## Abstract

Aims

Rising atmospheric CO2 has been shown to increase aboveground net primary productivity (ANPP) in water-limited perennial grasslands, in part by reducing stomatal conductance and transpiration, thereby reducing depletion of soil moisture. However, the benefits of CO2 enrichment for ANPP will vary with soil type and may be reduced if water limitation is low. Little is known about CO2 effects on ANPP of Panicum virgatum, a perennial C4 tallgrass and potential bioenergy crop. We hypothesized that if water limitation is minimized, (i) CO2 enrichment would not increase P. virgatum ANPP because photosynthetic rates of this C4 grass would not increase and because decreased transpiration at elevated CO2 would provide little additional benefit in increased soil moisture and (ii) soil type will have little effect on P. virgatum CO2 responses because of high overall soil moisture.

Methods

Growth and leaf physiology of P. virgatum cv. ‘Alamo’ were studied as plants established for 4 years on silty clay and clay soils along a 250 to 500 μl l-1 gradient in atmospheric CO2 located in central Texas, USA. Plants were watered to replace evapotranspiration, fertilized with NO3NH4 and P2O5 and clipped to standard height during mid-season.

Important Findings

ANPP increased through the third year of growth. Soil moisture (0–20 cm), ANPP, tiller numbers and leaf area index were 8–18% higher on the clay than on the silty clay soil. ANPP did not increase with CO2 except in the planting year. However, biomass removed with clipping strongly increased with CO2 in years 2 and 3, suggesting that CO2 enrichment increased the early- to mid-season growth of establishing P. virgatum but not later regrowth or that of fully established plants. Furthermore, CO2 enrichment differentially affected two components of ANPP in years 2 and 3, increasing tiller mass and reducing tiller numbers. This reallocation of resources in clipped P. virgatum suggested increased meristem limitation of productivity with CO2 enrichment. CO2 enrichment had little effect on photosynthesis but increasingly reduced stomatal conductance and transpiration as the plants established. As a result, water use efficiency became increasingly coupled to CO2 as leaf area increased during establishment. These results suggest that for well-watered and clipped P. virgatum, ANPP differed between soil types, was not affected by CO2 enrichment when fully established but interacted with clipping to alter allocation patterns during establishment. Soil type effects on ANPP-CO2 responses will likely become more apparent when water is more limiting.

## INTRODUCTION

Atmospheric CO2 concentrations have increased from around 270 μl l-1 in the pre-industrial period to the present level of 390 μl l-1 and are expected to surpass 500 μl l-1 by 2050 (Forster et al. 2007). Rising CO2 represents a chronic and cumulative change (Smith et al. 2009) in the availability of an essential resource for plant growth and productivity. CO2 effects on ecosystem primary productivity will depend on the direct effects of CO2 on plant carbon gain, on the indirect effects of CO2 on other ecosystem controls such as soil moisture and nitrogen availability and on the extent to which the indirect CO2 effects either reinforce or offset the direct effects (Polley et al. 2011b).

Several mechanisms have been advanced to explain how CO2 enrichment may increase plant growth and productivity. CO2 enrichment directly affects photosynthetic processes in several ways, including increased carboxylation efficiency (C3 species) and reduced stomatal conductance and transpiration (C3 or C4 species), which can increase photosynthetic water use efficiency (Ainsworth and Long 2005; Anderson et al. 2001; Drake et al. 1997). Indirect effects of reduced transpiration at elevated CO2 can include increase soil moisture availability (Lecain et al. 2003; Morgan et al. 2001) and soil nitrogen mineralization (Austin et al. 2004, Dijkstra et al. 2008). Furthermore, in multispecies systems, species abundances may begin to change once CO2 enrichment exceeds the capacity of some species to respond, and when other, potentially more productive species are favored by new combinations of CO2 and resource availability (Smith et al. 2009).

However, several factors could limit productivity increases from CO2 enrichment. Lack of water limitation, such as in high rainfall years or on soils with high water holding capacity, could limit soil moisture gains accruing from reduced transpiration. In addition, other resources, such as mineral nitrogen (N) availability (Gill et al. 2002; Norby et al. 2010 but see Dijkstra et al. 2008) may limit productivity gains from CO2 enrichment (Luo et al. 2004). Dilution of the N concentration in plant tissues (Ainsworth and Long 2005) may result in lower quality litter inputs. Finally, in monoculture production systems, species reordering is not available as a mechanism by which CO2 enrichment can increase productivity.

While these mechanisms are largely supported by studies in natural mixed-species ecosystems (Knapp et al. 1996, Owensby et al. 1996) and in annual agricultural crops (Kimball et al. 1999; Leakey 2009), an important gap remains regarding the CO2 response of perennial grasses in monoculture. Dijkstra et al. (2010) found that the effects of CO2 enrichment on the productivity of monocultures of perennial C4 grasses were not predictable from species responses in mixtures, but there have been few other studies (Oliver et al. 2009). The importance of studying perennial monocultures is heightened by the increased interest in their potential for bioenergy production (Perlack et al. 2005).

We studied the growth and productivity of monocultures of a mesic C4 tallgrass, Panicum virgatum L. (cv. Alamo), growing at CO2 ranging from 250 to 500 μl l-1. The study was conducted over 4 years (2007–2010) as P. virgatum established on two soils collected from upper (silty clay) and lower (clay) topographic positions in the Blackland Prairie region of central Texas, USA. These soils represent the typical range of texture, N and C contents, and hydrologic properties on which native grasslands (which would have included P. virgatum) would be found and on which P. virgatum would likely be grown for bioenergy production in the Blackland Prairie region. The monocultures were well watered and fertilized. With relatively abundant nitrogen and soil moisture, we hypothesized that biomass production would show little response to CO2 enrichment because the enhancement of physiological efficiencies that may arise from CO2 enrichment (i.e., increased water use efficiency or nitrogen use efficiency) would accrue little additional soil moisture, and thus provide little additional benefit to plant growth. We also hypothesized that soil type effects on biomass production and CO2 responses would be minimal because of the well-watered conditions.

## MATERIALS AND METHODS

### Study site

The study was conducted in the Lysimeter CO2 Gradient (LYCOG) facility, located at Temple, TX, USA (31°05’ N, 97°20’ W) in the southern US Central Plains. Panicum virgatum is a warm season C4 grass and native throughout the region. The climate at this site is subtropical, falling in the transition between humid and sub-humid zones. Mean annual precipitation is 914 mm (1971–2000), with growing season wet periods in May–June and September–October and a pronounced July–August dry period. Temperatures range from a July–August mean maximum of 35°C to a December mean minimum of 2.9°C. The mean frost-free period is ∼ 250°, from mid-March to late November.

The LYCOG facility consists of closed outdoor chambers that maintain a continuous linear gradient in CO2 concentration of 250 to 500 μl l-1, spanning pre-industrial to anticipated mid-21st century values. The facility is described in detail in Fay et al. (2009). LYCOG consists of two linear chambers. Each chamber is 1.2-m wide, 1.5-m tall and consists of 10–5 m long sections. Each section contained four steel-encased 1 × 1 × 1.5 m deep intact soil monoliths enclosed in water-tight steel boxes. One chamber contains the superambient portion of the gradient, created by introducing air enriched to 500 μl l-1 CO2. Blowers advect the air, and photosynthesis by the enclosed vegetation progressively depletes the air of CO2, which exits the chamber at 380 μl l-1. Similarly, on the second linear chamber, the subambient portion of the gradient is created when ambient air is introduced and progressively depleted of CO2, exiting the chamber at 250 μl l-1.

