the nitrogen demand for electron transport with the same level of

21

nitrogen allocated for carboxylation. The model implies that the storage nitrogen slightly

22

increases (Figure 5 b), mainly as a result of increased photosynthetic nitrogen use efficiency

17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

(NUEp) with higher nitrogen allocation for Rubisco and a lower rate of photorespiration [36].

2

The observed increase of LMA under the low growing temperature leads to slightly higher

3

structural nitrogen allocation (Figure 5 h). This is because we set a constant level of structural

4

nitrogen in the model (0.001 g N/g biomass; see Table 2) and a higher LMA can result in a lower

5

level of mass-based leaf nitrogen content ( LNCm and LNCm  LNCa / LMA ) with the same area-

6

based leaf nitrogen content ( LNCa ).

7

Notice that, due to higher nitrogen allocation for carboxylation (Figure 5 c) and the lower

8

photorespiration rate resulting from higher specificity of Rubisco for CO2 relative to that for O2

9

at low growing temperatures [36], the predicted ratio of photosynthetic nitrogen use efficiency

10

(i.e., NUE p , umol CO2/g photosynthetic N/day) to the respiratory nitrogen use efficiency (i.e.,

11

NUEr , umol CO2/g N for respiration/day) will increase with reduced growing temperature. In

12

order to compensate for the increase in photosynthetic nitrogen use efficiency relative to

13

respiratory nitrogen use efficiency, our model predicts higher respiratory nitrogen allocation (see

14

eq. (7) and Figure 5 g).

15

3.4. Radiation effects on leaf nitrogen allocations

16

The Vcmax data for Japanese plantain treated with 1-2 months of reduced radiation shows that

17

lower radiation down-regulates the Vcmax for the same level of leaf nitrogen content (Figure 6 a).

18

We use the nitrogen allocation model tuned for the Vcmax data at a high growing temperature

19

(30oC) and high radiation (control case for test case 3, see Table 2 for details) to predict nitrogen

20

allocation coefficients at both low and high radiation with a low growing temperature (15 oC). In

21

view of the short duration of the experiment (1-2 months), we assume that plants were not able

22

to fully optimize the nitrogen allocation for the low radiation environment. To illustrate the 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

likely direction of acclimation, we therefore predict the Vcmax at the low radiation environment

2

with the fitted model using a radiation level of 300 umol photon /m2/s (Figure 6 a dashed lines),

3

which is higher than the actual radiation of 50 umol photon /m2/s. See dotted lines in Figure 6 a

4

for model results with a full acclimation to the low radiation.

5

In the model, there are four important factors that can contribute to the response of nitrogen

6

allocation to a lower level of radiation. First, if lower radiation leads to a lower photosynthesis

7

rate, it can result in a lower nitrogen demand for storage and thus lower storage nitrogen

8

allocation with the same level of leaf nitrogen content (Figure 6 c). Second, low radiation will

9

reduce the potential electron transport rate given the same amount of nitrogen invested in light

10

capture. To balance electron transport and carboxylation, plants will need to increase the amount

11

of nitrogen allocated for light absorption (Figure 6e) and electron transport (Figure 6f). Third, a

12

lower rate of photosynthesis resulting from low radiation can also reduce the nitrogen allocated

13

for respiration (Figure 6 g). Finally, the reduction of LMA under low radiation environment can

14

also lead to a lower proportion of nitrogen allocated for cell structures (Figure 6 h).

15

4.

16

Discussion 4.1. Method implications

17

Given the observed Vcmax and leaf nitrogen content, our model can generate a complete

18

nitrogen investment inventory for major metabolic components. Compared with previous

19

studies, our model simultaneously considers nitrogen allocations to storage, carboxylation,

20

respiration, light harvesting and structure. Thus, it can provide a complete view of plant nitrogen

21

allocations to major growth components. In addition, we also considered the main environmental

22

factors including temperature, CO2 and radiation. The model accurately predicted the

19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

relationship of Vcmax to leaf nitrogen for each of the manipulative studies, providing confidence

2

that it can be used as a comprehensive model for a better understanding of nitrogen allocation

3

under future climate conditions.

4

Tuning nitrogen storage duration with comprehensive datasets of Vcmax and leaf

5

characteristics can allow the model to be used as a tool to predict photosynthesis rates for

6

different levels of leaf nitrogen content under different growing environments. Specifically,

7

given the functional nitrogen content and the environmental conditions, we are able to predict

8

the amount of nitrogen allocated for carboxylation and thus the relationship between Vcmax and

9

leaf nitrogen content. When the nitrogen allocation model is coupled with the Farquhar model

10

[35], we are able to predict how photosynthesis rates acclimated to different environmental

11

conditions.

12

With this nitrogen allocation model, we do not need to simulate the specific nitrogen

13

assignment for different plant tissues (with the exception of structural nitrogen requirement).

