Printer Friendly

Productivity patterns of C3 and C4 functional types in the U.S. Great Plains.

INTRODUCTION

Vegetation patterns are expected to be altered by global climate change (Melillo et al. 1991, Field et al. 1992, Post 1993). One approach for predicting the response of vegetation is to apply a climate change scenario to existing vegetation-environment relationships (Melillo et al. 1991). Large-scale relationships between vegetation and environmental variables have traditionally used biomes or vegetation types as units of aggregation (Holdridge 1947, Whittaker 1975, Prentice 1990). However, there may be considerable variation in the structure of vegetation within a given biome or type (Smith et al. 1993). Examining more resolute units of vegetation (i.e., less aggregated than biome or vegetation type) at large spatial scales will increase our understanding of the environmental controls over plant community structure and may improve our capacity to predict vegetation responses.

The use of species to describe regional vegetation patterns is constrained by availability of data and the limited ranges of species (Smith et al. 1993). Large-scale questions are best addressed by grouping species with similar characteristics into plant functional types (Box 1981, Prentice et al. 1992, Smith et al. 1993, Lauenroth et al. 1996). Aggregating species into functional types creates a component of vegetation that potentially ranges over regions and continents (Woodward 1987) yet preserves the physiological characteristics of the constituent species (Chapin 1993). The plant functional type is gaining widespread acceptance among ecologists interested in the response of ecosystems to global change (Woodward 1987, Schimel 1993).

A major distinction among species in temperate grassland communities of North America is photosynthetic pathway (Teeri and Stowe 1976, Brown 1993). Differential temperature responses result in a geographic separation of species having the [C.sub.3] or [C.sub.4] photosynthetic pathway, with [C.sub.4] species generally occupying warmer locations. However, the extent to which environmental factors relate to the production of [C.sub.3] and [C.sub.4] functional types at regional scales has yet to be determined.

Several studies have attempted to relate the abundances of [C.sub.3] and [C.sub.4] functional types to abiotic variables within central North America (Teeri and Stowe 1976, Sims et al. 1978, Boutton et al. 1980, Barnes et al. 1983, Fan 1993). The spatial extent of these studies varied from portions of states (Boutton et al. 1980, Barnes et al. 1983, Fan 1993) to the central and western United States (Sims et al. 1978) to most of North America (Teeri and Stowe 1976). Data sources have also varied in these correlational analyses from regional flora (Teeri and Stowe 1976) to biomass samples (Sims et al. 1978, Boutton et al. 1980, Barnes et al. 1983) to rangeland surveys (Fan 1993). Most studies on the abundance and distribution of [C.sub.3] and [C.sub.4] plants for grasslands throughout the globe have relied on floristic data (Rundel 1980, Werger and Ellis 1981, Hattersley 1983, Cavagnaro 1988). An exception being Tieszen et al. (1979) who used isotopic data as an indicator of proportional production along an altitudinal gradient in Kenya. No study, however, has assessed the productivities of [C.sub.3] and [C.sub.4] plants throughout the Great Plains of the United States or any similar large temperate grassland region using an extensive production data set. Developing quantitative relationships between the productivities of these functional types and environmental variables is crucial to understanding the dynamics of plant communities in the Great Plains and other grassland regions.

The main goal of our analysis is to quantify the production of [C.sub.3] and [C.sub.4] functional types in terms of the major environmental factors that govern their distribution and abundance in the Great Plains. Our specific objectives are to (1) quantify [C.sub.3] and [C.sub.4] production in terms of climate and soil variables; (2) isolate effects of climate on [C.sub.3] and [C.sub.4] production by maintaining a constant soil texture; and (3) isolate effects of soil texture on [C.sub.3] and [C.sub.4] production by examining specific locations in the Great Plains with different average climatic conditions.

METHODS

We constructed a spatial database of plant species production and environmental variables for the Great Plains of the United States. Plant species production data were collected from USDA Natural Resource Conservation Service (NRCS) range site descriptions. NRCS range sites represent the potential native plant community of well-managed grazing lands in the absence of abnormal disturbances and other management regimes (USDA 1967). Range sites are unique in the combination of total annual yield and plant community composition. Range site descriptions include the relative production (percent) attributed to each species in the plant community, as well as total community production in favorable, normal, and unfavorable years. These data are based on total growth during a single growing season and determined by harvesting plant material at various growth stages (USDA 1967). The NRCS uses range sites to inventory rangelands and to assist users in their management and conservation. We collected and automated [approximately equal to]1700 range site descriptions for our database.

We calculated absolute production (grams per square meter) for each species within a range site by multiplying the proportion of production for a species by the total site production (grams per square meter) during normal years. Species in the database were labeled as either [C.sub.3] or [C.sub.4] based on information from the literature (Waller and Lewis 1979, Stubbendieck et al. 1992). The relative and absolute production for [C.sub.3] and [C.sub.4] functional types were obtained for each range site by summing the values of individual species.