Each section is enclosed with clear polyethylene (0.006″/0.15 mm). This film transmits >90% of incident light with minimal effects on spectral quality and is similar to polyethylene films used in other global change experiments. Zippered openings in the polyethylene backed by draft flaps allow access to the monoliths for sampling. During winter, the polyethylene enclosures are removed and replaced with rain exclusion covers, exposing the dormant vegetation to the ambient atmosphere but continuing to exclude precipitation. The chambers are then reclosed with new polyethylene at the beginning of the next growing season to minimize the effects of photodegradation on light transmittance.

Average air temperature in the chambers is controlled to match outside ambient air temperature by passing the chamber airflow through a chilled water cooling coil separating each 5-m section. CO2 treatments are maintained for the portion of the growing season when the vegetation has adequate assimilation capacity to maintain the gradient, typically early May to late October.

The chambers contained 80 hydrologically isolated intact soil monoliths (1.5 m3), of which 20 contained P. virgatum. The remaining 60 monoliths support tallgrass prairie communities that did not include P. virgatum and are considered elsewhere (Polley et al. 2008, 2011a).

The 20 P. virgatum monoliths came from two soil orders, a silty clay soil typical of uplands (n = 8, Austin series, Mollisol, Udorthentic Haplustol) and a moist high organic matter heavy clay found downslope from the silty clay (n = 12, Houston Black series, Vertisol, Udic Haplustert). These soils were representative of their respective landscape positions in the Blackland Prairie region where the experiment was conducted. The soils were collected as intact soil monoliths (1 × 1 × 1.5 m deep) in 2002, during construction of LYCOG. The monoliths originally supported mixed native prairie vegetation which was removed by hand weeding and glyphosate application. Panicum virgatum was planted in May 2007 at a density of 50 live tillers per monolith. Tillers that died in 2007 were replaced in May 2008. The monoliths were arranged in a stratified random design along the CO2 gradient, interspersed among the prairie monoliths. Each P. virgatum monolith was individually drip irrigated. We irrigated each monolith to field capacity at the start of each growing season and irrigated weekly during each growing season to replace evapotranspiration (ET). We estimated ET from the average change in weight of three monoliths of each soil type exposed to either subambient or elevated CO2 during 2007 through 2009. Beginning in 2010, all the monoliths were weighed, and ET was calculated for each monolith. Irrigation applications were controlled and logged by a data logger. Each monolith was fertilized with the equivalent of 42 kg N/ha and 22.4 kg P/ha in April 2007 and with 168 kg N/ha and 44.8 kg P/ha in April 2008–2010. N was applied as NH4NO3. P was applied as P2O5. Fertilizer amounts were chosen to minimize N limitation.

Plants in all monoliths were clipped to 70-cm height two (2009–2010) or three (2008) times per growing season during June–August to prevent plants from outgrowing the chamber volume. The height was chosen to leave ample leaf area for regrowth and not to mimic complete harvests.

### Soil and plant measurements

Volumetric soil water content for 0–20 cm (Θ20) of each monolith was measured biweekly with a calibrated neutron attenuation probe (503DR Hydroprobe, CPN International, Concord, CA, USA) at permanently installed access tubes.

Aboveground net primary productivity (ANPP) was determined from the mid-season clipped biomass plus the end of year (November) standing biomass. Tiller numbers were counted at the final harvest. The end of year biomass was clipped at 10 cm above the soil surface. All biomass was dried to constant mass for 72 h at 60°C and then weighed. The leaf area index (LAI) in each monolith was measured each July and August using a SunScan canopy analysis system (Delta-T Devices Ltd., Burwell, Cambridge, UK).

Leaf-level carbon and water exchange were measured in each monolith during June 2008, May and July 2009 and June 2010. Gas exchange was measured on one or two leaves on two tillers per monolith between 0900 and 1400 local time. The chosen tillers had typical vigor for that soil type and CO2, and the selected leaves were recently fully expanded and also of typical vigor. Leaves were measured for net carbon assimilation (ACO2), stomatal conductance (gs), and transpiration (E) with an infrared gas analyzer (LI-6400 LICor Biosciences, Inc., Lincoln, NE, USA) using a 2 × 3 cm leaf cuvette, CO2 mixer and 85:15 red:blue light source. Leaf chamber illumination was controlled at 1500 μmol m-2 s-1 photon flux density. Leaf temperature varied between 30 and 35°C, and the leaf cuvette H2O mole fraction was controlled at 15, 17, and 20 mmol mol-1 in 2008, 2009 and 2010, respectively, corresponding to humidity levels in the gradient during the measurements. Cuvette CO2 was controlled to values corresponding to the position along the gradient. Instantaneous photosynthetic water use efficiency (WUEt) was computed as ACO2 E-1.

Leaves were collected immediately following the gas exchange measurements and measured for leaf water potential (Ψl) using a pressure chamber (Model 1000, PMS Instruments, Corvallis, OR, USA). The leaves were then dried, ground to a fine powder and assayed for carbon and nitrogen content in an elemental analyzer (Flash 2000, Thermo Scientific, Waltham, MA, USA).

### Data analysis

Statistical analyses of the effects of CO2 and soil type on soil moisture and plant growth and physiology of P. virgatum were conducted in SAS 9.2. Weekly Θ20 measurements and the July/August LAI measurements were averaged to yield a single value per monolith for each growing season, to correspond to the ANPP data. We computed the ratio of the sum of mid-season biomass removed to the end of season biomass as an index of clipping intensity. The monolith was the experimental unit in these analyses.

Analyses proceeded in three steps. First, we fit a repeated measures model (Equation 1) to test for soil differences in ANPP, LAI, tiller number and tiller mass using a mixed models procedure (Proc Mixed):

(1)
$yijkl=intercept+soili+monolithj(soili)+yeark+soil×yearik+eijkl$

Second, to test for CO2 effects and their interaction with soil type and year, we fit an expanded model to the plant growth and physiology variables (Equation 2):

(2)
$yijkl=intercept+soili+monolithj(soili)+α(CO2)+βi(CO2×soili)+yeark+soil×yearik+γK(CO2×yeark)+δik(CO2×soili×yeark)+eijkl$

In both models, soil was a fixed effect, monoliths nested within soils [monolith(soil)] a random effect, CO2 a covariate and year a repeated effect. Non-significant terms in this initial model were removed when this improved the fit statistics (Akaike’s Information Criteria).

Third, we fit linear regressions of each response variable versus CO2 for the soils separately and combined and for the years separately and combined (Table 2). When the ANCOVA model (equation 2) returned significant CO2, soil × CO2, soil × year or soil × CO2 × year effects, the corresponding regressions were plotted in Figs 3 and 4.

## RESULTS

### Soil moisture

Θ20 was higher on the clay (34.4% ± 0.4), compared to the silty clay soil (31.6 ± 0.4%, P < 0.0001) and increased weakly with CO2 for both soils combined (R2 = 0.04, P = 0.005, Fig. 1). Θ20 also varied between years (31.5 ± 0.4 in 2009 to 33.8 ± 0.4 in 2010, P = 0.013). We found no evidence for interactions among CO2, soil and year (0.09 < P < 0.68).

