More intense, more frequent, and longer heat-waves are expected in the future due to global warming, which could have dramatic ecological impacts. Increasing nitrogen (N) availability and its dynamics will likely impact plant responses to heat stress and carbon (C) sequestration in terrestrial ecosystems. This field study examined the effects of N availability on plant response to heat-stress (HS) treatment in naturally-occurring vegetation. HS (5 d at ambient or 40.5 °C) and N treatments (±N) were applied to 16 1 m2 plots in restored prairie vegetation dominated by Andropogon gerardii (warm-season C4 grass) and Solidago canadensis (warm-season C3 forb). Before, during, and after HS, air, canopy, and soil temperature were monitored; net CO2 assimilation (Pn), quantum yield of photosystem II (ΦPSII), stomatal conductance (gs), and leaf water potential (Ψw) of the dominant species and soil respiration (Rsoil) of each plot were measured daily during HS. One week after HS, plots were harvested, and C% and N% were determined for rhizosphere and bulk soil, and above-ground tissue (green/senescent leaf, stem, and flower). Photosynthetic N-use efficiency (PNUE) and N resorption rate (NRR) were calculated. HS decreased Pn, gs, Ψw, and PNUE for both species, and +N treatment generally increased these variables (±HS), but often slowed their post-HS recovery. Aboveground biomass tended to decrease with HS in both species (and for green leaf mass in S. canadensis), but decrease with +N for A. gerardii and increase with +N for S. canadensis. For A. gerardii, HS tended to decrease N% in green tissues with +N, whereas in S. canadensis, HS increased N% in green leaves. Added N decreased NRR for A. gerardii and HS increased NRR for S. canadensis. These results suggest that heat waves, though transient, could have significant effects on plants, communities, and ecosystem N cycling, and N can influence the effect of heat waves.
Global mean surface temperatures have risen by 0.6 °C from 1900 to 2000, mainly caused by increases in atmospheric CO2 and other greenhouse gases, and are projected to increase by another 1.4–5.8 °C by year 2100 (Houghton et al. 2001; IPCC 2007). In addition to rising mean annual temperatures, there will also be increases in the frequency, duration, and severity of periods with exceptionally high temperatures (Wagner 1996). An increased trend in the frequency of extreme heat stress events has been reported in various parts of the world (Henderson and Muller 1997; Gaffen and Ross 1998; Gruza and Ran'kova 1999; Yan 2002). Thus, plants in the future will not only be exposed to higher mean temperatures, but will also likely experience more frequent heat stress, which can greatly impact ecosystem productivity (Ciais et al. 2005) and biodiversity (Thomas et al. 2004). An extreme stress event is an episode in which the acclimatory capacities of an organism are substantially exceeded (Gutschick and BassiriRad 2003). Extreme events, in spite of their ephemeral nature, can cause shifts in the structure of plant communities. The environmental impacts from extreme events can be significantly greater than those associated with mean increases (Karl et al. 1997).
In addition to temperatures, human activities are increasing global N availability (IPCC 2007). N availability is likely to affect plant, community, and ecosystem responses to increasing heat stress, which will then impact ecosystem C sequestration. Understanding the effects of N on the responses of vegetation to heat stress requires insight into how stress physiology and community structure interact. While the influence of plant N status on response to acute heat stress has been previously examined, past studies have largely focused on laboratory experiments examining physiological responses (Heckathorn et al. 1996a,1996b; Lu and Zhang 2000). Further, because of the difficulties of imposing heat stress on naturally-occurring vegetation, little experimental work has been conducted on responses to acute heat stress in field-grown plants (Weis and Berry 1987; Morison and Lawlor 1999). To date, there have been only a handful of studies in which plant communities were exposed to extreme high temperatures, and these focused on community processes (e.g., recolonization, competition, invasion, and the role of species richness during extreme events) and were conducted on grassland (White et al. 2001; Van Peer et al. 2004) or arctic species (Marchand et al. 2005, 2006). Also, N availability had significant effects on plant N-relations responses to moderate warming (rather than acute heat stress) in a tallgrass prairie (An et al. 2005). Thus, little is known as to how heat stress in general, and N interactions with heat stress in particular, will affect natural plant communities. In this study, we concentrate on physiological and growth responses of two dominant warm-season tall-grass prairie species with contrasting photosynthetic pathways (a C4 grass and a C3 forb) in experimental field plots receiving heat and N treatments.