Range site descriptions were spatially located using NRCS State Soil Geographic (STATSGO) databases, organized in a geographic information system (GIS) (Environmental Systems Research Institute 1992). STATSGO databases aggregate county soil survey maps to represent soil patterns at the scale of a state or region (1:250 000) (USDA 1991). The databases are used primarily for large-scale resource management and monitoring. In STATSGO databases, states are divided into polygons, or Soil Associations (SA), representing areas of similar aggregate soil characteristics. The minimum size of a SA is 625 ha. Each SA is a weighted average (by area) of up to 21 components, which are either named soil series or nonsoil designations; each component is also characterized by a range site that can be uniquely related to a range site description. The range site attributed to each SA component is independent of the soil type (i.e., there is a many-to-many relationship between soil type and range site). Production for a SA was calculated as the weighted average of the production values for the component range sites. Data were only used for SAs in which every component could be related to a range site description.

The use of rangeland survey data limited the database to states in which at least some land is managed as rangeland. For the purpose of this analysis, it was also essential that range site descriptions included the relative production of each species in the plant community. The database for this study included Kansas, Nebraska, North Dakota, South Dakota, Texas, and the Great Plains portions of Colorado, Montana, New Mexico, and Wyoming. Oklahoma has [approximately equal to]37.5% of its area as rangeland (Lauenroth et al. 1994), but NRCS range site descriptions for the state did not include the relative production of individual species. Therefore, it was not included in the analysis.

The degree of aggregation in this data set makes it appropriate for regional-scale studies. Data from our range site database compared favorably with data compiled from other studies throughout the Great Plains (Paruelo and Lauenroth 1996). The correlation coefficients (r) were 0.78 for models of both [C.sub.3] and [C.sub.4] relative abundance. Relationships were not significantly different from the 1:1 line. We consider this database to be the most extensive, and perhaps the best, data set on productivity currently available for the Great Plains.

We used climatic data from 296 weather stations throughout the region (CLIMATEDATA 1988). For each station, daily values of precipitation, and daily minimum and maximum temperatures were collected for the period 1969-1988. Annual precipitation and average annual temperature were calculated for each station by year and averaged over the 20-yr period. The GIS was used to generate mean annual temperature isotherms of 0.25 [degrees] C intervals and mean annual precipitation isohyets of 1-cm intervals for the region. Mean annual temperature increases from north to south, whereas, mean annual precipitation increases from the west to the southeast. Values for areas between contours were determined by the midpoint of the surrounding contours. Soil texture data (percentage sand and clay) were generated from the texture classification of the surface soil layer for each STATSGO polygon (Burke et al. 1991). Thus, any geographic point in the Great Plains could be associated with the relative and absolute production of each functional type, and values for mean annual temperature, mean annual precipitation, and soil texture.

Statistical models

From our database we generated a set of spatially random points throughout the Great Plains using a sampling scheme within the GIS. This scheme essentially placed a fine grid over the study area and selected 200 random grid cells. We used 200 points in the analysis for several reasons; we wished to have enough points to adequately represent the region, yet not too many such that every independent variable was statistically significant. Additionally we tried to avoid the situation of having multiple points fall within the same STATS-GO polygon. A second set of 200 random points was selected and analyzed to confirm that the sampling scheme and number of points were representing well the variation in our database. We performed a stepwise multiple regression analysis of the production of [C.sub.3] and [C.sub.4] grasses with respect to climatic and soil texture variables. Our analysis focused on grasses, the dominant life-form in the Great Plains. Results based on photosynthetic pathway that included trees and shrubs may have been confounded by a greater component of woody species in the [C.sub.3] functional group. Both relative and absolute production were analyzed because there are important distinctions between the two variables. Relative production, the percentage of total aboveground net primary production (ANPP) accounted for by a functional type, is an indicator of dominance, whereas absolute production (ANPP in gram/square meter) represents total productivity of a functional type.

We used four predictor variables in the analysis: mean annual temperature (MAT), mean annual precipitation (MAP), percentage sand (SAND), and percentage clay (CLAY). We selected environmental variables that have substantial gradients within the Great Plains and that have been shown to relate to [C.sub.3] and [C.sub.4] functional type production (Sims et al. 1978, Boutton et al. 1980, Fan 1993). MAP and soil texture have also been shown to be correlated with total ANPP for this region (Sala et al. 1988). Seasonality of precipitation may be an important control over [C.sub.3] and [C.sub.4] productivity in grasslands and shrublands of North America (Paruelo and Lauenroth 1996). However, it does not vary substantially within the Great Plains (Paruelo et al. 1995) and was not included in our analysis. In contrast, thermal amplitude, or seasonality of temperature, did not relate significantly to the relative abundances of [C.sub.3] and [C.sub.4] grasses in North American grasslands and shrub-lands (Paruelo and Lauenroth 1996). We therefore restricted our analysis to the coarse climatic variables of MAT and MAP. We also included percentage sand and percentage clay as soil texture variables.

We refrained from using higher order or log functions unless the linear regressions were inadequate to significantly account (P [less than] 0.05) for variation in functional type production. Proper diagnostic checks were performed on model residuals. All models met the assumptions regarding residuals with the exception of the model for absolute production of [C.sub.3] grasses; the distribution of residuals for this model was significantly different than normal. Log transformations of the data relieved the problem, however this did not improve model predictions. We therefore decided to use all un-transformed linear models. After developing the regression models, we used them to examine patterns of functional type production within the Great Plains.

