Carbon dioxide enrichment alters plant community structure and accelerates shrub growth in the shortgrass steppe
Edited by Harold A. Mooney, Stanford University, Stanford, CA, and approved July 25, 2007
Abstract
A hypothesis has been advanced that the incursion of woody plants into world grasslands over the past two centuries has been driven in part by increasing carbon dioxide concentration, [CO2], in Earth's atmosphere. Unlike the warm season forage grasses they are displacing, woody plants have a photosynthetic metabolism and carbon allocation patterns that are responsive to CO2, and many have tap roots that are more effective than grasses for reaching deep soil water stores that can be enhanced under elevated CO2. However, this commonly cited hypothesis has little direct support from manipulative experimentation and competes with more traditional theories of shrub encroachment involving climate change, management, and fire. Here, we show that, although doubling [CO2] over the Colorado shortgrass steppe had little impact on plant species diversity, it resulted in an increasingly dissimilar plant community over the 5-year experiment compared with plots maintained at present-day [CO2]. Growth at the doubled [CO2] resulted in an ≈40-fold increase in aboveground biomass and a 20-fold increase in plant cover of Artemisia frigida Willd, a common subshrub of some North American and Asian grasslands. This CO2-induced enhancement of plant growth, among the highest yet reported, provides evidence from a native grassland suggesting that rising atmospheric [CO2] may be contributing to the shrubland expansions of the past 200 years. Encroachment of shrubs into grasslands is an important problem facing rangeland managers and ranchers; this process replaces grasses, the preferred forage of domestic livestock, with species that are unsuitable for domestic livestock grazing.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
Although atmospheric carbon dioxide concentrations ([CO2]) have increased from ≈280 volumetric ppm in preindustrial times to ≈380 ppm today and are projected to exceed 600 ppm by the end of this century, it is perhaps more important to point out that CO2 levels are higher today than they have been for at least 650,000 years (1). Furthermore, levels of atmospheric CO2 for the past half million years have tended to stay closer to the lowest glacial levels of ≈180 ppm compared with the ≈280–300 ppm of interglacial periods. These recent abrupt changes in atmospheric CO2 have tremendous implications for the adaptation and evolution of relatively modern ecosystems, such as C4 grasslands, that have evolved under relatively low atmospheric [CO2] by today's standards. This report focuses on the responses of vegetation in a Colorado semiarid grassland to growth at variable [CO2], but our report has implications for other rangelands around the world.
Rangelands comprise >40% of Earth's terrestrial surface (2). Although these lands are characteristically water-limited and unsuitable for intensive agriculture, they support one of the world's most extensive agricultural practices, domestic livestock grazing (3–5). Rangelands are important not only for the plant and animal products they provide but also as regions in which distinct pastoral cultures and societies have developed. Rising atmospheric [CO2] and predicted global change are expected to have an especially strong impact on water-limited regions like rangelands (6, 7), with potential consequences for long-established grazing practices. For instance, although grassland plant production often increases with rising atmospheric CO2, forage quality can decline because of lower N concentrations (8–11) or because of an increased abundance of lower quality plant species (11).
The alteration of plant community dynamics because of differential plant species or to functional group sensitivities to rising atmospheric CO2 is of particular concern for native ecosystems, including rangelands, in which the economic value of the land depends in large part on the species composition of the plant community. Species sensitivities to CO2 are driven in large part by different photosynthetic metabolisms. There is abundant evidence that productivity of plants with the C3 photosynthetic metabolism responds more to CO2 than C4 plants (12–14) because the photosynthetic metabolism of the former is not CO2-saturated at present-day atmospheric concentrations as it is in C4 plants (12). However, this trend can be complicated by other plant responses to CO2, like improved water use efficiency (7) or seedling recruitment (10), both of which are not simply related to species differences in the photosynthetic pathway. The performance of species (and presumably functional groups like photosynthetic class) depends on interactions of the CO2 responses with soil resources and interspecific plant competition, greatly complicating the predictions of species/plant community responses (14).