14

Instead, we can track the functional nitrogen content as determined by plant growth, nitrogen

15

uptake and plant tissue turn-over. The nitrogen allocation model determines the amount of

16

nitrogen allocated for photosynthetic enzymes based on the current functional nitrogen content

17

and guarantees that there is always enough storage nitrogen available for carbon sink. This

18

avoids the unrealistic situation in many ecosystem process models where nitrogen supply is not

19

sufficient to support rates of carbon assimilation and/or minimum stoichiometric ratios and an

20

empirical adjusting factor is introduced to down-regulate the photosynthesis rate accordingly.

21

Although we attempted to obtain appropriate datasets for nitrogen allocation, there are still

22

three important limitations in the model. First, the nitrogen in storage is difficult to measure

20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

since it is present as different forms of nitrogen [23,47]. Furthermore, Rubisco can function as

2

both carboxylation enzyme or storage nitrogen. Therefore, it is difficult to compare simulation

3

results with observations; however, this should not affect the predictability of the model since

4

the storage duration is tuned against the Vcmax dataset. Second, in this paper, we assume that the

5

storage nitrogen is mainly used to produce new plant tissues including both structural and

6

photosynthetic components; however, it may also be used to build defense enzymes, which can

7

be important for plant survival [17,19]. Our model may have implicitly incorporated this type of

8

investment in defense by overestimating the nitrogen storage duration; however, for a better

9

understanding of the tradeoffs between plant growth and persistence, it is important that future

10

field and modeling study quantify this type of nitrogen investment. Third, in the current version

11

of the model, we did not resolve the vertical distribution of light attenuation through the canopy

12

structure because most observations of Vcmax are from sunlit leaves near the canopy top.

13

Resolving the vertical structure of light attenuation would be required to implement a similar

14

scheme in a multi-layered vegetation model, and might also provide insight into vertical patterns

15

of leaf physiologica traits (e.g. Lloyd et al.[48]).

16

4.2. Result implications

17

One key parameter in the model is the nitrogen storage duration, which determines

18

plants’ nitrogen allocation strategy and builds its foundation on the plant strategy trade-off

19

between growth and persistence [17,49]. Our first test case shows that the deciduous Populus

20

euroamericana allocates more nitrogen for photosynthesis, while the coniferous Pseudotsuga

21

menziesii allocates more nitrogen for persistence or storage under a high level of nitrogen

22

availability (Figure 3 c, d). The difference in nitrogen storage allocation between the coniferous

23

and deciduous tree is likely to be general given the observation from large datasets that 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

deciduous trees have a much stronger relationship between leaf nitrogen and maximum rate of

2

photosynthesis (Vcmax) than coniferous trees [41], which indicates a higher proportion of nitrogen

3

allocated for carboxylation. In addition to plants’ nitrogen allocation strategy, an alternative

4

hypothesis to explain the high level of storage nitrogen allocation for the coniferous species

5

could be the limitation of growth by other resources, e.g. by phosphorous deficiencies [50,51],

6

which can result in low nitrogen allocation for carboxylation and high amount of nitrogen in

7

storage. Since the deciduous trees growing with similar soil conditions did not show an increase

8

in nitrogen storage duration, the limitation by other resources appears unlikely in our test case.

9

Because the nitrogen allocation strategy affects carboxylation nitrogen allocation and

10

subsequently the photosynthetic capacity, it can be an important factor determining species’

11

responses with nitrogen availability change. The expected nitrogen availability increase in the

12

arctic under global warming [52] might therefore benefit species that have nitrogen allocation

13

strategies favoring growth over persistence. This may be an important reason why deciduous

14

shrubs become the dominant species under fertilization experiments in the arctic [8,53], which

15

can have a stronger feedback to future climate [54].

16

Under CO2 enrichment, our model implies decreased carboxylation nitrogen investment but

17

increased storage nitrogen investment. This decreased carboxylation nitrogen may result from

18

the excess carbon supply to carbon sink capacity due to CO2 enrichment, which can be sensed in

19

mesophyll cells by a mechanism that possibly involves hexokinase acting as a flux sensor and

20

operates to reduce Rubisco content [5]. Meanwhile, plants may also increase the storage nitrogen

21

investment to up-regulate carbon sink for the balance of higher carbon supply (see eq (1)). The

22

nitrogen saved for decreased carboxylation nitrogen investment can be distributed for storage

23

nitrogen, electron transport and light absorption for best nitrogen use efficiency. Combined with 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

the leaf nitrogen reduction by carbon dilution with increased photosynthesis rate and reduced

2

passive nitrogen uptake with reduced stomata conductance [55], down-regulation of the

3

photosynthetic capacity can be substantial under elevated CO 2 concentrations [56].