Relationships with climatic variables

To examine controls of climate on [C.sub.3] and [C.sub.4] production within the Great Plains, we held soil texture constant. We assumed a sandy loam as the soil texture since it is representative of rangelands and remnant native grasslands (SAND = 55% and CLAY = 15%) (Cosby et al. 1984, Muhs and Maat 1993). The multiple regression models were used to examine productivity patterns of [C.sub.3] and [C.sub.4] grasses within the environmental space of MAT and MAP. Two characteristics of the patterns of [C.sub.3] and [C.sub.4] grasses that can potentially change with climate are the ecotone between areas dominated by [C.sub.3] or [C.sub.4] grasses and the range limits of these functional types. To examine the ecotone between [C.sub.3] and [C.sub.4] grasses, we intersected the models of [C.sub.3] and [C.sub.4] production (i.e., we set the regression equations for [C.sub.3] and [C.sub.4] grasses equal to each other and solved for MAP and MAT). This identified the climatic values at which production of [C.sub.3] and [C.sub.4] grasses was equivalent. To examine range limits, we quantified the climatic values [TABULAR DATA FOR TABLE 1 OMITTED] at which production of these functional types reached zero.

Relationships with soil texture variables

To analyze the importance of soil texture in determining regional patterns of [C.sub.3] and [C.sub.4] production, we examined four geographic locations within the Great Plains (northwest, northeast, southwest, and southeast) that represented extremes of MAT and MAP. The northwest location (northeastern Montana) had MAT of 7 [degrees] C and MAP of 30 cm. The northeast location (southeastern North Dakota) had MAT of 7 [degrees] C and MAP of 50 cm. The southwest location (southeastern New Mexico) had MAT of 15 [degrees] C and MAP of 45 cm, and the southeast location (eastern Oklahoma) had MAT of 15 [degrees] C and MAP of 100 cm. We used the multiple regression models to analyze [C.sub.3] and [C.sub.4] production at each location along a gradient of percentage sand. To simplify this analysis, we assumed percentage sand plus percentage clay equal to 100%. Thus, the gradient of increasing sand content can also be viewed as a gradient of decreasing clay content.

RESULTS

Statistical models

The environmental variables used in the analysis explained 67-81% of the variation in production of [C.sub.3] and [C.sub.4] grasses (Table 1). MAT was the most explanatory variable for relative production of [C.sub.3] grasses (C3PCT), relative production of [C.sub.4] grasses (C4PCT), and absolute production of [C.sub.3] grasses (C3PROD), by explaining [greater than]64% of the variation in these dependent variables. It was also a significant factor in explaining the absolute production of [C.sub.4] grasses (C4PROD). MAP was the most important variable in the model for C4PROD, explaining 73% of the variation. It was also a significant factor for C3PCT, C3PROD, and C4PCT A soil texture variable was never the most explanatory variable. However, one or both soil texture variables were always significant factors in each model. The models were used to generate maps of [C.sub.3] and [C.sub.4] relative and absolute production for the Great Plains [ILLUSTRATION FOR FIGURE 1A-D OMITTED]. Relative production of [C.sub.3] grasses decreased, and relative production of [C.sub.4] grasses increased, from the north to the southeast. Absolute production of [C.sub.3] grasses increased from the south to the northeast, and absolute production of [C.sub.4] grasses increased from the northwest to the southeast.

We compared predictions from our regression models with actual data from Sims et al. (1978). Predicted values were strongly related to observed values for all four models; correlation coefficients (r) ranged from 0.68 to 0.97. Intercepts, however, were significantly greater than zero, and slopes were significantly less than one. This is to be expected due to the aggregation of our data. We would expect our predictions to be slightly "muted," not quite capturing extreme values of productivity due to averaging.

Relationships with climatic variables

Holding soil texture constant as sandy loam, we examined in more detail the relationships between production and climate. Relative and absolute production of [C.sub.3] grasses declined with increasing MAT (Fig. 2C AND B). Relative production of [C.sub.3] grasses also decreased with increasing MAP; absolute production of [C.sub.3] grasses, however, increased slightly with MAP. Relative and absolute production of [C.sub.4] grasses increased with MAT and also with MAP [ILLUSTRATION FOR FIGURES 2C AND D OMITTED].

By intersecting the planes of [C.sub.3] and [C.sub.4] production in relation to MAT and MAP, we isolated the climatic conditions where productivities of these functional types are equivalent. Soil texture was again held constant using the percentages for sandy loam. The lines of equivalent relative and absolute production were plotted on the two-dimensional environmental space of MAT and MAP [ILLUSTRATION FOR FIGURE 3 OMITTED].