Shrub encroachment into many world native grasslands over the past 200 years is a well documented species shift that has been attributed in part to this differential species sensitivity to CO2 (12, 15), with predominantly C3 woody vegetation displacing C4 grasses. Bond and Midgley (16) proposed that differences in plant carbon allocation strategies between woody and herbaceous plants in addition to differences in photosynthetic metabolism are involved in woody plant encroachment. However, no direct evidence of this theory has yet been reported, leading some to question the importance of CO2 in this phenomenon (17, 18).
We are embarking on a period in which plant community dynamics in many of Earth's ecosystems may be subject to relatively rapid changes due to increased atmospheric CO2 levels and altered climate. These changes may affect not only the traditional goods and services we have come to expect from these lands but may alter their biodiversity as well. Our group previously evaluated the production responses of native Colorado shortgrass steppe vegetation to 5 years of growth (1997–2001) at elevated [CO2] by using large open-top chambers (OTCs) (10). Here, we use data from this same experiment (http://sgslter.colostate.edu/data/acquisitionplcy.htm) to examine how a doubling of CO2 concentration for 5 years with OTCs placed over a native shortgrass steppe altered the botanical structure of this important semiarid grassland on the western edge of the North American Great Plains. Indices of diversity and a plant community association index were used to evaluate plant community structure. Analyses of functional plant group responses expand the discussion into plant mechanisms involved in the CO2 responses and consequences of botanical changes on rangeland agriculture.
Results and Discussion
A total of 34 different plant species were observed in our nine experimental plots during the July species harvests over the 5 years of the CO2 enrichment experiment (1997–2001). However, the bulk of the aboveground plant biomass (81% averaged across years and CO2 treatments) was contributed by just three perennial grass species: Bouteloua gracilis (C4 photosynthetic pathway), Pascopyrum smithii (C3 photosynthetic pathway), and Stipa comata (C3). Biomass from other species included a 7% contribution of a sedge, Carex eleocharis Bailey; 4% contributed by the subshrub Artemisia frigida Willd; and a 7% contribution of the remaining 29 species, a variety of grasses and herbaceous dicots.
The Shannon diversity index (H′) was used to evaluate how growth at the two [CO2] values affected diversity of aboveground plant biomass among the 34 species. Diversity has both evenness (J′) and richness (species number) components, and Wilsey et al. (19) suggest that all three indices are needed to adequately describe diversity in grassland ecosystems. None of these indices distinguishes contributions or associations of particular species; instead, they rely on species numbers and abundance rankings (20). However, none was significantly affected by CO2 treatment (Fig. 1A, C, and E), although small but significant responses due to year were observed (Fig. 1 B, D, and F) and an apparent chamber effect was noted for richness (Fig. 1A). Thus, doubling CO2 for 5 years had very little impact on these common measures of plant species diversity.
Fig. 1.
Whittaker's (21) community association index was used as a community-wide dissimilarity metric for comparing differences in plant species composition between the elevated and ambient CO2 plots. Values range from 0 to 1, with 0 indicating identical communities and 1 indicating dissimilar communities with no species in common. In contrast to the diversity indices, a clear and significant (P = 0.04) linear trend was observed in the index, indicating increasing dissimilarity in the association of species between the ambient and elevated CO2 plant communities over time, with the index increasing from 0.17 in the year before CO2 fumigation to 0.33 by 2001 (Fig. 2). No significant change was observed for Whittaker's index in a comparison between ambient CO2 and nonchambered control plots, which was conducted to evaluate possible chamber effects on plant species associations (data not shown).
Fig. 2.
To further evaluate plant community changes over time, we partitioned the 34 species observed at our site into four functional groups based on photosynthetic pathway and morphology (13, 14): C3 grasses (composed of six grass species and one sedge), C4 grasses (four species), herbaceous forbs (22 minor, mostly C3 species), and a single woody subshrub species (A. frigida). We examined how peak aboveground biomass was affected by CO2. ANOVA (Table 1) detected significant treatment effects for C3 grasses (P = 0.002), significant (P < 0.05) year effects for all groups but subshrub (P = 0.06), and significant or strong trends for all treatment by year interactions (P ranged from 0.03 to 0.11); means comparison tests therefore were determined among treatments within each year.