4

Our model infers that lower growing temperature causes investment of nitrogen in

5

carboxylation to increase and investment in light absorption and electron transport to decrease.

6

This is in agreement with field and lab experiment data showing that, when a plant is

7

transplanted to a lower growing temperature, the investment of nitrogen in active Rubisco

8

increases but the investment in chlorophyll decreases for most cold tolerant species

9

[39,43,57,58,59]. Reich et al. [51] observed that the relationship between photosynthesis and leaf

10

nitrogen is stronger in the arctic than in the tropics. They attributed this in their paper to the

11

higher ratio of phosphorus to nitrogen in leaves of arctic plants. Based on our model, an

12

alternative hypothesis is that lower growing temperatures in the arctic might result in lower

13

chlorophyll requirements, which may lead to higher carboxylation nitrogen allocation. The

14

increased nitrogen allocation for carboxylation can ultimately lead to a stronger relationship

15

between leaf nitrogen and photosynthetic capacity.

16

5.

Conclusions

17

A complete nitrogen allocation model based on the balance of light absorption, electron

18

transport, carboxylation, respiration and storage is developed to better understand nitrogen

19

effects on photosynthesis. The nitrogen allocation model is based on a key parameter (nitrogen

20

storage duration) that determines the trade-off between growth and persistence. Our four test

21

cases with changes in nitrogen availability, CO2 concentration, growing temperature and

22

radiation demonstrate the model’s capability to investigate nitrogen allocation patterns and to

23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

predict the impact of altered environmental conditions upon nitrogen allocations. By predicting

2

nitrogen allocation coefficients under different CO2, temperature and radiation conditions for

3

different life history strategies, this model provides a useful tool toward a mechanistic projection

4

of photosynthetic acclimation under altered environments. We expect that our developed model

5

can potentially improve our confidence in simulations of carbon-nitrogen interactions [4] and the

6

vegetation feedbacks to climates in Earth system models.

7

6.

Acknowledgements

8

This work is funded by DOE Office of Science, Office of Biological and Environmental

9

Research (BER) Program and Los Alamos National Lab (LANL) Laboratory Directed Research

10

and Development (LDRD) Program. This submission is under public release of LA-UR-11-

11

12108.

12

24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2

Tables

3

Table 1 Main model parameters Parameters

Cv Cgr Dns

FNCa FNCm FNCmtar J max MRd NUE p

Descriptions

Conversion factor from CO2 (umol) to biomass (g); Cv =2.4  10-5 Proportion of photosynthetic carbon for respiration; Cgr=0.25 [31]. Duration of time (days) that the nitrogen storage can support the current rate of carbon assimilation if nitrogen uptake were to cease altogether Functional nitrogen content (g N/m2 leaf) Functional nitrogen content (g N/g leaf) Target functional nitrogen content (g N/g leaf) Maximum electron transportation rate (umol electron/m2/s) Maintenance respiration demand per gram of nitrogen (umol CO2/m2/day) Photosynthetic nitrogen use efficiency (umol CO2/g photosynthetic N/day)

NUEr NUEJ PN g

Respiratory nitrogen use efficiency (umol CO2/g respiratory N/day)

PN p

Proportion of nitrogen allocated for photosynthesis in growth nitrogen pool

PN l PN chl PNCa PNFs PNFr PNFc PNFl PNFe Rc RJ Rm SNCm Vc max Wc Wj

Proportion of nitrogen allocated for light harvesting in photosynthetic nitrogen pool

Nitrogen use efficiency for maximum electron transport (umol electron/g N/s) Proportion of nitrogen allocated for growth in functional nitrogen pool

Proportion of nitrogen allocated for chlorophyll in light harvesting nitrogen pool Photosynthetic nitrogen content (g N/m2); PNC  PN PN FNC a p g a proportions of storage nitrogen within the functional nitrogen pool proportions of respiratory nitrogenwithin the functional nitrogen pool proportions of carboxylation nitrogen within the functional nitrogen pool proportions of light absorption nitrogen within the functional nitrogen pool proportions electron transport nitrogen within the functional nitrogen pool CO2 adjusting factor for Rubisco-limited carboxylation rate CO2 adjusting factor for electron-limited carboxylation rate Maintenance respiration (umol CO2/m2/day) Structural nitrogen content (g N/g biomass), SNCm =0.001[33] Rubisco-limited maximum rate of carboxylation rate (umol CO2/m2/s) Rubisco-limited carboxylation rate (umol CO2/m2/s) Electron-limited carboxylation rate (umol CO2/m2/s)