C4 species typically have higher temperature optima for photosynthesis than C3 species (Sage and Monson 1999) as a consequence of lower photorespiration, which increases with temperature. This may contribute to greater tolerance to heat waves for C4 species than co-occurring C3 species (Coleman and Bazzaz 1992; Ehleringer et al. 1997; Wang et al. 2008). Thus, heat stress can potentially affect the relative distribution of C4 and C3 species. In natural systems, the significance of climate warming for C4 vegetation can depend less on the mean increase in global temperature, and more on the spatial and temporal variation of the temperature increase (Sage and Kubien 2003). In New Zealand, for example, episodic heat events inhibit C3 plants more than C4 grasses, and as a result, facilitate C4 grass invasion of C3-dominated grasslands (White et al. 2000, 2001). On the other hand, because of their greater nitrogen (N) investment in rubisco and photorespiratory enzymes, C3 plants have lower N-use efficiency of photosynthesis (PNUE) than C4 plants (Sage and Pearcy 1987; Li 1993). The ecological consequences of greater PNUE in C4 species have been studied to only a limited degree. For example, in grasslands, when soils are low in N, C4 grasses can be superior competitors to C3 grasses and can dominate (Wedin and Tilman 1996). When soils become N-enriched, the advantage in PNUE is offset and C3 species can match the photosynthetic potential of C4 species, and thereby increase in cover. However, whether N availability interacts with heat stress differently in C3 versus C4 species remains to be determined, but will have a bearing on the relative impact of global environmental change on C3 and C4 species abundance and distribution.
To examine the influence of N on plant response to heat stress in naturally-occurring mixed C3-C4 vegetation, we conducted a field study with the following three major objectives: (1) to determine how heat stress affects the ecophysiological and morphological variables of naturally-occurring co-dominant C4 and C3 species; (ii) to determine the effect of N on resistance and resilience of each species to heat stress; and (iii) to investigate how heat stress affects plant C and N concentration, N-use efficiency, and N resorption rate. We predicted that: (i) heat stress will have a more pronounced negative effect on the C3 than the C4 species; (ii) supplemental N will help both the C3 and C4 species to better tolerate heat stress, especially for the C3 species; and (iii) heat stress will increase leaf N concentration and decrease N-use efficiency, as a result of decreased leaf expansion and photosynthesis, but more so for the C3 species.
During heat stress (HS), air temperature in the heated plots was increased on average to 40.5 ± 2.8 °C. During the 5 d of heat treatment, leaf temperature of A. gerardii and S. canadensis in heated plots was higher than that in control plots, but returned to control levels after heat stress (Figure 1, Table 1). Nitrogen treatment had no effect on leaf temperature for A. gerardii, but for S. canadensis, plants with N treatment had a lower leaf temperature. Soil temperature (at 10 cm) was not altered by heat or N treatment (not shown). C% of both rhizosphere and bulk soil was not changed by heat stress, and N% of both rhizosphere and bulk soil was not affected by heat stress, but was increased by N treatment (P < 0.01) (not shown).
Table 1. Degrees of freedom (numerator and denominator d.f.) and F-statistics from ANOVA on individual plant variables in response to heat and nitrogen treatment in a restored prairie in Toledo, OH, with days (d 1 to d 5), heat, and nitrogen and their interactions as independent factors
F-statistics with a single asterisk indicate significance at P < 0.10, whereas a double asterisk indicates P < 0.05. gs, stomatal conductance; PNUE, photosynthetic N-use efficiency; Pn, CO2 assimilation;ΦPSII, quantum yield of photosystem II; Tleaf, leaf temperature; Ψw, leaf water potential.