The temperature at which [C.sub.3] production equals [C.sub.4] production decreases with increasing MAP. For relative production, the crossover point in terms of MAT ranges from over 8 [degrees] C at [less than] 30 cm of rainfall to 5 [degrees] C at 65 cm of MAP. For absolute production, the crossover point ranges from over 11 [degrees] C at [less than]30 cm of MAP to just over 3 [degrees] C at 54 cm. Lines of equivalent relative and absolute production should effectively be equal (i.e., if relative productions of the two functional groups are equal, absolute productions must also be equal). The difference between these lines represents a measure of discrepancy between the models of relative and absolute production. This discrepancy occurs within the process of aggregating data from several range sites, described in the Methods.

The analysis of functional type production with respect to climatic variables illustrated the relatively large proportion of the environmental space of the Great Plains dominated by [C.sub.4] grasses as opposed to [C.sub.3] grasses [ILLUSTRATION FOR FIGURE 3 OMITTED]. However, the proportion of environmental space dominated by these functional types differs somewhat from the proportion of geographic space dominated. In order to examine the geographic dominance of [C.sub.3] and [C.sub.4] grasses, we used the regression models (soil texture held constant as sandy loam) to generate the isopleth of equivalent [C.sub.3] and [C.sub.4] relative production [ILLUSTRATION FOR FIGURE 4 OMITTED]. [C.sub.3] grasses dominate a greater proportion of geographic space than environmental space. They dominate [approximately equal to]26% of the Great Plains, as compared with 74% for [C.sub.4] grasses. [C.sub.3] grasses dominate the northern Great Plains, including most of Montana, North Dakota, and eastern Wyoming, and some of South Dakota and northern Colorado; [C.sub.4] grasses dominate the remainder of the region.

We also examined the values of MAT and MAP at which the production of [C.sub.3] and [C.sub.4] grasses reached zero [ILLUSTRATION FOR FIGURE 5 OMITTED]. Based on the models, production of [C.sub.4] grasses does not reach zero within the Great Plains of the United States. At 30 cm of MAP, [C.sub.4] absolute production reaches zero below 0 [degrees] C MAT, corresponding to locations close to the boundary between grassland and boreal forest in Canada. This, however, is an extrapolation beyond the range of our data. Relative production of [C.sub.4] grasses also does not reach zero within the U.S. Our models predicted [C.sub.4] grasses to be found further north in areas of higher precipitation. The value of MAT at which [C.sub.3] relative production equals zero ranged from 17.5 [degrees] C at [less than]40 cm of rainfall to 12 [degrees] C at close to 110 cm. This corresponds geographically to an area extending from central Texas to northeastern Kansas. The relative production model predicted [C.sub.3] grasses to be found further south in the western portion of the region. The model of absolute production of [C.sub.3] grasses showed the opposite pattern. The positive relationship between [C.sub.3] absolute production and MAP yielded this discrepancy in the models.

Relationships with soil texture variables

[C.sub.4] production was negatively related to CLAY and positively related to SAND. [C.sub.3] production showed the opposite pattern. C3PCT was positively associated with CLAY and negatively associated with SAND, and C3PROD was also negatively related to SAND.

We analyzed [C.sub.3] and [C.sub.4] production in relation to soil texture for four geographic locations within the Great Plains. At all four locations, we found the trends described above; [C.sub.3] abundance decreased and [C.sub.4] abundance increased with increasing sand and decreasing clay content. At the northwest location (MAT = 7 [degrees] C, MAP = 30 cm), absolute production of [C.sub.3] grasses was always greater than [C.sub.4] grasses. Relative production of [C.sub.3] grasses was greater than [C.sub.4] grasses when sand content was less than [approximately equal to] 87% (clay content [greater than] 13%). There are some areas in the northwest Great Plains where sand content can be [greater than]87%, in particular the westward extension of the Nebraska sandhills in southeastern Wyoming. Thus, whereas most of the northwestern Great Plains is dominated by [C.sub.3] grasses, the potential exists for [C.sub.4] grasses to dominate sites with high sand content.

In the northeast location (MAT = 7 [degrees] C, MAP = 50 cm), relative production of [C.sub.3] grasses was higher than relative production of [C.sub.4] grasses when sand content was less than [approximately equal to]70% (clay content [greater than]30%). Absolute production of [C.sub.3] grasses was greatest when sand content was less than [approximately equal to]55% (clay content [greater than]45%). There are many areas in the northeast portion of the region that have sand content [less than]55%, and clay content can exceed 45% in the region. Based on the model, plant communities in the northeast Great Plains could potentially be dominated by either [C.sub.3] or [C.sub.4] grasses, depending on soil texture. [C.sub.3] grasses would have greater production than [C.sub.4] grasses on sites with fine-textured soils, and [C.sub.4] grasses would dominate sites with coarse-textured soils. However, areas with fine-textured soils in the eastern Great Plains are usually cultivated.

In the southwest portion of the Great Plains (MAT = 15 [degrees] C, MAP = 45 cm), [C.sub.3] grasses were predicted to be more abundant than [C.sub.4] grasses only when soil sand content was less than [approximately equal to]5%. Sand content below 5% or clay content exceeding 95% is unlikely on any given site. Thus, the models suggest that [C.sub.4] grasses dominate the plant community for practically all soils. In the southeast portion of the region (MAT = 15 [degrees] C, MAP = 100 cm), [C.sub.4] species are predicted to be more abundant than [C.sub.3] species regardless of soil texture.