Table 1.
| Source of variation | Num df | Den df | F value | Pr > F |
|---|---|---|---|---|
| C3 grasses | ||||
| Year | 5 | 21 | 32 | <0.0001 |
| Treatment | 2 | 6.0 | 20 | 0.0023 |
| Year × treatment | 10 | 21 | 2.7 | 0.0278 |
| C4 grasses | ||||
| Year | 5 | 23 | 26 | <0.0001 |
| Treatment | 2 | 6.1 | 1.2 | 0.3691 |
| Year × treatment | 10 | 23 | 1.9 | 0.0972 |
| Forbs | ||||
| Year | 5 | 21 | 3.1 | 0.0294 |
| Treatment | 2 | 5.8 | 1.5 | 0.3006 |
| Year × treatment | 10 | 21 | 2.2 | 0.0586 |
| Subshrub | ||||
| Year | 5 | 18 | 2.6 | 0.0585 |
| Treatment | 2 | 3.9 | 2.7 | 0.1881 |
| Year × treatment | 10 | 19 | 1.9 | 0.1113 |
For most of the years of the experiment, C3 grasses experienced significantly greater productivity under elevated CO2, whereas C4 grasses and forbs were unresponsive to CO2 (Fig. 3 A–C). Although the differences between C3 and C4 grasses were predicted based on differences in photosynthetic pathway, greater seedling recruitment of S. comata, a dominant C3 grass and the only grass species to respond to CO2, was likely an important aspect of the C3 reaction to CO2 (10). Low growing season precipitation in 2000 (Fig. 3D) and lack of soil water recharge in 2001 resulted in consistently low soil water content in the final 2 years of the experiment (22), which likely was an important factor in the correspondingly low productivity of C3 grasses in 2000 and 2001 (Fig. 3A). Forb species appeared too sporadically among the different study years and experimental plots for a robust test of their CO2 responsiveness (Fig. 3C). However, biomass of the subshrub A. frigida showed a temporal trend of increasing productivity in elevated CO2 plots compared with ambient and control plots that became significant in the final year of the experiment, when biomass in elevated CO2 plots exceeded ambient plots by ≈9-fold (Fig. 3D). Compared with the first year of CO2 enrichment, 1997, biomass increased 40-fold from 0.72 g·m−2 in 1997 to 28.7 g·m−2 in 2001 in the elevated CO2 plots.
Fig. 3.
Measurements of plant cover when averaged over seven or eight time periods during each growing season showed patterns similar to the biomass data. Cover is a different metric of plant growth than biomass because it incorporates the extent to which a species or a group of plants occupies the aerial environment. Trends emerged within most years, suggesting greater cover of C3 grasses and forbs and less cover of C4 grasses under elevated CO2 (Fig. 4 A–C). However, no significant temporal trends (changing cover with time) were observed for C3 grasses for any treatment, and forb cover increased only slightly with years in unchambered plots. Cover of C4 grasses in chambers maintained at both ambient and elevated [CO2] declined similarly from 1997 through 2001 (Fig. 4B), and cover of the subshrub increased for all treatments (Fig. 4D). The temporal decline in C4 cover was likely due to low soil water content in the final 2 years of the experiment (22) and may reflect a commonly observed drought escape strategy of the system C4-dominant B. gracilis to shed its leaves during prolonged drought. Increased cover of the subshrub A. frigida over the same period suggests greater competitive ability of this tap-rooted subshrub for deeper soil water, especially under elevated CO2. Only A. frigida exhibited clear treatment differences, with plant cover under elevated CO2 increasing significantly faster from 1997 through 2001 compared with both ambient (P = 0.006) and unchambered (P = 0.001) plots (Fig. 4D). Although the absolute changes in A. frigida cover under elevated CO2 were small, from 0.2% to 4.1%, they represent a >20-fold increase and movement over the course of the experiment out of the minor species category, representing ≈10% of the total community plant cover by the final year of the experiment.