Figure 1:

Volumetric soil water content of clay and silty clay soils along the CO2 gradient, and (inset) mean of all CO2 concentrations. Each point is the average of 2007–2010 (± standard error). Linear regression is for both soils combined (Table 2).

Figure 1:

Volumetric soil water content of clay and silty clay soils along the CO2 gradient, and (inset) mean of all CO2 concentrations. Each point is the average of 2007–2010 (± standard error). Linear regression is for both soils combined (Table 2).

### ANPP and components

Establishment year ANPP of P. virgatum averaged 160 g m-2, increased 8-fold in 2008, followed by a further doubling in 2009, with no further change in ANPP in 2010 (year P < 0.0001, Fig. 2a). Tiller numbers and mass both increased during establishment in a similar manner as ANPP (P < 0.0001, Fig. 2b and c), whereas LAI increased each year (Fig. 2d). Averaged across all years and CO2 concentrations, ANPP, tiller number and LAI were 12–18% higher on the clay soil than on the silty clay (P ≤ 0.046, Table 1). Clipping intensity was 38% in 2008, decreasing to 20% in 2010 (P = 0.0003, Fig. 3d).

Table 1:

biomass and physiological variables by soil type, averaged across years and CO2

 Silty clay (SE) Clay (SE) F P-value SWC (%vol) 31.59 (0.36) 34.43 (0.36) 28.8 <0.0001 ANPP (g m-2) 1383.03 (157.97) 1580.28 (133.66) 6.9 0.0180 Tiller # (m-2) 263.47 (23.35) 309.77 (20.86) 12.9 0.0006 Tiller mass (g) 4.46 (0.33) 4.48 (0.27) 1.3 0.2724 LAI (m2 m-2) 1.97 (0.16) 2.20 (0.13) 4.6 0.0465 ACO2 (μmol m-2 s-1) 22.62 (0.71) 21.38 (0.61) 1.8 0.1834 gs (mol m-2 s-1) 0.19 (0.02) 0.19 (0.01) 0.1 0.8180 E (mmol m-2 s-1) 4.21 (0.35) 4.23 (0.18) 0.0 0.9986 WUE (ACO2 E-1) 6.21 (0.49) 5.46 (0.31) 1.4 0.2513 Ψl (MPa) −1.19 (0.05) −1.33 (0.03) 3.3 0.0875 Leaf %C 45.89 (0.12) 45.30 (0.09) 0.7 0.4070 Leaf %N 1.86 (0.09) 1.79 (0.08) 1.0 0.3338 Leaf C:N Ratio 25.64 (1.27) 26.78 (1.25) 0.6 0.4653
 Silty clay (SE) Clay (SE) F P-value SWC (%vol) 31.59 (0.36) 34.43 (0.36) 28.8 <0.0001 ANPP (g m-2) 1383.03 (157.97) 1580.28 (133.66) 6.9 0.0180 Tiller # (m-2) 263.47 (23.35) 309.77 (20.86) 12.9 0.0006 Tiller mass (g) 4.46 (0.33) 4.48 (0.27) 1.3 0.2724 LAI (m2 m-2) 1.97 (0.16) 2.20 (0.13) 4.6 0.0465 ACO2 (μmol m-2 s-1) 22.62 (0.71) 21.38 (0.61) 1.8 0.1834 gs (mol m-2 s-1) 0.19 (0.02) 0.19 (0.01) 0.1 0.8180 E (mmol m-2 s-1) 4.21 (0.35) 4.23 (0.18) 0.0 0.9986 WUE (ACO2 E-1) 6.21 (0.49) 5.46 (0.31) 1.4 0.2513 Ψl (MPa) −1.19 (0.05) −1.33 (0.03) 3.3 0.0875 Leaf %C 45.89 (0.12) 45.30 (0.09) 0.7 0.4070 Leaf %N 1.86 (0.09) 1.79 (0.08) 1.0 0.3338 Leaf C:N Ratio 25.64 (1.27) 26.78 (1.25) 0.6 0.4653
Figure 2:

Biomass components on the clay and silty clay soil during the establishment of Panicum virgatum in the CO2 gradient experiment. Each point is the average over all CO2 concentrations (± standard error). (a) Aboveground net primary productivity (ANPP), (b) number of tillers per monolith, (c) individual tiller mass and (d) Leaf area index (LAI).

Figure 2:

Biomass components on the clay and silty clay soil during the establishment of Panicum virgatum in the CO2 gradient experiment. Each point is the average over all CO2 concentrations (± standard error). (a) Aboveground net primary productivity (ANPP), (b) number of tillers per monolith, (c) individual tiller mass and (d) Leaf area index (LAI).

Figure 3:

Biomass components of Panicum virgatum along the CO2 gradient on clay (open symbols) and silty clay (closed symbols) soils from the planting year (2007) through full establishment (2010). (a) Aboveground net primary productivity (ANPP), (b) individual tiller mass, (c) number of tillers per monolith, (d) clipping intensity, the ratio of June–August clipped biomass to end of season standing biomass. Regressions are shown for years when significant relationships (P < 0.05) were found for both soils combined (Table 2).

Figure 3:

Biomass components of Panicum virgatum along the CO2 gradient on clay (open symbols) and silty clay (closed symbols) soils from the planting year (2007) through full establishment (2010). (a) Aboveground net primary productivity (ANPP), (b) individual tiller mass, (c) number of tillers per monolith, (d) clipping intensity, the ratio of June–August clipped biomass to end of season standing biomass. Regressions are shown for years when significant relationships (P < 0.05) were found for both soils combined (Table 2).

Averaged over both soils, ANPP increased significantly with CO2 enrichment in the establishment year (R2 = 0.21) but not thereafter (CO2 × year P = 0.003, Fig. 3a, Table 2). The contributions of tiller number vs. tiller mass to ANPP changed as P. virgatum established. In the establishment year, ANPP increased with CO2 enrichment because tiller mass increased with CO2 (R2 = 0.29, P = 0.0143, Fig. 3b). As establishment progressed, tiller numbers decreased with CO2 enrichment (0.03 < P < 0.07, Fig. 3c), whereas tiller masses increased, most strongly on the clay soil in 2008 and 2009 (R2 = 0.55–0.79, P < 0.006). However, once P. virgatum was fully established in 2010, tiller number and mass no longer varied with CO2 (P > 0.19). Clipping intensity increased with CO2 enrichment in both 2008 and 2009 (R2 = 0.66–0.69, P < 0.0001) because clipped biomass increased with CO2 enrichment on the clay soil in 2008 and on both soils in 2009 (R2 = 0.64–0.72, P < 0.008, Table 2). In contrast, end of season standing biomass in general did not vary with CO2. No significant CO2, soil × CO2 or soil × year interactions were found for LAI.