DISCUSSION

Results of this study improve our understanding of how climate controls the geographic distribution of [C.sub.3] and [C.sub.4] functional types in the Great Plains. Teeri and Stowe (1976) related the number of [C.sub.4] species in the flora to climatic variables for North America. Using Eq. 1 from their results, the geographic area where 50% of species in the flora have the [C.sub.4] photosynthetic pathway is predicted to fall in central Kansas at [approximately equal to]38 [degrees] N latitude. Our study indicates that 50% of total production is attributed to [C.sub.4] species at [approximately equal to]43 [degrees] N latitude (Nebraska - South Dakota border). This implies that the percentage of species having [C.sub.3] and [C.sub.4] photosynthetic pathways throughout the Great Plains is quite different from the percentage of production accounted for by these functional types. Ehleringer (1978) used a physiological approach to model the daily carbon gain of [C.sub.3] and [C.sub.4] grasses vs. latitude in the Great Plains during the month of July. He found the geographic crossover point to be at 45 [degrees] N latitude. This result is closer to the transition between [C.sub.3]-dominated and [C.sub.4]-dominated areas predicted by our models.

While relationships between functional type production and MAT are supported by differential optimum temperatures for photosynthesis (Black 1973, Ehleringer 1978), relationships with MAP are not explicitly clean [C.sub.4] species have a higher water use efficiency than [C.sub.3] species (Black 1973) and thus are often predicted to be more abundant in drier areas (Brown and Simmons 1979, Pearcy and Ehleringer 1984). However, [C.sub.4] species dominate the wettest portions of the Great Plains. MAP is a strong determinant of aboveground net primary productivity in this region (Sala et al. 1988). Therefore MAP should be positively correlated with the absolute production of both of these functional types. But, why does [C.sub.3] relative production decrease and [C.sub.4] relative production increase with MAP?

Several hypotheses can be posed to explain why [C.sub.3] grasses become less dominant with increasing MAP. First, MAP and MAT are correlated in the Great Plains (r =0.46, P [less than] 0.0001). The MAP gradient in the Great Plains is from west to southeast, so MAT increases with MAP. MAT also increases along the MAP gradient as a result of a decrease in elevation from west to east in the Great Plains, and the effect of increased humidity on the diurnal variation in temperature. Summer temperatures, in particular normal daily minimum temperatures during the summer months, increase from west to east in this region (USDC 1968). The higher minimum temperatures during the growing season may favor [C.sub.4] plants over [C.sub.3] plants in the east. Teeri and Stowe (1976) explained [C.sub.4] species distribution using the average daily minimum temperature in July. Second, MAP may be related to other environmental variables that explain this pattern. Because winter tends to be the driest season throughout the Great Plains (Borchert 1950), the increase in MAP from west to east occurs mostly as an increase in summer precipitation. An eastward increase in summer rainfall is most pronounced in the northern Great Plains during July and August (Borchert 1950), therefore [C.sub.4] plants may have an advantage since they are more likely to benefit from summer rainfall.

There could be several biotic explanations for why relative production of [C.sub.3] grasses declined with increasing precipitation. Most of the dominant tallgrasses in the Great Plains are [C.sub.4] species: Andropogon gerardii, Andropogon scoparius, Panicum virgatum, Sorghastrum nutans, and Spartina pectinata (Kucera 1991). These grasses may have a competitive advantage in moist areas where light is a limiting resource. There are, however, some [C.sub.3] tallgrasses, the most common being Stipa spartea (Lauenroth et al. 1994). Studies on competitive abilities of [C.sub.3] and [C.sub.4] plants have been conflicting (Roush and Radosevich 1985, Wray and Strain 1987, Snaydon 1991) and inconclusive (Baskin and Baskin 1978).

Nitrogen may also be a limiting factor on sites with sufficient moisture. Recently, studies have shown that [C.sub.4] grasses persisted and displaced [C.sub.3] grasses on tallgrass prairie sites with low soil nitrogen (Tilman and Wedin 1991, Wedin 1990). This can be due to lower leaf nitrogen concentrations and greater nitrogen use efficiencies in [C.sub.4] species (Brown 1978, Schmitt and Edwards 1981, Field and Mooney 1986). Lower nitrogen concentrations in litter of [C.sub.4] species led to lower nitrogen mineralization rates in plots of [C.sub.4] species relative to plots of [C.sub.3] species (Wedin and Tilman 1990, Wedin and Pastor 1993). This provides a positive feedback for the stability of [C.sub.4]-dominated communities (Wedin and Tilman 1990, Wedin and Pastor 1993, Wedin 1995). Conversely, [C.sub.3] grasses had greater biomass relative to [C.sub.4] grasses when nitrogen was added to tallgrass prairie plots (Tilman 1996).