4 5 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2 3

Table 2 Main model inputs for four test cases Test cases

4 5 6 7 8 9 10 11 12 13 14

Time

PAR 1

CO2 (ppm) 370

Daytime hours 15

Day T (oC) 25

Night T (oC) 16

RH

LMA (g/m2)

l 2

3

FMCmtar

Test case 1 ~2 years 1150 0.5 80 0.3 0.038 (Pseudotsuga menziesii) Test case 1 ~2 years 1150 370 15 25 16 0.5 76 0.3 0.026 (Populus euroamericana) Test case 2 control 8-9 1290 370 14 284 234 0.8 85 5 0.2 0.014 (Ambient CO2) years Test case 2 treatment 8-9 1290 570 14 28 4 234 0.8 85 5 0.2 0.014 (Elevated CO2) years Test case 3 control 1-2 450 370 14 30 30 0.8 55 6 0.6 0.02 (High growing T) months Test case 3 treatment 1-2 450 370 14 15 15 0.8 66 7 0.6 0.02 (Low growing T) months Test case 4 control 1-2 450 370 14 15 15 0.8 66 0.6 0.02 (High radiation) months Test case 4 treatment 1-2 50 370 14 15 15 0.8 46 8 0.6 0.02 (Low radiation) months Note 1: PAR=photosynthetic active radiation for nitrogen allocation among carboxylation, light absorption and electron transport (umol photon/m2/s). Data for test case 1 and 2 is from the 10-km gridded data from the SUNNY model [60] averaged during the time 8:00AM-4:00PM in July and data for test case 3, 4 is from the experimental controlled radiation. See Figure S3 for details of hourly radiation. 2:

l

is the proportion of net photosynthesis allocated to leaf. For leaf allocation for test case 1, we set l

be 0.3 using the tree allometry from the ED model [61] based on a seedling of 1cm diameter and 1 meter height. We set

l be

0.2 for test case 2 [62] and 0.6 for test case 3 and 4 based on fast-growing plants non-woody plants [63].3: estimated from mean leaf nitrogen content with eq. ( 5); 4: data are based on July daily minimum and maximum temperature from the DAYMET website. 5: LMA is calculated based on the mean values of old and new leaves in July [64]. 6: data is from Kobayashi et al. [65]. 7: We assume a 20% increase in LMA at the low growing temperature given that the area based leaf nitrogen increased by about 20% at the low growing temperature (see Figure 5 a). 8: We assume a 30% reduction in LMA at the low radiation compared to high radiation [65].

15

26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2

Figures

3 4 5 6 7 8 9 10 11 12 13

Figure 1 Schemes of nitrogen allocation model. The allocation model is based on nitrogen trade-offs among light absorption, electron transport, carboxylation, respiration and storage. Rectangles indicate pools and ovals represent processes. Dashed arrows indicate feedback effects. The plant nitrogen is first portioned into nitrogen for photosynthesis, storage nitrogen, respiratory nitrogen (nitrogen in mitochondria) and structural nitrogen (nitrogen in cell walls and nuclear). Then, photosynthetic nitrogen is further partitioned into nitrogen for light absorption (nitrogen in chlorophyll), electron transport (N in thylakoid except for chlorophyll), carboxylation (nitrogen in Calvin Cycle enzymes). In our model, we propose to balance the allocation of nitrogen among light absorption, electron transport, carboxylation and carbon sink to maximize plant growth given their strategies of trade-off between growth and persistence. We calculate the nitrogen allocation based on the environmental conditions (temperature, radiation, water, and nitrogen availability) that plants have experienced in the past.

14 15 16 17 18 19 20 21

Figure 2 Hierarchical nitrogen allocations at the individual-plant level. The allocation starts from the bottom to top. The whole plant nitrogen is first divided into functional nitrogen and structural nitrogen. Functional nitrogen is then divided into growth nitrogen and storage nitrogen. The growth nitrogen is further divided into photosynthetic nitrogen and respiratory nitrogen, with the photosynthetic nitrogen divided into nitrogen for light harvesting and nitrogen for carboxylation. Notice that light harvesting process includes light absorption and electron transport. Finally, nitrogen allocated for light harvesting is divided into nitrogen for light absorption and nitrogen for electron transport. The parameter in the parenthesis indicates the proportion of nitrogen invested for its category in the same row.

22 23 24 25 26 27 28 29 30 31

Figure 3 Nitrogen fertilization effects on leaf nitrogen allocations for test case one. Panel (a) shows the nitrogen allocation model fitted to observed relationship between Vcmax (scaled to 25oC) and leaf nitrogen for a poplar species (Populus euroamericana) and Douglas-fir (Pseudotsuga menziesii). Data are from Ripullone et al. [13]. The open and filled circles represent observed Vcmax for poplar and Douglas-fir, respectively, with the dashed and solid lines representing the fitted Vcmax by the tuned nitrogen allocation model for poplar and for Douglas-fir. We tune the nitrogen storage duration by varying parameters a, b and c of equation (31) in Text S1 so that the estimates of Vcmax by the allocation model fits to the observations of Vcmax. Panel (b) shows the tuned nitrogen storage duration. Panels (c)-(h) show the predicted proportion of nitrogen allocated for storage, carboxylation, light absorption, electron transport, respiration and structure, respectively. See Table 2 for main model inputs.