Table 2:

regression parameters and statistical tests for the growth and physiological responses to CO2 in Panicum virgatum

 Silty clay Clay Both soils Year Intercept Slope R2 F P Intercept Slope R2 F P Intercept Slope R2 F P ANPP (g m-2) 2007 −67.3 0.469 0.45 4.9 0.0681 −263.6 1.276 0.45 8.3 0.0165 −81.8 0.646 0.21 4.9 0.0396 2008 1670.2 −1.506 0.24 1.9 0.2133 860.3 1.599 0.26 3.1 0.1092 1511.5 −0.634 0.03 0.5 0.4702 2009 1792.4 1.028 0.33 3.0 0.1349 2037.4 1.045 0.11 1.2 0.3006 2143.6 0.499 0.03 0.6 0.4432 2010 2176.6 −0.098 0.00 0.0 0.9394 2481.8 −0.488 0.03 0.3 0.6195 2490.0 −0.666 0.05 0.9 0.3621 All years 1334.1 0.120 0.00 0.0 0.9525 1009.9 1.614 0.01 0.7 0.4167 1364.1 0.366 0.00 0.1 0.7819 # tillers 2007 2.1 0.172 0.38 3.7 0.1034 12.8 0.241 0.25 3.3 0.1007 39.4 0.128 0.10 2.0 0.1724 2008 307.2 −0.184 0.15 1.0 0.3472 369.0 −0.208 0.12 1.4 0.2600 384.0 −0.303 0.23 5.4 0.0327 2009 332.1 0.097 0.02 0.1 0.7187 639.3 −0.608 0.44 7.8 0.0188 532.8 −0.345 0.18 4.0 0.0607 2010 378.1 −0.001 0.00 0.0 0.9923 585.8 −0.454 0.21 2.6 0.1360 525.2 −0.318 0.17 3.7 0.0701 All years 246.3 0.042 0.00 0.0 0.8874 361.5 −0.146 0.00 0.2 0.6379 347.0 −0.149 0.01 0.5 0.4646 Tiller mass (g) 2007 0.75 0.002 0.31 2.7 0.1539 −1.06 0.008 0.49 9.6 0.1130 0.07 0.004 0.29 7.4 0.0143 2008 5.49 −0.002 0.16 1.1 0.3310 1.68 0.009 0.55 12.2 0.0059 3.80 0.003 0.70 1.4 0.2602 2009 5.45 0.001 0.03 0.2 0.6924 0.17 0.012 0.79 38.3 0.0001 3.49 0.006 0.37 10.5 0.0045 2010 5.75 0.000 0.00 0.0 0.9544 3.88 0.005 0.16 1.9 0.1940 4.68 0.002 0.06 1.3 0.2789 All years 4.22 0.001 0.00 0.0 0.8838 1.10 0.010 0.13 6.6 0.0133 2.74 0.005 0.04 3.1 0.0807 Clipped mass (g m-2) 2008 214.0 0.024 0.04 0.3 0.6143 −332.4 1.958 0.69 22.2 0.0008 −7.2 0.922 0.27 6.6 0.0195 2009 170.1 0.948 0.72 15.4 0.0078 −158.6 1.873 0.64 17.8 0.0018 11.9 1.368 0.60 26.5 <0.0001 2010 272.4 0.195 0.01 0.1 0.7784 455.5 −0.194 0.02 0.2 0.7005 405.7 −0.089 0.00 0.1 0.8070 All years 198.5 0.512 0.07 1.8 0.1991 −63.9 1.356 0.33 16.4 0.0003 104.0 0.822 0.16 11.4 0.0013 End of year mass (g m-2) 2008 1456.0 −1.745 0.49 5.7 0.0542 1192.7 −0.359 0.03 0.3 0.6165 1518.7 −1.555 0.28 7.0 0.0163 2009 1622.4 0.079 0.00 0.0 0.8880 2196.0 −0.828 0.09 1.1 0.3307 2131.7 −0.869 0.12 2.4 0.1387 2010 1904.2 −0.293 0.03 0.2 0.6639 2026.3 −0.294 0.02 0.2 0.6637 2084.3 −0.578 0.08 1.6 0.2163 All years 1193.4 −0.285 0.00 0.03 0.8579 1089.2 0.504 0.00 0.1 0.7514 1303.6 −0.299 0.00 0.1 0.7763 Clipped: end of year 2008 −0.05 0.001 0.57 8.0 0.0303 −0.40 0.002 0.77 33.6 0.0002 −0.25 0.002 0.69 40.5 <0.0001 2009 0.11 0.001 0.64 10.7 0.0172 −0.15 0.001 0.71 24.1 0.0006 −0.05 0.001 0.66 35.4 <0.0001 2010 0.15 0.000 0.02 0.1 0.7522 0.22 0.000 0.00 0.0 0.8705 0.19 0.000 0.00 0.0 0.9510 All years 0.04 0.001 0.16 4.31 0.0497 −0.14 0.001 0.36 19.03 0.0001 −0.06 0.001 0.29 23.6 <0.0001 ACO2 (μmol m-2 s-1) 2008 17.5 0.006 0.03 0.1 0.7292 3.8 0.043 0.47 7.1 0.0283 10.0 0.025 0.25 4.9 0.0425 2009 28.8 −0.011 0.06 0.4 0.5488 19.8 0.008 0.06 0.6 0.4609 22.4 0.003 0.00 0.1 0.7834 2010 28.2 −0.012 0.47 5.4 0.0592 17.6 0.012 0.13 1.5 0.2487 21.0 0.004 0.02 0.4 0.5507 All years 24.1 −0.004 0.01 0.2 0.6901 13.7 0.022 0.18 6.9 0.0132 17.7 0.011 0.06 3.4 0.0691 gS (mol m-2 s-1) 2008 0.42 −0.001 0.67 10.0 0.0251 0.12 0.000 0.05 0.4 0.5245 0.24 0.000 0.09 1.5 0.2479 2009 0.50 −0.001 0.86 37.5 0.0009 0.29 0.000 0.46 8.4 0.0158 0.36 0.000 0.55 22.5 0.0002 2010 0.73 −0.001 0.80 23.9 0.0027 0.77 −0.002 0.79 37.1 0.0001 0.72 −0.001 0.77 61.9 <0.0001 All years 0.54 −0.001 0.70 49.5 <0.0001 0.38 −0.001 0.22 9.0 0.0053 0.43 −0.001 0.39 34.5 <0.0001 E (mmol m-2 s-1) 2008 11.0 −0.015 0.68 10.5 0.0230 4.3 0.001 0.01 0.1 0.7703 6.9 −0.005 0.15 2.6 0.1291 2009 11.6 −0.018 0.86 36.8 0.0009 7.2 −0.008 0.50 10.1 0.0098 8.6 −0.011 0.58 25.1 <0.0001 2010 10.1 −0.017 0.82 27.5 0.0019 10.3 −0.019 0.88 72.7 <0.0001 9.8 −0.017 0.83 87.3 <0.0001 All years 10.9 −0.017 0.64 36.7 <0.0001 7.0 −0.008 0.25 10.7 0.0026 8.3 −0.011 0.39 35.6 <0.0001 WUEt 2008 −0.58 0.012 0.93 66.4 0.0005 0.76 0.009 0.53 9.1 0.0167 0.35 0.010 0.68 31.6 <0.0001 2009 −1.79 0.019 0.93 76.7 0.0001 1.00 0.012 0.88 70.1 <0.0001 −0.12 0.015 0.88 127.2 <0.0001 2010 −3.22 0.028 0.93 74.2 0.0001 −5.15 0.033 0.95 200.6 <0.0001 −4.21 0.030 0.94 285.0 <0.0001 All years −1.84 0.020 0.48 19.5 0.0002 −0.62 0.017 0.43 24.6 <0.0001 −1.06 0.018 0.47 49.4 <0.0001 Ψl (MPa) 2009 −1.90 0.002 0.74 17.5 0.0058 −1.89 0.002 0.67 20.5 0.0011 −1.96 0.002 0.73 49.8 <0.0001 2010 −1.77 0.001 0.47 5.4 0.0593 −1.66 0.001 0.25 3.4 0.0941 −1.75 0.001 0.42 13.3 0.0018 All years −1.87 0.002 0.43 10.7 0.0056 −1.79 0.001 0.36 12.6 0.0018 −1.89 0.002 0.46 32.1 <0.0001