Leaf water potential (Ψw) was decreased for heated plants, but N had little effect on Ψw; and Ψw was recovered 1 week after heat-stress (d 12) and was similar among treatments (Figure 2, Table 1). Soil (=soil + root) respiration (Rsoil) was decreased by both heat and N, but there was no interactive effect of N and heat, and after heat-stress Rsoil was similar for ±HS in +N plants but still lower in +HS versus −HS plants with no added N (Figure 3, Table 1). There was an overall negative effect of heat treatment on net photosynthesis (Pn). For both A. gerardii and S. canadensis, Pn was significantly lower in heated plots than in control plots during heat stress. N had no significant effect on Pn for A. gerardii, but for S. canadensis, N increased Pn and there was a significant heat × N interaction (Figure 4, Table 1). Further, Pn remained depressed 1 week after HS in +HS plants with added N, relative to un-heated controls, but this was not observed in plants receiving no added N. Variation in stomatal conductance to water vapor (gs) was a function of both heat and N. For A. gerardii and S. canadensis, gs was lower in heated plots and higher in N-treated plots. There was also a significant interactive effect of heat and N on gs for S. canadensis (Figure 4, Table 1). Also, gs remained depressed 1 week after HS in +HS plants (more so in +N), relative to un-heated controls, especially for S. canadensis. Quantum yield of electron transport (ΦPSII) was not decreased by heat stress for A. gerardii and S. canadensis, but N had a significant positive effect on ΦPSII for A. gerardii and S. canadensis (Figure 4, Table 1).
For A. gerardii, N treatment increased specific leaf area (SLA), but heat did not affect SLA, whereas for S. canadensis, neither N or heat affected SLA (Figure 5, Tables 2, 3). Total aboveground biomass (Wa) was not significantly affected by heat or N in either species, but there was a decrease in Wa with heat in both species (ANOVA including both species; P= 0.024), and increases in Wa with added N in S. canadensis and decreases in Wa with added N in A. gerardii (Figure 6, Tables 2, 3). Biomass of green leaf, stem, and senescent leaf was not significantly altered by either heat or N treatment for both A. gerardii and S. canadensis. However, flower biomass was increased significantly by N treatment for S. canadensis and decreased by heat stress for A. gerardii. For A. gerardii without N treatment, the percentage of senescent leaf was significantly higher in heated plots (18.2%) than at control plots (12.3%).
Table 2. Degrees of freedom (numerator and denominator d.f.) and F-statistics from ANOVA on individual plant variables for Andropogon gerardii, in response to heat and nitrogen treatment in a restored prairie in Toledo, OH, with heat, N and their interactions as independent variables
F-statistics with a single asterisk indicated significance at P < 0.10, whereas a double asterisk indicates P < 0.05. Cf, C% of flower; Cg, C% of green leaf; Cs, C% of senescent leaf; Cst, C% of stem; Nf, N% of flower; Ng, green leaf N concentration; Ns, senescent leaf N concentration; Nst, N% of stem; SLA, specific leaf area; Wa, total aboveground biomass; Wf, flower biomass; Wg, biomass of green leaf; Ws, biomass of senescent leaf; Wst, biomass of stem.
Table 3. Degrees of freedom (numerator and denominator d.f.) and F-statistics from ANOVA on individual plant variables for Solidago canadensis, in response to heat and nitrogen treatment in a restored prairie in Toledo, OH, with heat, N and their interactions as independent variables
F-statistics with a single asterisk indicated significance at P < 0.10, whereas a double asterisk indicates P < 0.05. Cf, C% of flower; Cg, C% of green leaf; Cs, C% of senescent leaf; Cst, C% of stem; Nf, N% of flower; Ng, green leaf N concentration; NRR, N resorption rate; Ns, senescent leaf N concentration; Nst, N% of stem; SLA, specific leaf area; Wa, total aboveground biomass; Wf, flower biomass; Wg, biomass of green leaf; Ws, biomass of senescent leaf; Wst, biomass of stem.