32 33 34 35 36 37 38 39

Figure 4 CO2 fertilization effects on leaf nitrogen allocations for test case two. In panel (a), closed and open circles represent observed Vcmax for loblolly pine (Pinus taeda) with ambient (370 ppm) and elevated CO2 concentration (570 ppm), respectively. Solid lines are estimates of Vcmax by the nitrogen allocation model tuned to ambient CO2 data, while dashed lines are predictions of Vcmax by the tuned nitrogen allocation model using elevated CO2 concentration. Panel (b) shows the nitrogen storage duration. Panels (c)-(h) show the fitted (solid lines) and predicted (dashed lines) proportion of nitrogen allocated for storage, carboxylation, light absorption, electron transport, respiration and structure, respectively. See Table 2 for main model inputs.

40 41 42 43 44 45 46

Figure 5 Growing temperature effects on leaf nitrogen allocations for test case three. In panel (a), open and filled circles indicate observed Vcmax for a Japanese plantain (Plantago asiatica) growing at temperatures of 15oC and 30oC, respectively, both of which are scaled to the reference temperature of 25 o C. Plants in both treatments were growing at a relatively high radiation exposure (450 umol/m2/s for 4 hours and 50 umol photon/m2/s for 10 hours). Solid lines are estimates of Vcmax by the nitrogen allocation model tuned to data at the high growing temperature (30 oC), while dashed lines are predictions of Vcmax by the tuned nitrogen allocation model using the low growing temperature (15 oC). Panel (b) shows the 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2 3

tuned nitrogen storage duration. Panels (c)-(h) show the predicted proportion of nitrogen allocated for storage, carboxylation, light absorption, electron transport, respiration and structure, respectively. See Table 2 for main model inputs.

4 5 6 7 8 9 10 11 12 13

Figure 6 Radiation effects on leaf nitrogen allocations for test case four. In panel (a), open and closed circles represent observed Vcmax for a Japanese plantain (Plantago asiatica) growing at a low and high radiation exposure, respectively. See Figure S3 and Table 2 for details of radiation. Plants in both treatments were growing at a relatively low temperature (15oC). Solid lines are predictions of Vcmax by the nitrogen allocation model tuned to data at a high growing temperature (30oC, see filled circles in Figure 5 a) and dashed lines are predictions of Vcmax by the tuned nitrogen allocation model using a radiation level of 300umol/m2/s. Dotted grey lines are predictions of Vcmax by the tuned nitrogen allocation model using a radiation level of 50umol/m2/s. Panel (b) shows the nitrogen storage duration. Panels (c)-(h) show the predicted proportion of nitrogen allocated for storage, carboxylation, light absorption, electron transport, respiration and structure, respectively. See Table 2 for main model inputs.

14 15 16 17 18 19 20 21

Figure S 1 Relationship between Vcmax and leaf nitrogen (g/m2) by incorporating leaf-level variability. The closed circles show the observed Vcmax (scaled to 25oC) for leaves of a popular species (Populus euroamericana, a) and douglas fir (Pseudotsuga menziesii , b) under fertilization [13]. Black solid line indicates the fitted Vcmax by the nitrogen allocation model using mean leaf nitrogen concentration and mean Rubisco proportion for leaves on a single tree. Open circles show the simulated Vcmax by taking into consideration the variability in leaf nitrogen and Rubisco proportion. The variability is assumed to be 10% of the mean value. The grey solid line shows the relationship between Vcmax and leaf nitrogen by fitting a linear model to the simulated Vcmax (open circles).

22 23 24 25 26 27 28

Figure S 2 Hourly photosynthetically active radiation (PAR) for test cases. Data for test case 1 are from the 10-km gridded data estimated by the SUNNY model [60] averaged for July, obtained from the National Solar Radiation Data Base (NSRDB). Data for test case 3, 4 is from the experimental controlled radiation. For test case 2, since no hourly radiation data available, we get the data from NSRDB based on a location with a similar latitude (78.05 W, 40.65 N) assuming a10% filtration by plastic roof. The radiation for test case 4 is for the low radiation environment. For the high radiation environment, the radiation is same as that for test case 3.

29 30 31 32

Figure S 3 CO2 effects on the nitrogen allocated for electron transport (a) and light absorption (b) relative to that for carboxylation. Data are from Crous et al. [38] with 2-3 years of CO2 fertilizations. The solid lines indicate nitrogen allocations for ambient CO2 environment and dashed lines indicate nitrogen allocations for elevated CO2 environment. See Figure 3 for details.