 Silty clay Clay Both soils Year Intercept Slope R2 F P Intercept Slope R2 F P Intercept Slope R2 F P ANPP (g m-2) 2007 −67.3 0.469 0.45 4.9 0.0681 −263.6 1.276 0.45 8.3 0.0165 −81.8 0.646 0.21 4.9 0.0396 2008 1670.2 −1.506 0.24 1.9 0.2133 860.3 1.599 0.26 3.1 0.1092 1511.5 −0.634 0.03 0.5 0.4702 2009 1792.4 1.028 0.33 3.0 0.1349 2037.4 1.045 0.11 1.2 0.3006 2143.6 0.499 0.03 0.6 0.4432 2010 2176.6 −0.098 0.00 0.0 0.9394 2481.8 −0.488 0.03 0.3 0.6195 2490.0 −0.666 0.05 0.9 0.3621 All years 1334.1 0.120 0.00 0.0 0.9525 1009.9 1.614 0.01 0.7 0.4167 1364.1 0.366 0.00 0.1 0.7819 # tillers 2007 2.1 0.172 0.38 3.7 0.1034 12.8 0.241 0.25 3.3 0.1007 39.4 0.128 0.10 2.0 0.1724 2008 307.2 −0.184 0.15 1.0 0.3472 369.0 −0.208 0.12 1.4 0.2600 384.0 −0.303 0.23 5.4 0.0327 2009 332.1 0.097 0.02 0.1 0.7187 639.3 −0.608 0.44 7.8 0.0188 532.8 −0.345 0.18 4.0 0.0607 2010 378.1 −0.001 0.00 0.0 0.9923 585.8 −0.454 0.21 2.6 0.1360 525.2 −0.318 0.17 3.7 0.0701 All years 246.3 0.042 0.00 0.0 0.8874 361.5 −0.146 0.00 0.2 0.6379 347.0 −0.149 0.01 0.5 0.4646 Tiller mass (g) 2007 0.75 0.002 0.31 2.7 0.1539 −1.06 0.008 0.49 9.6 0.1130 0.07 0.004 0.29 7.4 0.0143 2008 5.49 −0.002 0.16 1.1 0.3310 1.68 0.009 0.55 12.2 0.0059 3.80 0.003 0.70 1.4 0.2602 2009 5.45 0.001 0.03 0.2 0.6924 0.17 0.012 0.79 38.3 0.0001 3.49 0.006 0.37 10.5 0.0045 2010 5.75 0.000 0.00 0.0 0.9544 3.88 0.005 0.16 1.9 0.1940 4.68 0.002 0.06 1.3 0.2789 All years 4.22 0.001 0.00 0.0 0.8838 1.10 0.010 0.13 6.6 0.0133 2.74 0.005 0.04 3.1 0.0807 Clipped mass (g m-2) 2008 214.0 0.024 0.04 0.3 0.6143 −332.4 1.958 0.69 22.2 0.0008 −7.2 0.922 0.27 6.6 0.0195 2009 170.1 0.948 0.72 15.4 0.0078 −158.6 1.873 0.64 17.8 0.0018 11.9 1.368 0.60 26.5 <0.0001 2010 272.4 0.195 0.01 0.1 0.7784 455.5 −0.194 0.02 0.2 0.7005 405.7 −0.089 0.00 0.1 0.8070 All years 198.5 0.512 0.07 1.8 0.1991 −63.9 1.356 0.33 16.4 0.0003 104.0 0.822 0.16 11.4 0.0013 End of year mass (g m-2) 2008 1456.0 −1.745 0.49 5.7 0.0542 1192.7 −0.359 0.03 0.3 0.6165 1518.7 −1.555 0.28 7.0 0.0163 2009 1622.4 0.079 0.00 0.0 0.8880 2196.0 −0.828 0.09 1.1 0.3307 2131.7 −0.869 0.12 2.4 0.1387 2010 1904.2 −0.293 0.03 0.2 0.6639 2026.3 −0.294 0.02 0.2 0.6637 2084.3 −0.578 0.08 1.6 0.2163 All years 1193.4 −0.285 0.00 0.03 0.8579 1089.2 0.504 0.00 0.1 0.7514 1303.6 −0.299 0.00 0.1 0.7763 Clipped: end of year 2008 −0.05 0.001 0.57 8.0 0.0303 −0.40 0.002 0.77 33.6 0.0002 −0.25 0.002 0.69 40.5 <0.0001 2009 0.11 0.001 0.64 10.7 0.0172 −0.15 0.001 0.71 24.1 0.0006 −0.05 0.001 0.66 35.4 <0.0001 2010 0.15 0.000 0.02 0.1 0.7522 0.22 0.000 0.00 0.0 0.8705 0.19 0.000 0.00 0.0 0.9510 All years 0.04 0.001 0.16 4.31 0.0497 −0.14 0.001 0.36 19.03 0.0001 −0.06 0.001 0.29 23.6 <0.0001 ACO2 (μmol m-2 s-1) 2008 17.5 0.006 0.03 0.1 0.7292 3.8 0.043 0.47 7.1 0.0283 10.0 0.025 0.25 4.9 0.0425 2009 28.8 −0.011 0.06 0.4 0.5488 19.8 0.008 0.06 0.6 0.4609 22.4 0.003 0.00 0.1 0.7834 2010 28.2 −0.012 0.47 5.4 0.0592 17.6 0.012 0.13 1.5 0.2487 21.0 0.004 0.02 0.4 0.5507 All years 24.1 −0.004 0.01 0.2 0.6901 13.7 0.022 0.18 6.9 0.0132 17.7 0.011 0.06 3.4 0.0691 gS (mol m-2 s-1) 2008 0.42 −0.001 0.67 10.0 0.0251 0.12 0.000 0.05 0.4 0.5245 0.24 0.000 0.09 1.5 0.2479 2009 0.50 −0.001 0.86 37.5 0.0009 0.29 0.000 0.46 8.4 0.0158 0.36 0.000 0.55 22.5 0.0002 2010 0.73 −0.001 0.80 23.9 0.0027 0.77 −0.002 0.79 37.1 0.0001 0.72 −0.001 0.77 61.9 <0.0001 All years 0.54 −0.001 0.70 49.5 <0.0001 0.38 −0.001 0.22 9.0 0.0053 0.43 −0.001 0.39 34.5 <0.0001 E (mmol m-2 s-1) 2008 11.0 −0.015 0.68 10.5 0.0230 4.3 0.001 0.01 0.1 0.7703 6.9 −0.005 0.15 2.6 0.1291 2009 11.6 −0.018 0.86 36.8 0.0009 7.2 −0.008 0.50 10.1 0.0098 8.6 −0.011 0.58 25.1 <0.0001 2010 10.1 −0.017 0.82 27.5 0.0019 10.3 −0.019 0.88 72.7 <0.0001 9.8 −0.017 0.83 87.3 <0.0001 All years 10.9 −0.017 0.64 36.7 <0.0001 7.0 −0.008 0.25 10.7 0.0026 8.3 −0.011 0.39 35.6 <0.0001 WUEt 2008 −0.58 0.012 0.93 66.4 0.0005 0.76 0.009 0.53 9.1 0.0167 0.35 0.010 0.68 31.6 <0.0001 2009 −1.79 0.019 0.93 76.7 0.0001 1.00 0.012 0.88 70.1 <0.0001 −0.12 0.015 0.88 127.2 <0.0001 2010 −3.22 0.028 0.93 74.2 0.0001 −5.15 0.033 0.95 200.6 <0.0001 −4.21 0.030 0.94 285.0 <0.0001 All years −1.84 0.020 0.48 19.5 0.0002 −0.62 0.017 0.43 24.6 <0.0001 −1.06 0.018 0.47 49.4 <0.0001 Ψl (MPa) 2009 −1.90 0.002 0.74 17.5 0.0058 −1.89 0.002 0.67 20.5 0.0011 −1.96 0.002 0.73 49.8 <0.0001 2010 −1.77 0.001 0.47 5.4 0.0593 −1.66 0.001 0.25 3.4 0.0941 −1.75 0.001 0.42 13.3 0.0018 All years −1.87 0.002 0.43 10.7 0.0056 −1.79 0.001 0.36 12.6 0.0018 −1.89 0.002 0.46 32.1 <0.0001