Carbon concentration of plant tissue (C%) was altered by both heat and N treatment, but the effects differed with species and among different plant parts (Figure 7, Tables 2, 3). For A. gerardii, C% of N-treated plants was lower in green leaves and flowers, but higher in non-heated senescent leaves. Heat decreased C% only in senescent leaves for A. gerardii with N treatment. C% was not altered by either heat or N in green leaf, senescent leaf, stem, and flower for S. canadensis. In general, nitrogen concentration (N%) was increased in N-treated plant tissues (excluding flowers) for both A. gerardii and S. canadensis (Figure 7, Tables 2, 3). Heat had little effect on N% in A. gerardii, though there was a tendency for decreases in N% in N-treated plants, while in S. canadensis, heat increased N% in green leaves, and for A. gerardii, there was also an interactive effect of heat and N on N% in green leaves.
Photosynthetic nitrogen-use efficiency (PNUE) was significantly lower for S. canadensis than A. gerardii and was decreased by heat for both A. gerardii and S. canadensis (Figure 8, Table 1). In heated plots, plants with N treatment tended to have a higher PNUE for A. gerardii, though the effect was not statistically significant. Further, recovery of PNUE was incomplete after one week post-HS, relative to un-heated controls, for both species and N levels. Nitrogen resorption rate (NRR) was decreased by N treatment, but not changed by heat, for A. gerardii, whereas for S. canadensis, NRR was significantly higher for heated plants, but was not different due to N treatment (Figure 9, Tables 2, 3).
During heat stress, both A. gerardii (C4) and S. canadensis (C3) experienced decreased Pn, gs, Ψw, PNUE, and soil (=soil + root) respiration decreased too; decreases in Pn, gs, PNUE, and soil respiration were still evident 1 week after heat treatment ended (d 12). In general, N addition affected these physiological variables in both heated and unheated-plants (increasing Pn, gs, Ψw, PNUE, but decreasing soil respiration). With few exceptions, during heat stress, N did not alter the nature of the heat-stress effect on these variables (i.e., there was no significant heat × N interaction). In contrast, 1 week after heat-stress recovery, residual heat-stress-related decreases in Pn and gs were evident only (Pn), or greater (gs), in high-N plants, and decreases in soil respiration were only evident in low-N plants. Both species exhibited trends of decreasing aboveground biomass with heat treatment, whereas added N tended to increase biomass in S. canadensis but decrease biomass in A. gerardii. Carbon concentration (C%) of tissues was affected only by heat treatment in A. gerardii (leaves and stems), whereas N% of green leaves in A. gerardii decreased with heat stress (+N only) but increased with heat stress in S. canadensis. Finally, heat treatment increased N resorption rate (NRR) in S. canadensis, but not in A. gerardii, whereas added N decreased NRR for A. gerardii, but not in S. canadensis.
Collectively, these results indicate the heat waves imposed here were of moderate severity (as evidenced by the magnitude of HS effects), yet such moderate heat waves can affect plant C and N relations and biomass growth and allocation, and the heat effects can still be evident after 1 week of post-heat recovery. Further, many plant responses to the heat treatment were influenced by N availability and differed between the C3 species, S. canadensis, and the C4 species, A. gerardii. Specifically, these results suggest that in a future warmer world with increasing N availability, S. canadensis may be affected less by heat waves than A. gerardii, but in the absence of more N, the reverse may be true. It is also worth noting that heat and N effects in this study may be smaller than likely to occur, as our heat treatment was a single heat wave (and plants in north-west Ohio experience approximately three to five heat waves per summer) and our N treatment was initiated only 2 weeks prior to heat stress. Thus, predictions of N effects on heat-stress responses and differences between C3 and C4 plants based on this study may be conservative.