33

28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Literatures 1. Reich PB, Hobbie SE, Lee T, Ellsworth DS, West JB, et al. (2006) Nitrogen limitation constrains sustainability of ecosystem response to CO2. Nature 440: 922-925. 2. Luo Y, Su BO, Currie WS, Dukes JS, Finzi A, et al. (2004) Progressive nitrogen limitation of ecosystem responses to rising atmospheric carbon dioxide. Bioscience 54: 731-739. 3. Thornton PE, Doney SC, Lindsay K, Moore JK, Mahowald N, et al. (2009) Carbon-nitrogen interactions regulate climate-carbon cycle feedbacks: results from an atmosphere-ocean general circulation model. Biogeosciences 6: 2099-2120. 4. Zaehle S, Dalmonech D (2011) Carbon–nitrogen interactions on land at global scales: current understanding in modelling climate biosphere feedbacks. Current Opinion in Environmental Sustainability 3: 311-320. 5. Ainsworth EA, Rogers A (2007) The response of photosynthesis and stomatal conductance to rising [CO2]: mechanisms and environmental interactions. Plant, Cell and Environment 30: 258-270. 6. LeBauer DS, Treseder KK (2008) Nitrogen limitation of net primary productivity in terrestrial ecosystems is globally distributed. Ecology 89: 371-379. 7. Chapin FS, Shaver GR (1985) Individualistic growth response of tundra plant species to environmental manipulations in the field. Ecology 66: 564-576. 8. Shaver GR, Bret-Harte SM, Jones MH, Johnstone J, Gough L, et al. (2001) Species composition interacts with fertilizer to control long-term change in tundra productivity. Ecology 82: 31633181. 9. Aber JD, Ollinger SV, Driscoll CT (1997) Modeling nitrogen saturation in forest ecosystems in response to land use and atmospheric deposition. Ecological Modelling 101: 61-78. 10. Thornton PE, Law BE, Gholz HL, Clark KL, Falge E, et al. (2002) Modeling and measuring the effects of disturbance history and climate on carbon and water budgets in evergreen needleleaf forests. Agricultural and Forest Meteorology 113: 185-222. 11. Reich PB, Walters MB, Kloeppel BD, Ellsworth DS (1995) Different photosynthesis-nitrogen relations in deciduous hardwood and evergreen coniferous tree species. Oecologia 104: 24-30. 12. Friend AD (1991) Use of a model of photosynthesis and leaf microenvironment to predict optimal stomatal conductance and leaf nitrogen partitioning. Plant Cell and Environment 14: 895-905. 13. Ripullone F, Grassi G, Lauteri M, Borghetti M (2003) Photosynthesis–nitrogen relationships: interpretation of different patterns between Pseudotsuga menziesii and Populus × euroamericana in a mini-stand experiment. Tree Physiology 23: 137-144. 14. Tang J, Zhuang Q (2009) A global sensitivity analysis and Bayesian inference framework for improving the parameter estimation and prediction of a process-based Terrestrial Ecosystem Model. Journal of Geophysical Research 114: D15303. 15. Bonan GB, Lawrence PJ, Oleson KW, Levis S, Jung M, et al. (2011) Improving canopy processes in the Community Land Model version 4 (CLM4) using global flux fields empirically inferred from FLUXNET data. Journal of Geophysical Research-Biogeosciences 116. 16. Marschner H (1995) Mineral nutrition of higher plants. San Diego, CA: Academic Press. 889 p. 17. Herms DA, Mattson WJ (1992) The dilemma of plants: to grow or defend. The Quarterly Review of Biology 67: 283-335. 18. Verkroost AWM, Wassen MJ (2005) A simple model for nitrogen-limited plant growth and nitrogen allocation. Annals of Botany 96: 871-876. 19. Chapin FS, Schulze E, Mooney HA (1990) The ecology and economics of storage in plants. Annual Review of Ecology and Systematics 21: 423-447.