### Leaf carbon and water fluxes

ANPP and tiller responses to CO2 were accompanied by changes in several measures of photosynthetic carbon uptake and water loss. gs and E were 14% lower on the clay soil compared to the silty clay, averaged across CO2 and year (P < 0.01, Table 1). However ACO2, WUEt, Ψl and leaf C:N did not differ between the soils (P > 0.09). ACO2 and gs increased from 2008 to 2010 while E decreased, leading to a 73% increase in WUEt, averaged over soils and CO2 (P < 0.0001, Fig. 4e). Leaf %N decreased from 2.2 ± 0.4% in 2009 to 1.5 ± 0.4% in 2010, resulting in a 50% increase leaf C:N (P ≤ 0.0001).

ACO2 showed a weak increase with CO2 enrichment in 2008 (R2 = 0.25, P = 0.04, Fig. 4a) but not in 2009–2010. In contrast, gS and E decreased ∼2-fold with CO2 enrichment, leading to a ∼2-fold increase in WUEt (Fig. 4a–d). Ψl also increased ∼40% with CO2 enrichment (Fig. 4e). Furthermore, as P. virgatum established, Ψl decreased (Fig. 4e), indicating increased plant water stress. gs and E were unresponsive to CO2 in 2008 (P = 0.13–0.25, Fig. 4b and c), but they declined more steeply and were increasingly tightly correlated with CO2 from 2008 to 2010 (Fig. 4b and c, Table 2). As a result, WUEt became increasingly responsive to CO2 enrichment from 2008 to 2010, with R2 reaching 0.94 in 2010.

Figure 4:

Physiological performance of Panicum virgatum along the CO2 gradient on clay (open symbols) and silty clay (closed symbols) soils from the second year of growth (2008) to full establishment (2010, symbols as in Fig. 3). (a) Leaf net photosynthesis (ACO2), (b) stomatal conductance (gS) (c), transpiration (E), (d) photosynthetic WUEt and (e) leaf water potential (Ψl). Regressions are shown for years when significant relationships (P < 0.05) were found for both soils combined (Table 2).

Figure 4:

Physiological performance of Panicum virgatum along the CO2 gradient on clay (open symbols) and silty clay (closed symbols) soils from the second year of growth (2008) to full establishment (2010, symbols as in Fig. 3). (a) Leaf net photosynthesis (ACO2), (b) stomatal conductance (gS) (c), transpiration (E), (d) photosynthetic WUEt and (e) leaf water potential (Ψl). Regressions are shown for years when significant relationships (P < 0.05) were found for both soils combined (Table 2).

These changes in photosynthetic carbon and water fluxes through time as P. virgatum established were accompanied by significant but less dramatic differences between soils in the CO2 responsiveness of a subset of the parameters. ACO2 increased with CO2 enrichment on the clay soil (R2 = 0.18, P = 0.01) but not on the silty clay (P = 0.69, Fig. 4a). In contrast, gS and E decreased with CO2 enrichment more steeply and were more strongly correlated with CO2 on the silty clay soil (R2 = 0.64–0.70) than on the clay soil (R2 = 0.22–0.25; Table 2). There were no differences between soils overall in the correlations of WUEt and Ψl with CO2.

## DISCUSSION

The findings of this study support the hypothesis that ANPP of P. virgatum was mostly unaffected by a 250 to 500 μl l-1 gradient of CO2 concentration. ANPP of established P. virgatum in this study was higher than that reported for field-grown plants in this region (Kiniry et al. 1996), suggesting that water and nutrient limitation were relatively low compared to field conditions. However, the findings also show that this range of CO2 concentrations caused offsetting effects on several components that contribute to ANPP and that the physiological coupling of P. virgatum to CO2 concentration increased as establishment proceeded.

### CO2 altered aboveground biomass allocation toward fewer, larger tillers

The finding of a trade-off between tiller mass and numbers suggests a reallocation of resources with CO2 enrichment, increasingly toward current tiller growth at the expense of tiller production in years 2 and 3. This represents a potential increase in meristem limitation of ANPP for establishing P. virgatum at elevated CO2 (Benson et al. 2004; Dalgleish and Hartnett 2009). There are few studies of tiller biomass responses to CO2 enrichment in P. virgatum. The finding contrasts with a meta-analysis of C3 and C4 grasses that found increased tiller mass and density with CO2 enrichment (Wand et al. 1999). CO2 enrichment would be expected to increases resource availability for tiller production and growth. Previous meta-analyses found that CO2 enrichment had little effect on plant biomass allocation, even when increasing total plant biomass (Poorter and Nagel 2000; Poorter et al. 2012; Wand et al. 1999).

The clipping applied to the experimental plants is likely the primary explanation for the unexpected pattern in tiller production. Clipping removed 20–40% of total biomass, including some fraction of leaf area, which would lower overall assimilation capacity and contribute to the apparent resource limitation. The biomass removed by clipping increased with CO2 enrichment during years 2 and 3; however final harvest biomass did not vary with CO2. This means the early to mid-season growth of P. virgatum increased with CO2 enrichment but not the later-season growth. Plant regrowth following defoliation typically declines through the growing season and increases with light, water and nutrient availability (Whitham et al. 1991). Mechanisms by which tiller regrowth after clipping took increasing precedence with CO2 enrichment over allocation to additional tillers may include (i) a strong tendency for plants to replacing the removed aboveground biomass, returning to pre-clipping allocation patterns (Poorter et al. 2012), (ii) the activation of new plant sinks for regrowth, which allows for stronger growth responses with CO2 enrichment (Kirschbaum 2011; Körner 2011), (iii) reduced activation and growth of rhizome buds following leaf removal (Beaty et al. 1978) and (iv) developmental constraints on growth patterns, such as continued stem growth to support later inflorescence production (Körner 2011).

The CO2 effects on early/midseason growth and tiller masses/numbers were absent in year 4 when the plants were fully established. This suggests that as leaf area continued to increase, increased whole-plant photosynthate supply may have alleviated resource limitations that likely caused the allocation trade-off between tiller numbers and size. In addition, other resources, such as light, soil moisture or N may have become more limiting than in preceding years. The decline in leaf N from 2009 to 2010 may be an early indication of N limitation of photosynthesis and/or biomass production, as suggested by previous studies in C4 grassland (Schimel et al. 1991; Turner and Knapp 1996). Nutrient effects on allocation are likely stronger than CO2 effects (Poorter and Nagel 2000).