Fig. 4.
The lack of detectable changes in diversity due to variable [CO2] indicates no important alterations in the number and abundance structure of plant species, and the decreasing similarity between plant communities exposed to elevated and ambient [CO2] suggests shifts in species composition and their relative contributions to the plant community. Whittaker's community association index is apparently able to detect CO2-induced changes in community structure because of its reliance on the relative contributions of particular species, whereas diversity, evenness, and richness, which are species-blind, may change more slowly. Increases in Whittaker's index were no doubt partially due to enhanced growth of S. comata (10) and A. frigida (Fig. 3D) but also were likely influenced by a number of other species that collectively became more dissimilar over time but when assayed individually did not display significant responses to CO2.
The progressively enhanced biomass and cover of the subshrub A. frigida under elevated CO2 provide evidence from a native grassland that supports the notion that rising atmospheric [CO2] may be contributing to shrubland expansions of the past 100–200 years (12, 15, 16). The increased proportion of woody plants has reduced significantly the available forage in many world grasslands and, without proactive management measures like burning, has rendered these lands less suitable for livestock grazing. It has long been argued that livestock grazing, fire suppression, and climate change have been responsible for this shrub encroachment, but sorting out the relative importance of these competing forces from the CO2-induced invasion is challenging (18). No field experiment has confirmed the CO2-induced shrub expansion hypothesis, and, consequently, the role of CO2 in shrubland expansion has been questioned (17).
A. frigida, commonly called fringed sage, is an aromatic, mat-forming perennial shrub, ranging 10–60 cm tall (23). Although small in stature, it is the most widely distributed and abundant Artemisia species in the world, occurring from Mexico north through primarily the western United States and Canada to Alaska. Although considered a native to the United States, it also can be found in Siberia, Mongolia, and Kazakhstan. The superior CO2 response of A. frigida, which ranks among the highest reported in the literature (24, 25), may be attributed to one or more mechanisms. High CO2 has been hypothesized to speed up recovery of woody plants from fire or grazing (16). Our experiment, which included simulated grazing through alternate year defoliation, posits that the benefit of higher atmospheric CO2 to woody plants may be extended to improved access to soil water, possibly through increased carbon allocation to its roots. A. frigida is not characteristically a deeply rooted plant, but Coupland and Johnson (26) point out that A. frigida root systems will adjust to periods of more abundant water supply by developing deeper tap roots. Such conditions may have prevailed in the early years of the experiment, when growing season precipitation was relatively abundant (3-year average of 435 mm for years 1997–1999 compared with a long-term value of 280 mm) (Fig. 3D) and soil water content was generally high, especially in the elevated CO2 treatment (22). Compared with shallow-rooted grasses like the site-dominant B. gracilis, the tap-root of A. frigida may enhance its ability to extract soil water at depths that can increase under elevated CO2 (10, 22, 26, 27). Greater access to soil water may have been especially important under the conditions of low soil water content that prevailed in the final 2 years of the experiment (22). Other characteristics that may explain greater CO2 sensitivity of A. frigida are its C3-photosynthetic metabolism, especially in comparison to the C4 system-dominant B. gracilis; long-lived aboveground woody tissues; and high nitrogen use efficiency, which has been hypothesized to confer greater CO2 sensitivity in woody plants compared with the less nitrogen use-efficient grasses they are displacing (28).
A. frigida tends to increase under heavy grazing and other disturbances, is unpalatable to livestock, invades deteriorated grasslands, and is considered a weed (23, 29). It is also one of the least desirable Artemisia species for wildlife (23, 30), although it may be used by small hoofed mammals, like pronghorn (Antilocapra americana Ord.) or elk (Cervus canadensis Erxleben), especially in the overwintering months, and can be an important food source for sage-grouse (Centrocercus urophasianu). Our results, which indicate that growth of A. frigida can be enhanced dramatically simply by increasing ambient [CO2], suggest that rising atmospheric [CO2] already may be causing important changes in the ecology of the semiarid grasslands of the western Great Plains. Further research is needed to determine whether these results can be extended to the performance of A. frigida in other North American or Asian grasslands or to other major shrub species (12, 15, 16), including other Artemesia spp. (31), whose expansion into many world grasslands is already well documented.