29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

20. Evans JR, Poorter H (2001) Photosynthetic acclimation of plants to growth irradiance: the relative importance of specific leaf area and nitrogen partitioning in maximizing carbon gain. Plant, Cell & Environment 24: 755-767. 21. Evans JR (1989) Photosynthesis and nitrogen relationships in leaves of C 3 and C4 plants. Oecologia 78: 9-19. 22. Bloom AJ, Chapin FS, Mooney HA (1985) Resource limitation in plants - an economic analogy. Annual Review of Ecology and Systematics 16: 363-392. 23. Millard P (1988) The accumulation and storage of nitrogen by herbaceous plants. Plant, Cell & Environment 11: 1-8. 24. Kleijn D, Treier UA, Müller-Schärer H (2005) The Importance of Nitrogen and Carbohydrate Storage for Plant Growth of the Alpine Herb Veratrum album. New Phytologist 166: 565-575. 25. Hikosaka K, Terashima I (1995) A model of the acclimation of photosynthesis in the leaves of C3 plants to sun and shade with respect to nitrogen use. Plant, Cell & Environment 18: 605-618. 26. Niinemets U, Tenhunen JD (1997) A model separating leaf structural and physiological effects on carbon gain along light gradients for the shade-tolerant species Acer saccharum. Plant Cell and Environment 20: 845-866. 27. Makino A, Osmond B (1991) Effects of nitrogen nutrition on nitrogen partitioning between chloroplasts and mitochondria in pea and wheat. Plant Physiology 96: 355-362. 28. Coley PD, Bryant JP, Chapin FS (1985) Resource Availability and Plant Antiherbivore Defense. Science 230: 895-899. 29. Dewar RC, Franklin O, Mäkelä A, McMurtrie RE, Valentine HT (2009) Optimal Function Explains Forest Responses to Global Change. Bioscience 59: 127-139. 30. Baker DN, Hesketh JD, Duncan WG (1971) Simulation of growth and yield in cotton.I. gross photosynthesis, respiration, and growth. Crop Science 12: 431-435. 31. Williams K, Percival F, Merino J, Mooney HA (1987) Estimation of tissue construction cost from heat of combustion and organic nitrogen content. Plant, Cell & Environment 10: 725-734. 32. Ryan MG (1991) A simple method for estimating gross carbon budgets for vegetation in forest ecosystems. Tree Physiology 9: 255-266. 33. White MA, Thornton PE, Running SW, Nemani RR (2000) Parameterization and sensitivity analysis of the BIOME–BGC terrestrial ecosystem model: net primary production controls. Earth Interactions 4: 1-85. 34. Tenhunen JD, Weber JA, Yocum CS, Gates DM (1976) Development of a Photosynthesis Model with an Emphasis on Ecological Applications. II. Analysis of a Data Set Describing the $P_{M}$ Surface. Oecologia 26: 101-119. 35. Farquhar GD, Caemmerer S, Berry JA (1980) A biochemical model of photosynthetic CO 2 assimilation in leaves of C3 species. Planta 149: 78-90. 36. Long SP (1991) Modification of the response of photosynthetic productivity to rising temperature by atmospheric CO2 concentrations - has its importance been underestimated. Plant Cell and Environment 14: 729-739. 37. Oleson KW, Lawrence DM, B.Bonan G, Flanner MG, Kluzek E, et al. (2010) Technical description of version 4.0 of the Community Land Model (CLM). Boulder, CO: National Center for Atmospheric Research. 257 p. 38. Crous KY, Walters MB, Ellsworth DS (2008) Elevated CO2 concentration affects leaf photosynthesisnitrogen relationships in Pinus taeda over nine years in FACE. Tree Physiology 28: 607-614. 39. Hikosaka K (2005) Nitrogen partitioning in the photosynthetic apparatus of Plantago asiatica leaves grown under different temperature and light conditions: Similarities and differences between temperature and light acclimation. Plant and Cell Physiology 46: 1283-1290.