### Physiological coupling to CO2 increased as P. virgatum established

The pattern of decreasing CO2 effects on the components of ANPP were accompanied by increased CO2 effects on stomatal conductance and transpiration as the plants established and leaf area increased. Overall photosynthetic rates of P. virgatum, the absence of increased photosynthesis and lower stomatal conductance and transpiration at elevated CO2 found in this study agree with previous findings for this and other C4 tallgrasses (Dohleman et al. 2009; Knapp 1985; Skeel and Gibson 1996). The resulting reduced plant water loss can increase soil moisture compared to that at lower CO2, providing an indirect mechanism by which CO2 enrichment may stimulate ANPP in water-limited, C4-dominated vegetation (Morgan et al. 2011). Indeed, we found a modest increase in soil moisture with CO2 enrichment. A larger increase in soil moisture would likely have occurred if the plants had been less well watered.

The increased soil moisture and decreased stomatal conductance with CO2 enrichment was accompanied by increased leaf water potential, indicating improved plant water status at elevated CO2. However, leaf water potential was generally lower and responded less to CO2, in year 4 compared to year 3. This suggests that the continued gain in leaf area during establishment increased the transpirational demand for soil moisture. This is the likely mechanism explaining why stomatal conductance and transpiration were progressively more strongly reduced by CO2 enrichment as plants established, increasing the coupling of WUE to CO2. We found little evidence to suggest photosynthetic down-regulation with CO2 enrichment as the plants established. Similar rates of ACO2 were observed each year at high CO2.

### Soil type effects on CO2 responses were inconsistent

Soil type affected several measures of P. virgatum growth and productivity. ANPP, tiller production and LAI averaged 15% higher on the clay than on the silty clay soil. ANPP and root mass of prairie assemblages studied on these same soils were also higher on the clay soil (Fay et al. 2009), suggesting that it was inherently more productive than the silty clay soil. Stout (1992) also reported differences in P. virgatum biomass among soil types, in part because of differences in plant water use. However, Sanderson et al. (1999) suggested that soil type had little influence on P. virgatum biomass production across several sites in Texas, suggesting that differences in climate can outweigh effects of soil type. Soil texture is a major reason for soil type differences in P. virgatum growth (Parrish and Fike 2005). Fine-textured clay soils have higher water holding capacity than more coarse textured soils, resulting in higher soil water contents (Fig. 1; Fay et al. 2009). Soil type effects on ANPP in this study suggest that water limitation was not fully eliminated despite the ET-replacement watering regime.

Differences between soils in the CO2 responses of some growth and physiology parameters were suggested by the analyses (Table 2). However, these differences were not apparent in the data (Figs 3 and 4), where overall CO2 effects predominated in years where they occurred. Also, the CO2 responses were sometimes inconsistent. For example, on the clay soil, stronger tiller mass and clipped mass increases with CO2 enrichment (compared to the silty clay) were accompanied by weaker stomatal conductance responses (Table 2). Moreover, these soil-specific responses did not culminate in soil-specific ANPP-CO2 responses. However, at lower soil moisture levels, the lower water holding capacity in the more coarse-textured silty clay soil will likely increase the soil moisture benefits (compared to the clay soil) of reduced transpiration with CO2 enrichment (Epstein et al. 1997, Morgan et al. 2011).

## CONCLUSIONS

The results from this study suggest that ANPP in clipped, well-watered, fertilized P. virgatum differed between these soils because of differences in soil moisture. ANPP of established P. virgatum did not vary with CO2 enrichment. CO2 enrichment affected canopy architecture and tiller growth while plants were establishing, likely because of clipping induced reallocation of resources that varied with CO2. Soil type effects on ANPP-CO2 responses will likely become more apparent when water is more limiting. These findings could have important implications for natural P. virgatum populations and for bioenergy production systems based on P. virgatum.

## FUNDING

United States Department of Agriculture-Agricultural Research Service and the National Science Foundation Plant Genome Research Program (IOS-0922457).

We thank A. Gibson, K. Jones, C. Kolodziejczyk, A. Naranjo and K. Tiner for technical support, and J. Kiniry for providing the P. virgatum rhizomes. USDA is an equal opportunity provider and employer. Mention of trade names or commercial products does not imply recommendation or endorsement by the USDA.

## References

Ainsworth
EA
Long
SP
What have we learned from 15 years of free-air CO2 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2
New Phytol
,
2005
, vol.
165
(pg.
351
-
72
)
Anderson
LJ
Maherali
H
Johnson
HB
, et al.  .
Gas exchange and photosynthetic acclimation over subambient to elevated CO2 in a C-3-C-4 grassland
Glob Change Biol
,
2001
, vol.
7
(pg.
693
-
707
)
Austin
AT
Yahdjian
L
Stark
JM
, et al.  .
Water pulses and biogeochemical cycles in arid and semiarid ecosystems
Oecologia
,
2004
, vol.
141
(pg.
221
-
35
)
Beaty
ER
Engel
JL
Powell
JD
Tiller development and growth in switchgrass
J Range Manage
,
1978
, vol.
31
(pg.
361
-
65
)
Benson
EJ
Hartnett
DC
Mann
KH
Belowground bud banks and meristem limitation in tallgrass prairie plant populations
Am J Bot
,
2004
, vol.
91
(pg.
416
-
21
)
Dalgleish
HJ
Hartnett
DC
The effects of fire frequency and grazing on tallgrass prairie productivity and plant composition are mediated through bud bank demography
Plant Ecol
,
2009
, vol.
201
(pg.
411
-
20
)
Dijkstra
FA
Blumenthal
D
Morgan
JA
, et al.  .
Elevated CO2 effects on semi-arid grassland plants in relation to water availability and competition
Funct Ecol
,
2010
, vol.
24
(pg.
1152
-
61
)
Dijkstra
FA
Pendall
E
Mosier
AR
, et al.  .
Long-term enhancement of N availability and plant growth under elevated CO2 in a semi-arid grassland
Funct Ecol
,
2008
, vol.
22
(pg.
975
-
82
)
Dohleman
FG
Heaton
EA
Leakey
, et al.  .
Does greater leaf-level photosynthesis explain the larger solar energy conversion efficiency of Miscanthus relative to switchgrass?
Plant Cell Environ
,
2009
, vol.
32
(pg.
1525
-
537
)
Drake
BG
González-Meler
MA
, et al.  .
More efficient plants: a consequence of rising atmospheric CO2?
Ann Rev Plant Physiol Plant Mol Biol
,
1997
, vol.
48
(pg.
609
-
39
)
Epstein
HE
Lauenroth
WK
Burke
IC
Effects of temperature and soil texture on ANPP in the U.S. Great Plains
Ecology
,
1997
, vol.
78
(pg.
2628
-
631
)
Fay
PA
Kelley
AM
Procter
AC
, et al.  .
Primary productivity and water balance of grassland vegetation on three soils in a continuous CO2 gradient: initial results from the lysimeter CO2 gradient experiment
Ecosystems
,
2009
, vol.
5
(pg.
4
-
714
)
Forster
P
Ramaswamy
V
Artaxo
P
, et al.  .
Solomon
S
Qin
D
Manning
M
, et al.  .
Changes in atmospheric constituents and in radiative forcing
(eds). Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change
,
2007
Cambridge, United Kingdom and New York, NY
Cambridge University Press
(pg.
129
-
234
)
Gill
RA
Polley
HW
Johnson
HB
, et al.  .
Nonlinear grassland responses to past and future atmospheric CO2
Nature
,
2002
, vol.
417
(pg.
279
-
82
)
Kimball
BA
LaMorte
RL
Pinter
PJ
Jr
, et al.  .
Free-air CO2 enrichment and soil nitrogen effects on energy balance and evapotranspiration of wheat
Water Resour Res
,
1999
, vol.
35
(pg.
1179
-
90
)
Kiniry
JR
Sanderson
MA
Williams
JR
, et al.  .
Simulating Alamo switchgrass with the ALMANAC model
Agron J
,
1996
, vol.
88
(pg.
602
-
06
)
Kirschbaum
MUF
Does enhanced photosynthesis enhance growth? Lessons learned from CO(2) enrichment studies
Plant Physiol
,
2011
, vol.
155
(pg.
117
-
24
)
Knapp
AK
Effect of fire and drought on the ecophysiology of Andropogon gerardii and Panicum virgatum in a tallgrass prairie
Ecology
,
1985
, vol.
66
(pg.
1309
-
20
)
Knapp
AK
Hamerlynck
EP
Ham
JM
, et al.  .
Responses in stomatal conductance to elevated CO2 in twelve grassland species that differ in growth form
Vegetatio
,
1996
, vol.
125
(pg.
31
-
41
)
Körner
Ch
The grand challenges in functional plant ecology
Front Plant Sci
,
2011
, vol.
2