The results of this experiment clearly highlight the importance of evaluating ecological changes from several perspectives. Although diversity, evenness, and richness seemed unaffected by 5 years of CO2 enrichment, Whittaker's index showed a clear trend of community change. However, the only significant species responses to emerge from this experiment were the previously reported greater productivity of the C3 grass S. comata under elevated CO2 (10) and the finding that growth of A. frigida increased more rapidly at elevated CO2 than at ambient CO2. Like A. frigida, S. comata also has potentially negative consequences for forage quality (11). It seems plausible that the gradual increase in atmospheric [CO2] that has been underway for >200 years may induce important changes in diversity, but documenting such changes in a relatively short-term experiment on small plots is a challenge.
Materials and Methods
Site Description.
The experiment was conducted at the United States Department of Agriculture–Agricultural Research Service Central Plains Experimental Range (latitude 40° 50′ N longitude 104° 43′ W) at the northern limit of the shortgrass steppe, a semiarid grassland on the western edge of the North American Great Plains used extensively for livestock grazing. Long-term annual precipitation and growing season precipitation amounts are 320 and 280 mm, respectively (22). The effect of elevated CO2 on this native ecosystem was investigated using OTCs (4.5 m in diameter, ≈3.8 m in height, enclosing 15.5 m2 of ground area or a volume of 60.5 m3). A portion of the pasture initially was divided into three blocks, and three 15.5-m2 circular plots per block were randomly selected as the experimental treatment plots: a chambered, ambient CO2 treatment (360 ± 20 ppm CO2); a chambered, elevated CO2 treatment (720 ± 20 ppm CO2); and an unchambered control plot of equal ground area, which was used to assess the chamber effect. [CO2] was not controlled at night and was generally <450 ppm of CO2 in all treatments. The chamber [CO2] was checked approximately weekly, and the range was generally ±20 ppm from the target treatment concentrations. From late March until mid-October from 1997 to 2001, the OTCs surrounded two plots in each of the three blocks (for a total of six chambers). CO2 fumigation proceeded in the elevated chamber CO2 treatment as soon as the chambers were placed on the plots each spring and was continued until they were removed in the autumn when vegetation was dormant. For more detail, see ref. 10.
Plant Biomass and Plant Cover.
Aboveground plant biomass was measured by species in 1996 (the year before CO2 treatments) and from 1997 to 2001 (the years of CO2 treatments) during the period of peak standing phytomass (late July). A metal wire grid containing 56 quadrats, with each quadrat measuring 40.5 × 15.3 cm (3.46 m2 total), was placed over the southern half of each plot, and vegetation in every other quadrat (28 quadrats) was clipped to the crown, separated by species, dried at 60°C, and weighed. This defoliation protocol removes 50% of the green vegetation and was incorporated into the CO2 treatment experimental design primarily as a means to represent the nominal grazing conditions for these grasslands (10).
The percentage of plant cover by species was estimated visually monthly from April through October in 10 quadrats, each measuring 10 × 10 cm, placed within the biomass metal wire sampling grid in each of the OTCs.
Community Measures.
From the plant biomass data, we calculated richness as the number of species, the widely used Shannon diversity index (H′), as
where pi is the proportion of the ith species (s) in the population and eveness (J′) is
according to Pielou (20).
We used Wittaker's index of community association (21), defined here as A, as a dissimilarity metric and calculated it as
where pi(a) and pi(e) are the proportions of the ith species (s) in the ambient and elevated plots, respectively. Values range from 0 to 1, with values of 0 indicating identical communities and 1 being dissimilar communities, having no species in common.