Relationships between soil texture and [C.sub.3] and [C.sub.4] functional type production have not received much attention. We found [C.sub.3] grasses to be positively related to clay content and negatively related to sand content; relationships were opposite for [C.sub.4] grasses. Although these relationships were derived from random points throughout the Great Plains, we need to consider that soil textures are not distributed uniformly across the region. Eolian deposition of sand from the west led to the formation of sandy soils in the semiarid and sub-humid portions of the Great Plains (Muhs and Maat 1993). Thus the majority of soils with high sand content in the Great Plains are found in the western and central portions of the region. The heterogeneous distribution of sand in the Great Plains and the greater range of sand content in the west could imply that our soil texture results are biased toward soil-vegetation relationships found in the western part of the region.

Quantitative models derived for [C.sub.3] and [C.sub.4] functional type production in the Great Plains have the potential to be used as predictive tools to analyze the response of vegetation in the region to climate change. While there is considerable uncertainty about predicted climates, global circulation models suggest that changes in central North America will be relatively large compared to other regions (Mitchell et al. 1990). Based on the regression models, the relative production of [C.sub.3] grasses could potentially decrease [approximately equal to]4% for each additional degree of MAT. Absolute production of [C.sub.3] grasses could potentially decrease [approximately equal to]11 g/[m.sup.2], and absolute production of [C.sub.4] grasses could increase [approximately equal to]6 g/[m.sup.2], for each additional degree of MAT. This could correspond to a northward expansion of [C.sub.4] grasses. We can use the models to predict the geographic dominance of [C.sub.3] and [C.sub.4] grasses under current climatic conditions and with a 2 [degrees] C increase in MAT. [C.sub.3] grasses dominate 35% of the region under current conditions as opposed to 19% after the temperature increase. On the other hand, predicted changes in the precipitation regime, with more rainfall in winter and less in summer (Mitchell et al. 1990), could favor [C.sub.3] grasses over [C.sub.4] grasses.

Both of these functional types are essentially found throughout the entire Great Plains. Our models predict the absence of [C.sub.3] grasses in the very southern portion of the region, however Stipa leucotricha, a [C.sub.3] grass with a winter growing season, is native to the southern Great Plains (Stubbendieck et al. 1992). The ubiquity of these functional types suggests that a response to climate change could occur rapidly. The relationships formulated in this study essentially average the productivity patterns of component species within [C.sub.3] and [C.sub.4] functional types. Species or subtypes may have distinct productivity patterns that will not always coincide with the patterns of their associated functional groups (Ellis et al. 1980, Hattersley 1983, Epstein et al. 1996; H. E. Epstein et al., unpublished manuscript).

The relationships formulated in this study are an important step in understanding the environmental controls over plant functional type production. We have identified large-scale environmental variables that explain much of the variation in [C.sub.3] and [C.sub.4] productivity for the Great Plains. These variables are important inclusions in models that predict the future abundances of these functional types and the impact of climate change on vegetation at the regional scale. A logical continuation of this work is to test the regression equations developed here for other grassland regions, as more data become available on the productivities of [C.sub.3] and [C.sub.4] grasses throughout the world.

ACKNOWLEDGMENTS

This project was supported by the National Science Foundation (BSR 91-06183, BSR 90-11659, and BSR 90-13888). We wish to thank Martha Coleman, Caroline Yonker, and Weihong Fan for their assistance in the development of the regional database and the Natural Resource Conservation Service for providing the range site descriptions. Jim Zumbrunnen provided statistical assistance. We thank Tim Seastedt and Larry Tieszen for providing valuable comments and improving the manuscript.

LITERATURE CITED

Barnes, P. W., L. L. Tieszen, and D. J. Ode. 1983. Distribution, production, and diversity of [C.sub.3]- and [C.sub.4]-dominated communities in a mixed prairie. Canadian Journal of Botany 61:741-751.

Baskin, J. M., and C. C. Baskin. 1978. A discussion of the growth and competitive ability of [C.sub.3] and [C.sub.4] plants. Castanea 43:71-76.

Black, C. C., Jr. 1973. Photosynthetic carbon fixation in relation to net C[O.sub.2] uptake. Annual Review of Plant Physiology 24:253-286.

Borchert, J. R. 1950. The climate of the central North American grassland. Annals of the Association of American Geographers 40:1-39.

Boutton, T W., A. T Harrison, and B. N. Smith. 1980. Distribution of biomass of species differing in photosynthetic pathway along an altitudinal transect in southeastern Wyoming grassland. Oecologia (Berlin) 45:287-298.

Box, E. O. 1981. Predicting physiognomic vegetation types with climate variables. Vegetatio 45:127-139.

Brown, D. A. 1993. Early nineteenth-century grasslands of the midcontinent plains. Annals of the Association of American Geographers 83:589-612.

Brown, R. H. 1978. A difference in N use efficiency in [C.sub.3] and [C.sub.4] plants and its implication in adaptation and evolution. Crop Science 18:93-98.

Brown, R. H., and R. E. Simmons. 1979. Photosynthesis of grass species differing in C[O.sub.2] fixation pathways: I. Water-use efficiency. Crop Science 19:375-379.

Burke, I. C., T. G. F. Kittel, W. K. Lauenroth, P. Snook, C. M. Yonker, and W. J. Parton. 1991. Regional analysis of the Central Great Plains. Bioscience 41:685-692.