30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

40. Peterson AG, Ball JT, Luo YQ, Field CB, Reich PB, et al. (1999) The photosynthesis leaf nitrogen relationship at ambient and elevated atmospheric carbon dioxide: a meta-analysis. Global Change Biology 5: 331-346. 41. Kattge J, Knorr W, Raddatz T, Wirth C (2009) Quantifying photosynthetic capacity and its relationship to leaf nitrogen content for global-scale terrestrial biosphere models. Global Change Biology 15: 976-991. 42. Collatz GJ, Ball JT, Grivet C, Berry JA (1991) Physiological and environmental regulation of stomatal conductance, photosynthesis and transpiration: a model that includes a laminar boundary layer. Agricultural and Forest Meteorology 54: 107-136. 43. Holaday AS, Martindale W, Alred R, Brooks AL, Leegood RC (1992) Changes in activities of enzymes of carbon metabolism in leaves during exposure of plants to low-temperature. Plant Physiology 98: 1105-1114. 44. Ehleringer J, Björkman O (1977) Quantum Yields for CO2 Uptake in C3 and C4 Plants: Dependence on Temperature, CO2, and O2 Concentration. Plant Physiology 59: 86-90. 45. Gilbert ME, Zwieniecki MA, Holbrook NM (2011) Independent variation in photosynthetic capacity and stomatal conductance leads to differences in intrinsic water use efficiency in 11 soybean genotypes before and during mild drought. Journal of Experimental Botany 62: 2875-2887. 46. Urban O, Ač A, Kalina J, Priwitzer T, Šprtová M, et al. (2007) Temperature dependences of carbon assimilation processes in four dominant species from mountain grassland ecosystem. Photosynthetica 45: 392-399. 47. Nordin A, Nasholm T (1997) Nitrogen storage forms in nine boreal understorey plant species. Oecologia 110: 487-492. 48. Lloyd J, Patiño S, Paiva RQ, Nardoto GB, Quesada CA, et al. (2010) Optimisation of photosynthetic carbon gain and within-canopy gradients of associated foliar traits for Amazon forest trees. Biogeosciences 7: 1833-1859. 49. Seiwa K (2007) Trade-offs Between Seedling Growth and Survival in Deciduous Broadleaved Trees in a Temperate Forest. Annals of Botany 99: 537-544. 50. Warren CR, Adams MA (2002) Phosphorus affects growth and partitioning of nitrogen to Rubisco in Pinus pinaster. Tree Physiology 22: 11-19. 51. Reich P, Oleksyn J, Wright I (2009) Leaf phosphorus influences the photosynthesis–nitrogen relation: a cross-biome analysis of 314 species. Oecologia 160: 207-212. 52. Mack MC, Schuur EAG, Bret-Harte MS, Shaver GR, Chapin FS (2004) Ecosystem carbon storage in arctic tundra reduced by long-term nutrient fertilization. Nature 431: 440-443. 53. Bret-Harte MS, Shaver GR, Zoerner JP, Johnstone JF, Wagner JL, et al. (2001) Developmental plasticity allows Betula nana to dominate tundra subjected to an altered environment. Ecology 82: 18-32. 54. Swann AL, Fung IY, Levis S, Bonan GB, Doney SC (2010) Changes in Arctic vegetation amplify highlatitude warming through the greenhouse effect. Proceedings of the National Academy of Sciences 107: 1295-1300. 55. Taub DR, Wang X (2008) Why are nitrogen concentrations in plant tissues lower under elevated CO 2? A critical examination of the hypotheses. Journal of Integrative Plant Biology 50: 1365-1374. 56. Long SP, Ainsworth EA, Rogers A, Ort DR (2004) Rising atmospheric carbon dioxide: plants face the future. Annual Review of Plant Biology 55: 591-628. 57. Leegood RC, Edwards GE (2004) Carbon Metabolism and Photorespiration: Temperature Dependence in Relation to Other Environmental Factors. In: Baker NR, editor. Photosynthesis and the Environment: Springer Netherlands. pp. 191-221.

31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

58. Makino A, Nakano H, Mae T (1994) Effects of growth temperature on the responses of ribulose-1,5biphosphate carboxylase, electron transport components, and sucrose synthesis enzymes to leaf nitrogen in rice, and their relationships to photosynthesis. Plant Physiology 105: 1231-1238. 59. Haldimann P (1998) Low growth temperature-induced changes to pigment composition and photosynthesis in Zea mays genotypes differing in chilling sensitivity. Plant, Cell & Environment 21: 200-208. 60. Perez R, Ineichen P, Moore K, Kmiecik M, Chain C, et al. (2002) A new operational model for satellitederived irradiances: Description and validation. Solar Energy 73: 307-317. 61. Moorcroft PR, Hurtt GC, Pacala SW (2001) A method for scaling vegetation dynamics: the ecosystem demography model (ED). Ecological Monographs 71: 557-586. 62. McCarthy HR, Oren R, Johnsen KH, Gallet-Budynek A, Pritchard SG, et al. (2010) Re-assessment of plant carbon dynamics at the Duke free-air CO(2) enrichment site: interactions of atmospheric [CO(2)] with nitrogen and water availability over stand development. New Phytologist 185: 514528. 63. Poorter H, Remkes C, Lambers H (1990) Carbon and Nitrogen Economy of 24 Wild-Species Differing in Relative Growth-Rate. Plant Physiology 94: 621-627. 64. Rogers A, Ellsworth DS (2002) Photosynthetic acclimation of Pinus taeda (loblolly pine) to long-term growth in elevated pCO2 (FACE). Plant, Cell & Environment 25: 851-858. 65. Kobayashi T, Okamoto K, Hori Y (2001) Variations in size structure, growth and reproduction in Japanese plantain (Plantago asiatica L.) between exposed and shaded populations. Plant Species Biology 16: 13-28.

32

Figure 1 Click here to download high resolution image

Figure 2 Click here to download high resolution image

Figure 3 Click here to download high resolution image

Figure 4 Click here to download high resolution image

Figure 5 Click here to download high resolution image

Figure 6 Click here to download high resolution image

Text S1 Click here to download Supporting Information: Text S1.doc

Text S2 Click here to download Supporting Information: Text S2.doc

Text S3 Click here to download Supporting Information: Text S3.doc

Text S4 Click here to download Supporting Information: Text S4.doc

Text S5 Click here to download Supporting Information: Text S5.doc

Figure S1 Click here to download Supporting Information: Figure S1.tiff

Figure S2 Click here to download Supporting Information: Figure S2.tiff

Figure S3 Click here to download Supporting Information: Figure S3.tiff