Statistical Analyses.
We used a linear mixed model with repeated measures (SAS Proc MIXED; SAS Institute, Cary, NC) to analyze for CO2 treatment affects on richness, species evenness (J′), and species diversity (H′). A linear mixed model (SAS Proc GLIMMIX; SAS Institute) also was used to evaluate aboveground biomass responses among four plant functional groups: C3 grasses (two dominant perennial grasses, four minor grass species, and one Carex species), C4 grasses (one dominant perennial grass, three minor species), herbaceous dicots (22 minor species, mostly C3), and one subshrub (A. frigida). In both the diversity and functional group analyses, “year” was used as a repeated measure variable, “block” was specified as a random effect (thereby removing the variability due to blocking), and block·CO2 treatment was used as the error term for CO2 treatment comparisons. A first-order, autoregressive covariance structure was used for the covariance structure of the aboveground biomass data set.
Whittaker's community association index (21) was analyzed for year effects in contrasts involving ambient vs. elevated plots by regressing year on the index using a repeated measures mixed model.
Temporal trends in plant cover were evaluated by averaging cover of the four functional groups, C3 grasses, C4 grasses, herbaceous forbs, and the subshrub, across each year's six or seven sampling dates and then regressing percentage cover on year and treatment by using a repeated measures mixed model for each functional group. A heterogeneous first-order, autoregressive covariance structure was used to account for autocorrelated responses and increasing variance over time. The subshrub was the only functional group in which temporal trends (P < 0.10) in plant cover were observed for all three treatments. Estimate statements were added to the Proc MIXED statements to test whether the linear increase in mean percentage cover for the elevated treatment exceeded the same for the ambient treatment and the control treatments of the subshrub group.
Abbreviation
- OTC
- open-top chamber.
Acknowledgments
We thank Mary Ashby, Jeff Thomas, Jim Nelson, Mary Smith, Susan Crookall, Larry Tisue, Stacey Poland, Jennifer King, and David Jensen for technical assistance and Brian Wilsey, Wayne Polley, and two anonymous reviewers for helpful comments on the manuscript. This research was supported in part by National Science Foundation Terrestrial Ecology and Global Change Award IBN-9524068, National Science Foundation Award DEB-9708596, and Shortgrass Steppe Long-Term Ecological Research Project DEB-9350273.
References
1
E Jansen, J Overpeck, KR Briffa, J-C Duplessy, F Joos, V Masson-Delmotte, D Olgao, B Otto-Bliesner, WR Peltier, S Rahmstorf, et al., Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Climate Change 2007: The Physical Science Basis, eds S Solomon, D Qin, M Manning, Z Chen, M Marquis, KB Avery, M Tignor, HL Miller (Cambridge Univ Press, Cambridge, UK), pp. 443–497 (2007).
2
BD Campbell, DM Stafford-Smith, GM McKeon Global Change Biol 3, 177–187 (1997).
3
IR Noble, H Gitay J Veg Sci 7, 329–336 (1996).
4
VR Squires, A Sidahmed Ann Arid Zone Res 36, 79–96 (1997).
5
HW Polley, JA Morgan, M Stafford-Smith, B Campbell Climate Change and Global Crop Productivity, eds KR Reddy, HF Hodges (CAB Intl, Oxfordshire, UK), pp. 293–314 (2000).
6
PA Niklaus, C Körner Ecol Monogr 74, 491–511 (2004).
7
JA Morgan, DE Pataki, C Körner, H Clark, SJ Del Grosso, JM Grünzweig, AK Knapp, AR Mosier, PCD Newton, PA Niklaus, et al. Oecologia 140, 11–25 (2004).
8
RA Gill, HW Polley, HB Johnson, LJ Anderson, H Maherall, RB Jackson Nature 417, 279–282 (2002).
9
CE Owensby, RM Cochran, LM Auen Carbon Dioxide, Populations, and Communities, eds C Körner, F Bazzaz (Academic, San Diego, CA), pp. 363–371 (1996).