Cavagnaro, R. 1988. Distribution of [C.sub.3] and [C.sub.4] grasses at different altitudes in a temperate arid region of Argentina. Oecologia (Berlin) 76:273-277.

Chapin, F. S. 1993. Functional role of growth forms in ecosystem and global processes. Pages 287-312 in J. R. Ehleringer and C. B. Field, editors. Scaling physiological processes. Academic Press, San Diego, California, USA.

CLIMATEDATA. 1988. U.S. West Optical Publishing, 90 Madison Street Suite 200, Denver, Colorado, USA.

Cosby, B. J., G. M. Hornberger, R. B. Clapp, and T. R. Ginn. 1984. A statistical exploration of the relationships of soil moisture characteristics to the physical properties of soils. Water Resources Research 20:682-690.

Ehleringer, J. R. 1978. Implications of quantum yield differences on the distributions of [C.sub.3] and [C.sub.4] grasses. Oecologia (Berlin) 31:255-267.

Ellis, R. P., J. C. Vogel, and A. Fuls. 1980. Photosynthetic pathways and the geographical distribution of grasses in South West Africa/Nambia. South African Journal of Science 76:307-314.

Environmental Systems Research Institute. 1992. ARC/INFO Version 6.1.1. Redlands, California, USA.

Epstein, H. E., W. K. Lauenroth, I. C. Burke, and D. P. Coffin. 1996. Ecological responses of dominant grasses along two climatic gradients in the Great Plains of the U.S. Journal of Vegetation Science 7(6): in press.

Fan, W. 1993. Regional analysis of plant species and environmental variables in eastern Colorado. Dissertation. Colorado State University, Fort Collins, Colorado, USA.

Field, C. B., F. S. Chapin III, P. A. Matson, and H. A. Mooney. 1992. Responses of terrestrial ecosystems to the changing atmosphere: a resource-based approach. Annual Review of Ecology and Systematics 23:201-235.

Field, C. B., and H. A. Mooney. 1986. The photosynthesis - nitrogen relationship in wild plants. Pages 25-55 in T. J. Givnish, editor. On the economy of plant form and function. Cambridge University Press, Cambridge, England.

Hattersley, P. W. 1983. The distribution of [C.sub.3] and [C.sub.4] grasses in Australia in relation to climate. Oecologia (Berlin) 57: 113-128.

Holdridge, L. R. 1947. Determination of world formations from simple climatic data. Science 105:367-368.

Kucera, C. L. 1991. Tallgrass prairie. Pages 227-263 in R. T. Coupland, editor. Ecosystems of the world 8A: Natural grasslands. Elsevier Science, New York, New York, USA.

Lauenroth, W. K., D. P. Coffin, I. C. Burke, and R. A. Virginia. 1996. Interactions between demographic and ecosystem processes: a challenge for functional types. In T. M. Smith, I. A. Woodward, and H. H. Shugart, editors. Plant functional types. Cambridge University Press, Cambridge, England.

Lauenroth, W. K., D. G. Milchunas, J. L. Dodd, R. H. Hart, R. K. Heitschmidt, and L. R. Rittenhouse. 1994. Effects of grazing on ecosystems of the Great Plains. Pages 69-101 in M. Vavra, W. A. Laycock, and R. D. Peiper, editors. Ecological implications of livestock herbivory in the west. Society for Range Management, Denver, Colorado, USA.

Melillo, J. M., T. V. Callaghan, F. I Woodward, E. Salati, and S. K. Sinha. 1991. Effects on ecosystems. Pages 287-310 in R. T. Coupland, editor. Ecosystems of the world 8A: Natural grasslands. Elsevier Science, New York, New York, USA.

Mitchell, J. F. B., S. Manabe, V. Meleshko, and T. Tokioka. 1990. Equilibrium climate change and its implications for the future. Pages 131-174 in J. T. Houghton, G. J Jenkins, and J. J. Ephraums, editors. Climate change: The IPCC scientific assessment. Cambridge University Press, New York, New York, USA.

Muhs, D. R., and P. B. Maat. 1993. The potential response of eolian sands to greenhouse warming and precipitation reduction on the Great Plains of the U.S.A. Journal of Arid Environments 25:351-361.

Paruelo, J. M., and W. K. Lauenroth. 1996. Climatic controls of the distribution of plant functional types in grasslands and shrublands of North America. Ecological Applications. 6:1212-1224.

Paruelo, J. M., W. K. Lauenroth, H. E. Epstein, I. C. Burke, M. R. Aguiar, and O. E. Sala. 1995. Regional climatic similarities in the temperate zones of North and South America. Journal of Biogeography 22:915-925.

Pearcy, R. W., and J. Ehleringer. 1984. Comparative eco-physiology of [C.sub.3] and [C.sub.4] plants. Plant, Cell and Environment 7:1-13.

Post, W. M. 1993. Uncertainties in the terrestrial carbon cycle. Pages 116-132 in A.M. Solomon and H. H. Shugart, editors. Vegetation dynamics and global change. Chapman and Hall, New York, New York, USA.