Any N-related influence on heat-induced changes in plant physiology, growth, biomass allocation, and tissue C and N concentration, and any such differences in N × heat effects between C3 and C4 species, will have important implications for plant herbivory and decomposition, and thus, for ecosystem N and C dynamics. For example, heat-related increases in tissue senescence and changes in C or N% will have a direct impact on herbivore feeding preference and growth rate, and on litter quantity and quality, and hence on decomposition rates. A shift in the ratio of C3:C4 species with increasing heat waves in the presence/absence of higher N would have a dramatic impact on ecosystem N or C cycling also, as C4 foliage is characterized by a higher carbon-to-nitrogen (C:N) ratio and higher fiber content than C3 foliage, thus contributing high C:N litter to soil organic matter, which results in low N mineralization rates (Wedin and Tilman 1996; Sage and Monson 1999). High C:N ratios also reduce decomposition rates, such that proportionally more N on a site may reside in the soil organic matter pool (Aerts 1997); thus N availability often declines when a C4-dominated sward replaces C3 vegetation (Reich et al. 2001).
Plants appear to be more susceptible to high day or night-time temperatures during later flower-to-early seed developmental stages (Cross et al. 2003). Notably, in this study, flower biomass (Wf) was significantly reduced for A. gerardii by heat stress, and this decrease was somewhat smaller in high-N plants. Heat stress had no effect on Wf for S. canadensis, suggesting that increases in heat waves in the future may affect the seed bank of this plant community, which might affect community structure in the longer term. Heat-stressed plants can compensate for decreases in flower production by producing later flowers on existing inflorescences (Sato et al. 2000; Cross et al. 2003), but whether plants can still compensate for decreased flower production after a late-growing-season heat stress remains to be investigated.
The heat-treatment effects on plant C and N relations observed in this study differ somewhat compared with results from previous studies. For example, although we observed increased N% in green leaves with heat stress in S. canadensis, past studies applying long-term warming observed decreases in N% of green leaves with warming (Tjoelker et al. 1999; An et al. 2005). In contrast, other warming studies showed that elevated temperature increased leaf N concentration due to enhanced soil N mineralization and increased plant N uptake (Nijs et al. 1996; Luomala et al. 2003). We also observed unique effects on N resorption (NRR) from senescing leaves in this study, compared with previous studies. Here, NRR was decreased by N treatment but not changed by heat stress for A. gerardii; but for S. canadensis, NRR was increased by heat stress but was not altered by N treatment. The decreased NRR of A. gerardii due to N treatment in our study is consistent with the observation that species from nutrient-poor environments often have a higher N resorption rate than species from nutrient-rich environments (Aerts 1997). The partitioning of N compounds between soluble and structural compounds is an important regulator of N resorption (Norby et al. 2001; Yuan et al. 2005). The increased NRR of S. canadensis in the heated plots might be caused by acceleration of the normal senescence and resorption process. Warming has been observed to accelerate the senescence of leaves, such that warmed plants completed the normal senescence and resorption process faster than those in un-warmed controls, resulting in plant litter in the warmed plots either having a lower N concentration or lower fraction of N in soluble compounds (Norby et al. 2000).
The present experiment showed that heat stress, though ephemeral, can potentially modify community composition and impact ecosystem nutrient cycling, via effects on plant growth and tissue N and C content. Further, increases in N availability may influence plant response to heat stress; for example, as slowing recovery of heat-related damage to photosynthesis, and benefiting C3 species more than C4 species during heat stress. This study only examined short-term plant responses to acute heat stress within one generation of perennial plants, but the results indicate that the impact of acute heat stress on plant communities and ecosystems should be studied more extensively, particularly in combination with other potentially-interactive aspects of global environmental change (e.g., CO2, O3, and precipitation).