DOI=10.3389/fpls.2011.00001
Leakey
Rising atmospheric carbon dioxide concentration and the future of C 4 crops for food and fuel
P Roy Soc B-Biol Sci
,
2009
, vol.
276
(pg.
2333
-
43
)
Lecain
DR
Morgan
JA
Mosier
AR
, et al.  .
Soil and plant water relations determine photosynthetic responses of C3 and C4 grasses in a semi-arid ecosystem under elevated CO2
Ann Bot
,
2003
, vol.
92
(pg.
41
-
52
)
Luo
Y
Su
B
Currie
WS
, et al.  .
Progressive nitrogen limitation of ecosystem responses to rising atmospheric carbon dioxide
BioScience
,
2004
, vol.
54
(pg.
731
-
39
)
Morgan
JA
LeCain
DR
Mosier
AR
, et al.  .
Elevated CO2 enhances water relations and productivity and affects gas exchange in C3 and C4 grasses of the Colorado shortgrass steppe
Glob Change Biol
,
2001
, vol.
7
(pg.
451
-
466
)
Morgan
JA
Lecain
DR
Pendall
E
, et al.  .
C4 grasses prosper as carbon dioxide eliminates desiccation in warmed semi-arid grassland
Nature
,
2011
, vol.
476
(pg.
202
-
05
)
Norby
RJ
Warren
JM
Iversen
CM
, et al.  .
CO2 enhancement of forest productivity constrained by limited nitrogen availability
PNAS
,
2010
, vol.
107
(pg.
19368
-
73
)
Oliver
RJ
Finch
JW
Taylor
G
Second generation bioenergy crops and climate change: a review of the effects of elevated atmospheric CO2 and drought on water use and the implications for yield
GCB Bioenergy
,
2009
, vol.
1
(pg.
97
-
114
)
Owensby
CE
Ham
JM
Knapp
A
, et al.  .
Koch
GW
Mooney
HA
Ecosystem-level responses of tallgrass prairie to elevated CO2
Carbon Dioxide and Terrestrial Ecosystems
,
1996
San Diego, CA
(pg.
147
-
62
)
Parrish
DJ
Fike
JH
The biology and agronomy of switchgrass for biofuels
Crit Rev Plant Sci
,
2005
, vol.
24
(pg.
423
-
59
)
Perlack
RD
Wright
LL
Turhollow
AF
, et al.  .
Biomass as Feedstock for a Bioenergy and Bioproducts Industry: The Technical Feasibility of a Billion-Ton Annual Supply
,
2005
Oak Ridge, TN
USDA and DOE, Oak Ridge National Laboratory
(pg.
1
-
60
)
Polley
HW
Jin
VL
Fay
PA
CO2-caused change in plant species composition rivals the shift in vegetation between mid-grass and tallgrass prairies
Glob Change Biol
,
2011
, vol.
18
(pg.
700
-
10
)
Polley
HW
Johnson
HB
Fay
PA
, et al.  .
Initial response of evapotranspiration from tallgrass prairie vegetation to CO2 at subambient to elevated concentrations
Funct Ecol
,
2008
, vol.
22
(pg.
163
-
71
)
Polley
HW
Morgan
JA
Fay
PA
Application of a conceptual framework to interpret variability in rangeland responses to atmospheric CO2 enrichment
J Agr Sci
,
2011
, vol.
149
(pg.
1
-
14
)
Poorter
H
Nagel
O
The role of biomass allocation in the growth response of plants to different levels of light, CO(2), nutrients and water: a quantitative review
Aust J Plant Physiol
,
2000
, vol.
27
(pg.
595
-
607
)
Poorter
H
Niklas
KJ
Reich
PB
, et al.  .
Biomass allocation to leaves, stems and roots: meta-analyses of interspecific variation and environmental control
New Phytol
,
2012
, vol.
193
(pg.
30
-
50
)
Sanderson
MA
Reed
RL
Ocumpaugh
WR
, et al.  .
Switchgrass cultivars and germplasm for biomass feedstock production in Texas
Bioresource Technol
,
1999
, vol.
67
(pg.
209
-
19
)
Schimel
DS
Kittel
TGF
Knapp
AK
, et al.  .
Physiological interactions along resource gradients in a tallgrass prairie
Ecology
,
1991
, vol.
72
(pg.
672
-
84
)
Skeel
VA
Gibson
DJ
Physiological performance of Andropogon gerardii, Panicum virgatum, and Sorghastrum nutans on reclaimed mine spoil
Restor Ecol
,
1996
, vol.
4
(pg.
355
-
67
)
Smith
MD
Knapp
AK
Collins
SL
A framework for assessing ecosystem dynamics in response to chronic resource alterations induced by global change
Ecology
,
2009
, vol.
90
(pg.
3279
-
89
)
Stout
WL
Water-use efficiency of grasses as affected by soil, nitrogen, and temperature
Soil Sci Soc Am J
,
1992
, vol.
56
(pg.
897
-
902
)
Turner
CL
Knapp
AK
Responses of a C-4 grass and three C-3 forbs to variation in nitrogen and light in tallgrass prairie
Ecology
,
1996
, vol.
77
(pg.
1738
-
49
)
Wand
SJE
Midgley
G
Jones
MH
, et al.  .
Responses of wild C4 and C3 grass (Poaceae) species to elevated atmospheric CO2 concentration: a meta-analytic test of current theories and perceptions
Glob Change Biol
,
1999
, vol.
5
(pg.
723
-
41
)
Whitham
TG
Maschinski
J
Larson
KC
, et al.  .
Price
PW
Lewinsohn
TM
Fernandes
GW
, et al.  .
Plant responses to herbivory: the continuum from negative to positive and underlying physiological mechanisms
(eds). Plant-Animal Interactions: Evolutionary Ecology in Tropical and Temperate Regions
,
1991
New York
Wiley
(pg.
227
-
56
)