10
JA Morgan, AR Mosier, DG Milchunas, DR LeCain, JA Nelson, WJ Parton Ecol Appl 14, 208–219 (2004).
11
DG Milchunas, AR Mosier, JA Morgan, DR LeCain, JY King, JA Nelson Agri Ecosyst Environ 111, 166–184 (2005).
12
HW Polley J Range Manage 50, 562–577 (1997).
13
H Poorter, M-L Navas N Phytol 157, 175–198 (2003).
14
PB Reich, D Tilman, J Craine, D Ellsworth, MG Tjoelker, J Knops, D Wedin, S Naeem, D Bahauddin, J Goth, et al. N Phytol 150, 435 (2001).
15
SB Idso Climate Change 22, 85–86 (1992).
16
WJ Bond, GF Midgley Global Change Biol 6, 865–869 (2000).
17
OW Van Auken Annu Rev Ecol Syst 31, 197–215 (2000).
18
S Archer, DS Schimel, EA Holland Climate Change 29, 91–99 (1995).
19
BJ Wilsey, DR Chalcraft, CM Bowles, MR Willig Ecology 86, 1178–1184 (2005).
20
EC Pielou J Theort Biol 13, 131–144 (1966).
21
RH Whittaker Ecol Monogr 22, 1–44 (1952).
22
JA Nelson, JA Morgan, DR LeCain, AR Mosier, DG Milchunas, BA Parton Plant Soil 259, 169–179 (2004).
23
, ed JK Francis (US Dep of Agric Forest Service, Provo, UT, General Technical Report IITF-WB-1. Wildland Shrubs of the United States and its Territories: Thanmic Description, available at www.fs.fed.us/global/iitf/pdf/shrubs/Artemisia%20frigida.pdf. (2002).
24
EA Ainsworth, SP Long N Phytol 165, 351 (2005).
25
JE Mohan, LH Ziska, WH Schlesinger, RB Thomas, RC Sicher, K George, JS Clark Proc Natl Acad Sci USA 103, 9086–9089 (2006).
26
RT Coupland, RE Johnson J Ecol 53, 475–507 (1965).
27
DR LeCain, JA Morgan, DG Milchunas, AR Mosier, JA Nelson, DP Smith Plant Soil 279, 219–228 (2006).
28
F Teyssonneyre, C Picon-Cochard, R Falcimagne, J-F Soussana Global Change Biol 8, 1034–1046 (2002).
29
TD Whitson Weeds of the West (Univ of Wyoming, Laramie, WY), pp. 1–630 (1997).
30
AY Cooperrider, JA Bailey, eds ED McArthus, BL Welch (US Dep Agric For Serv, Ogden, UT), pp. 46–54, General Technical Report INT-200. (1984).
31
PA Knapp, PT Soule Ann Assoc Am Geographers 86, 387–411 (1996).
Information & Authors
Information
Published in
Proceedings of the National Academy of Sciences
Vol. 104 | No. 37
September 11, 2007
September 11, 2007
PubMed: 17785422
Classifications
Copyright
© 2007 by The National Academy of Sciences of the USA. Freely available online through the PNAS open access option.
Submission history
Received: April 13, 2007
Published online: September 11, 2007
Published in issue: September 11, 2007
Keywords
Acknowledgments
We thank Mary Ashby, Jeff Thomas, Jim Nelson, Mary Smith, Susan Crookall, Larry Tisue, Stacey Poland, Jennifer King, and David Jensen for technical assistance and Brian Wilsey, Wayne Polley, and two anonymous reviewers for helpful comments on the manuscript. This research was supported in part by National Science Foundation Terrestrial Ecology and Global Change Award IBN-9524068, National Science Foundation Award DEB-9708596, and Shortgrass Steppe Long-Term Ecological Research Project DEB-9350273.
Notes
This article is a PNAS Direct Submission.
Authors
Competing Interests
The authors declare no conflict of interest.
Metrics & Citations
Metrics
Citation statements
Altmetrics
Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFGet Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to get full access to it.