Materials and Methods
Field site and treatments
The experiment site was located within restored prairie vegetation at the University of Toledo's Stranahan Arboretum (Toledo, Ohio, USA), which is located within the oak-savannah glacial-sand ecosystem referred to as the “Oak Openings” region (http://oakopen.org/). Andropogon gerardii (big bluestem), a warm-season C4 perennial grass, and Solidago canadensis (goldenrod), a warm-season C3 perennial herbaceous dicot, are the two dominant plant species in this field site. The experiment design was a 2 × 2 factorial (±heating ×±added N; with n= 4 replicates per treatment combination), using 16 ×1 m2 randomly selected and assigned-treatment plots. Eight of the plots received heat treatment for 5 d from 17 to 21 August 2006, and eight of the plots received added N treatment (NH4NO3) applied twice (1 and 2 weeks) before heat treatment at a rate of 5 g N/m2 per year. Heat treatment was applied by using eight top-vented 1 m3-chambers made with transparent plastic attached to a wooden frame. A portable electric heater with a maximum capacity of 1 500 W was installed in the chamber to increase air temperature and an electric fan was used to circulate warm air inside the chamber. The target treatment temperature was 41 °C, which is 10 °C higher than the average daytime temperature for August in Toledo, and 2–3 °C higher than the typical maximum temperature in the summer season in this area. Heat treatments were imposed during daytime for 5 d and for 10-h per day (08.00 am to 18.00 hours). Control plots were not covered by chambers and experienced ambient temperature. The air temperature and leaf temperatures inside and outside the chambers were monitored using data-loggers and fine-wire thermocouples during heat treatments; soil temperature at 10 cm depth was measured with a thermometer.
Before, during, and after heat stress, net photosynthesis, stomatal conductance to water vapor, quantum yield of electron transport of photosystem II (PSII), and leaf water potential were measured daily on randomly-chosen recently-expanded fully-lit leaves. Steady-state net photosynthesis (Pn; net CO2 exchange) and stomatal conductance (gs) of single leaves was measured with a portable photosynthesis system containing an infrared gas analyzer (model 6400, LiCOR, Lincoln, NE, USA), equipped with a 250 mm3 leaf chamber as in Heckathorn et al. (1997). Measurements were made at ambient light and temperature within 1 min of insertion of leaves into the cuvette (immediately after stabilization of CO2 and H2O fluxes). Quantum yield of PSII electron transport (ΦPSII) was measured with a pulse-amplitude-modulated (PAM) fluorometer (Model PAM 101/103, Walze, Germany), as in Wang et al. (2008). A pressure chamber (Model 600, PMS Instruments Co., Corvallis, OR, USA) was used to measure midday leaf water potential (Ψw).
Biomass and C, N measurements
One-week after heat stress, 40 × 50 cm2 of each plot was harvested. The clipped plants were sorted into different categories (green and senescent leaves, stems and flowers), oven-dried at 65 °C for 1 week and weighed. N and C concentration for different plant parts, as well as rhizosphere and bulk soil, was measured with a Perkin Elmer CHN Analyzer (Model 2400, MA, USA). NRR for each species was calculated by NRR = (Ng–Ns)/Ng, where Ng was the green leaf N concentration and Ns was the senescent leaf N concentration. Photosynthetic N-use efficiency (PNUE) was based on net photosynthesis per unit plant N.
Analyses were conducted within each species to determine whether the physiological variables differed as a function of different treatments. For daily-measured variables like Ψw, Pn, gs, PNUE, ΦPSII, Rsoil, and leaf temperature, three-way ANOVA (SAS 9.1) was used to test for significant effects of days, heat, N, and their interaction during heat stress (d 1 to d 5). A two-way ANOVA was used to test for significant effects of heat, N, and their interaction on Pn, gs, PNUE, ΦPSII, Rsoil, and leaf temperature after heat stress (i.e., on d 12), and on biomass, C%, and N% of plants and soil. In order to test for species effect, three-way ANOVA was conducted on biomass, C% and N%, with species, heat, N and their interactions as independent factors. Days, heat, and N were all treated as fixed effects.
(Handling editor: Jiquan Chen)
We thank Daryl Moorhead, Sandra Stutzenstein, and Walter Schulisch for providing access to experimental field sites, as well as for logistical support and assistance conducting the experiment. We thank Jiquan Chen for providing us vehicles and experimental equipment. Thanks to Sasmita Mishra and Rajan Tripathee for assistance with plant harvest and processing plant materials. We are indebted to Rachel Henderson for her advice on carbon and nitrogen analysis. We thank the reviewers for their helpful comments on this manuscript.