Prentice, I. C., W. Cramer, S. P. Harrison, R. Leemans, R. A. Monserud, and A. M. Solomon. 1992. A global biome model based on plant physiology and dominance, soil properties and climate. Journal of Biogeography 19:117-134.

Prentice, K. C. 1990. Bioclimatic distribution of vegetation for general circulation model studies. Journal of Geophysical Research 95:11811-11830.

Roush, M. L., and S. R. Radosevich. 1985. Relationships between growth and competitiveness of four annual weeds. Journal of Applied Ecology 22:895-905.

Rundel, P. W. 1980. The ecological distribution of [C.sub.4] and [C.sub.3] grasses in the Hawaiian Islands. Oecologia (Berlin) 45: 354-359.

Sala, O. E., W. J. Parton, L. A. Joyce, and W. K. Lauenroth. 1988. Primary production of the Central Grassland Region of the United States. Ecology 69:40-45.

Schimel, D. S. 1993. Population and community processes in the response of terrestrial ecosystems to global change. Pages 45-54 in P. M. Karieva, J. G. Kingsolver, and R. B. Huey, editors. Biotic interactions and global change. Sinauer Associates, Sunderland, Massachusetts, USA.

Schmitt, M. R., and G. E. Edwards. 1981. Photosynthetic capacity and nitrogen use efficiency of maize, wheat and rice: a comparison between [C.sub.3] and [C.sub.4] photosynthesis. Journal of Experimental Botany 128:459-466.

Sims, P. L., J. S. Singh, and W. K. Lauenroth. 1978. The structure and function of ten western North American grasslands: I. Abiotic and vegetational characteristics. Journal of Ecology 66:251-285.

Smith, T. M., H. H. Shugart, F. I. Woodward, and P. J. Burton. 1993. Plant functional types. Pages 272-292 in A.M. Solomon and H. H. Shugart, editors. Vegetation dynamics and global change. Chapman and Hall, New York, New York, USA.

Snaydon, R. W. 1991. The productivity of [C.sub.3] and [C.sub.4] plants: a reassessment. Functional Ecology 5:321-330.

Stubbendieck, J., S. L. Hatch, and C. H. Butterfield. 1992. North American range plants. Fourth edition. University of Nebraska Press, Lincoln, Nebraska, USA.

Teeri, J. A., and L. G. Stowe. 1976. Climatic patterns and the distribution of [C.sub.4] grasses in North America. Oecologia (Berlin) 23:1-12.

Tieszen, L. L., M. M. Senyimba, S. K. Imbamba, and J. H. Troughton. 1979. The distribution of [C.sub.3] and [C.sub.4] grasses and carbon isotope discrimination along an altitudinal and moisture gradient in Kenya. Oecologia (Berlin) 37:337-350.

Tilman, D. 1996. Biodiversity: population versus ecosystem stability. Ecology 77:350-363.

Tilman, D., and D. A. Wedin. 1991. Dynamics of nitrogen competition between successional grasses. Ecology 72: 1038-1049.

USDA. 1967. National handbook for range and related grazing lands. Soil Conservation Service, U.S. Department of Agriculture, Washington, D.C., USA.

USDA. 1991. State soil geographic data base (STATSGO): data users guide. Soil Conservation Service Miscellaneous Publication Number 1492.

USDC. 1968. Climate atlas of the United States. Environmental Science Service Administration, U.S. Department of Commerce, Washington, D.C., USA.

Waller, S.S., and J. K. Lewis. 1979. Occurrence of [C.sub.3] and [C.sub.4] photosynthetic pathways in North American grasses. Journal of Range Management 32:12-28.

Wedin, D. A. 1990. Nitrogen cycling and competition among grass species. Dissertation. University of Minnesota, Minneapolis, Minnesota, USA.

-----. 1995. Species, nitrogen, and grassland dynamics: the constraints of stuff. Pages 253-262 in C. W. Jones and J. H. Lawton, editors. Linking species and ecosystems. Chapman and Hall, New York, New York, USA.

Wedin, D. A., and J. Pastor. 1993. Nitrogen mineralization dynamics in grass monocultures. Oecologia 96:186-192.

Wedin, D. A., and D. Tilman. 1990. Species effects on nitrogen cycling: a test with perennial grasses. Oecologia (Berlin) 84:433-441.

Werger, M. J. A., and R. P. Ellis. 1981. Photosynthetic pathway in the arid regions of South Africa. Flora 171:64-75.

Whittaker, R. H. 1975. Communities and ecosystems. Second edition. Macmillan, New York, New York, USA.

Woodward, F. I. 1987. Climate and plant distribution. Cambridge University Press, Cambridge, UK.

Wray, S. M., and B. R. Strain. 1987. Competition in old-field perennials under C[O.sub.2] enrichment. Ecology 68:1116-1120.
COPYRIGHT 1997 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Epstein, H.E.; Lauenroth, W.K.; Burke, I.C.; Coffin, D.P.
Publication:Ecology
Date:Apr 1, 1997
Words:7130
Previous Article:Long-term CO2 enrichment of a pasture community: species richness, dominance, and succession.
Next Article:Litter decomposition and nitrogen dynamics in aspen forest and mixed-grass prairie.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters