Printer Friendly

Evaluating relationships between spatial heterogeneity and the biotic and abiotic environments.

INTRODUCTION

Many ecologists are beginning to recognize the role of heterogeneity in ecological systems by acknowledging its influence on population dynamics and biodiversity (MacArthur and Pianka, 1966; Wiens, 1976; Turner and Gardner, 1991; Sarnelle et al., 1993). In fact, it has been suggested that heterogeneity is actually the root of biological diversity at all levels of ecological organization and should serve as the foundation for conservation and ecosystem management (Christensen, 1997; Osffeld et al., 1997; Wiens, 1997). Concern regarding the role of spatial heterogeneity in influencing population and community dynamics has gained recent attention (Levin, 1992; Wiens, 1997; White and Walker, 1997; Sanderson and Harris, 2000; Fuhlendorf and Engle, 2001). Spatially discrete disturbance patterns are described historically as a shifting mosaic (Kay, 1998; Fuhlendorf and Engle, 2001), where the presence of alternative habitat types may have provided (1) complementary resources (i.e., food availability, cover from predators or climate) that improved habitat quality for some species and/or (2) unique habitat types capable of supporting a variety of species, both of which presumably enhance biodiversity. Because heterogeneity is largely associated with spatial and temporal variability, it is highly scale dependent and can be influenced by many factors including grazing and topo-edaphic features (Fuhlendorf and Smeins, 1996, 1999).

Increasing evidence suggests that specific microclimates within a species' range can have profound effects on population dynamics (survival, movements, fitness) and may dictate habitat utilization throughout the year (Sharp and Van Horne, 1999; Sedgeley, 2001). The influence of temperature on the biotic community has been widely documented (Daubenmire, 1974; Roads et al., 1994; Loik et al., 2000), but landscape temperature patterns are reported rarely in ecological studies (Chen et al., 1996). Thermal properties are critical aspects of animal survival throughout the summer months. To maintain body temperature, animals must balance heat gained from the environment by heat loss to the environment. Metabolic rates required to maintain body temperatures are influenced positively by the gradient between body and ambient temperature (Calder and King, 1974). Thus, animals are forced to entertain tradeoff decisions between foraging behavior, predator avoidance and thermal refuge. To tolerate extremely high temperatures, plants and animals must adapt physiologically, morphologically and/or behaviorally to their thermal environments (Calder and King, 1974; Jones, 1983). Animals that efficiently utilize microhabitats to conserve energy may gain an advantage by reallocating these reserves to other vital processes (Walsberg, 1985, 1986, 1993; Webb and Rogers, 1988).

Although vegetation characteristics (i.e., availability and distribution of food, structure and cover) dictate habitat selection for many wildlife species (Kendeigh and Fawver, 1981; Ostfeld, 1985; Pierson and Wight, 1991), they also are a critical moderator of local temperatures (Pianka, 1988). Further, livestock grazing may be the most influential process governing changes in vegetation structure and composition (Holechek et al., 1989; Milchunas and Lauenroth, 1993). Thus, direct and indirect influences of temperature and livestock grazing can have an influence on ecosystem function by influencing distribution and abundance of a wide range of animal species including insects, birds and mammals (Bock et al., 1984; Dennis et al., 1998; Ritchie, 2000). For instance, interactions between air temperature and livestock grazing can influence soil moisture, photosynthetic, respiration and decomposition rates creating unique habitat types that ultimately influence distribution and abundance of plant and animal populations (Geiger, 1965; Daniel et al., 1979; Perry, 1994). Consequently, distribution of landscape-level temperature gradients may have profound effects on distribution and abundance of biological communities, especially those occupying hot arid environments (Karr and Freemark, 1983; Schleucher, 1999). Hence, moderation of the thermal environment is a critical abiotic factor and landscape function governing all life processes and should be considered as an important factor in community ecology.

As levels of atmospheric C[O.sub.2] continue to increase (IPCC, 2001, 2007; Lal et al., 2003; Lal, 2008), debates about the influence of elevated atmospheric C[O.sub.2] and global temperature have gained recent attention and will likely continue in the foreseeable future. However, there is no debate regarding the ability of temperature to dictate distribution and abundance of biotic communities (Calder and King, 1974; Jones, 1983; Walsberg, 1993). Thus, quantifying relationships between landscapes and the thermal environment will likely play an increasing role in conservation and restoration of many ecological systems. Our study site provided us an excellent opportunity to examine relationships between soil-surface temperature, vegetation and grazing on patterns of landscape heterogeneity. Our site is located within the Southern Great Plains of North America where extremely high air temperatures (mean = 37 C) and livestock grazing are common throughout the summer months.

Although habitat selection is generally described by quantifying vegetation characteristics (i.e., % cover, structure, species composition) within a species home range, other potentially important habitat features such as soil moisture (Kendeigh and Fawver, 1981) and temperature (Pierson and Wight, 1991) should be further evaluated as key indicators of site selection for many biological organisms. It is unclear how patterns of soil-surface temperature might influence distribution and abundance of plant and animal populations. To better understand the ecological importance of these relationships to conservation and restoration of our native ecosystems, we designed an experiment to document the spatial relationships between thermal patterns of landscape variability and vegetation characteristics associated with livestock grazing. Thus, we characterize fine-scale heterogeneity at the landscape or pasture level to determine if external factors associated with grazing intensity are capable of influencing the heterogeneity of biotic and abiotic components at finer spatial scales. We hypothesized that soil-surface temperatures and heterogeneity would be influenced strongly by patterns of vegetation and grazing intensity.

METHODS

STUDY AREA

The study area was located approximately 15 km south of Clinton, Oklahoma in the Rolling Red Plains Resource Area of the southern Great Plains. Average annual precipitation was 77 cm and ranged from 51 to 82 cm (Fuhlendorf et al., 2002). Approximately 70% of the rainfall occurred during the primary growing season from Apr. to Sep. The 600-ha research station was largely rolling uplands cut by several steep drainages with a mean elevation of 490 m. Rock outcrops and bare areas were common. Soils were highly erosive and primarily classified as a Cordell silty clay loam with a depth of 25 to 36 cm over solid siltstone (Moffatt and Conradii, 1979; Fuhlendorf et al., 2002). The vegetation was typical of the mixed-grass prairie with variable dominant species dependent upon topo-edaphic effects and land use. On uplands the dominant species were a mixture of grasses with variable stature. Dominant mid-grasses included Bouteloua curtipendula, Aristida purpurea and Bothnochloa saccharoides. Short grasses were more abundant with shallow soils or heavy grazing and include Bouteloua gracilis, Buchloe dactyloides and Bouteloua hirsuta. Tallgrasses were less abundant and restricted to protected mesic sites but included Schizachydum scoparium, Sorghastrum nutans and Andropogon gerardii. There also was a high diversity of herbaceous dicots that varied with annual fluctuations in precipitation. Woody plant species included Rhus glabra and Prunus angustifolia in isolated portions of the landscape, as well as the widely distributed sub-shrub, Guiterrezia sarothrae. Riparian zones made up < 5% of the total area and were dominated by Populus deltoides, Ulmus americana, Bumelia lanuginosa and Sapindus Drummondii (Fuhlendorf et al., 2002). Taxonomic nomenclature followed Hatch et al. (1990).

DATA COLLECTION

Replicated treatments were established on 6 ca. 45 ha pastures that were subjected to heavy (n = 2), moderate (n = 2), and ungrazed (n = 2) treatments to determine associations between soil-surface temperature and grazing. Heavily and moderately grazed sites were stocked with stocker cattle (mean = 270 kg) at 2 ha/animal and 3.75 ha/animal, respectively, from 1 Apr. to 31 Sep. We established 200-m transects in each replication (n = 6) that were centered on and ran perpendicular through a riparian zone so that the effects of topographic position (upland, riparian) and spatial patterns of variability would be maximized for each transect. We used a 0.25-[m.sup.-2]) quadrat to characterize plant cover and soil-surface temperatures at 1-m intervals along the transect. Estimates of percent ground cover by plant functional group (grass, forb, shrub, litter and bare ground) were recorded using the mid-point of the following cover classes of 0-5, 6-25, 26-50, 51-75, 76-95 and 96-100% (Daubenmire, 1959). Litter was defined as the combined senesced shoots and leaves of previous years growth associated with grasses, forbs, shrubs and trees and is assessed as a percentage of total ground cover. Because we estimated % ground cover in 3-dimensional space, sums of all cover classes within a single quadrat were capable of exceeding 100%. Tree canopy cover within the riparian area was dominated by Populus deltoides and Ulmus americana and generally covered all plots within the riparian area. Soil-surface temperatures were recorded during the period of greatest physiological stress (Aug.) for plant and animal communities in this region. We restricted collection times between 1100 and 1700 h when air temperature and light intensity were maximized (i.e., 0% cloud cover). Air temperature and solar radiation ranged from 36.3-37.3 C and 733-894 watts/[m.sup.2], respectively, during our collection times. Soil-surface temperatures were recorded using an OMEGASCOPE OS531 hand held infra-red thermometer (One Omega Drive, Box 4047, Stamford, CT 06907). Temperature measurements were intended to represent the conditions important to ground-level organisms and were obtained by placing the thermometer ca. 3 cm above the surface of the ground. Soil-surface temperatures are defined as that layer of material between the thermometer and the ground surface but could include bare soil, rock, leaf litter or herbaceous material at the soil-surface. We recorded soil-surface temperatures at five locations within each quadrat: four at 6 cm from each quadrat corner and one directly in the center. To quantify relationships between vegetation structure and soil-surface temperatures, we also recorded the maximum plant vegetation height from the center of each quadrat and determined the angle of obstruction along 8 radii at 45[degrees] compass intervals (i.e., the circle space occupied by each quadrat was split into 8 sectors at every 45[degrees] around the center) to quantify vertical and horizontal vegetation structure, an important indicator of ground nesting bird habitat (Kopp et al., 1998; Harrell and Fuhlendorf, 2002). The angle of obstruction is a function of plant height and the ground distance of a particular plant from a point in space. Subsequently, the angle of obstruction integrates multiple habitat parameters associated with the vertical and horizontal dimension of habitat structure and is an excellent tool used as a measure of overall structural heterogeneity (Harrell and Fuhlendorf, 2002). Smaller angles of obstruction (i.e., < 30[degrees]) are structurally classified as more open, sparsely vegetated habitat, while larger angles of obstruction (i.e., > 75[degrees]) are structurally classified as more closed, densely vegetated habitat. All upland measurements were recorded on slopes < 3[degrees] and slopes along the upland-riparian edge were generally < 5[degrees] with an east-west orientation.

DATA ANALYSES

We compared percent cover of plant functional groups between grazing treatments (heavy, moderate and ungrazed) and landscape position (riparian, upland) using PROC ANOVA (SAS Institute Inc., 2001). Interactions were not significant (P > 0.05) so sources of variation were distributed among main factor effects (grazing treatment and landscape position). To assess the influence of vegetation cover variables on temperature, we used stepwise regression analysis (PROC REG; SAS Institute, Inc., 2001), with temperature as the dependent variable and cover of plant functional groups as the independent variables. Variables selected for inclusion in the model were significant when P [less than or equal to] 0.15 (Hosmer and Lemeshow, 1989; Sams et al., 1996; Cody and Smith, 1997). The suitability of this model was tested using the Hosmer-Lemeshow goodness-of-fit statistic at P > 0.05 (Hosmer and Lemeshow, 1989). We also examined relationships between cover of plant functional groups and temperature using Pearson correlation analysis (PROC CORR, SAS Institute Inc., 2001).

[FIGURE 1 OMITTED]

To determine spatial patterns of temperature along the line, we calculated semi-variograms to characterize spatial patterns of soil-surface temperatures within each grazing treatment (Clark, 1980; Turner et al., 1991). The semi-variance is the sum of squared differences between all possible pairs of points separated by a particular distance, typically where measurements are uniformly spaced along a straight line (Turner et al., 1991). Semi-variograms are defined by (1) the sill, which is the horizontal portion of the graph where the semi-variance tends to level off at the maximum semivariance, (2) the range of influence is determined by the lag distance associated with the sill (commonly referred to as patch size), where points farther apart than the range are independent (because at distances greater than the range the semivariance equals the sample variance) and points closer than the range bear some relationship to one another and (3) the nugget or y-intercept, which represents sampling error or variability occurring at scales smaller than the sampling interval (Clark, 1980). Semi-variograms typically fit 4 types of theoretical models (linear, gaussian, exponential and spherical) used to interpret spatial data (Fig. 1). A flat variogram, also called the 'nugget effect', indicates random variation, where all measurements along the line are unrelated to one another, implying zero spatial correlation (Trangmar et al., 1985).

RESULTS

Grazing altered the composition and cover of these grassland ecosystems. Amount of bare ground increased with grazing and was 3-fold greater in heavily grazed than in ungrazed pastures (Table 1). In contrast, leaf litter decreased with grazing intensity and was nearly 10-fold greater in ungrazed sites (Table 1). Grass cover was higher in ungrazed than in heavily grazed sites, but there was no difference in forb or shrub cover between grazing treatments (Table 1). Vegetation height and angle of obstruction were greatest in ungrazed sites and decreased with grazing intensity (Table 1).

Mean soil-surface temperatures did not differ between heavy (mean = 48.5 C), moderate (mean = 47.4 C) and ungrazed (mean = 49.1; P > 0.10) treatments. As a result, data were pooled across treatments to evaluate differences between landscape positions (upland and riparian). Analysis of variance showed that upland (n = 1045, mean = 51 C, c.v. = 14.51) soil-surface temperatures were greater (P < 0.001) than those within the riparian area (n = 155, mean = 30 C, c.v = 22.78; Fig. 2). Only 6% of upland vs. 96% of riparian soil-surface temperatures were < 40 C (Table 2).

Best-fit multiple regression models that predicted soil-surface temperature from vegetation characteristics within the riparian ([R.sup.2] = 0.09 F = 2.43, P = 0.121, 2 variable model) and uplands ([R.sup.2] = 0.22, F = 12.27, P = 0.001, 3 variable model) explained only a small percentages of variation in soil-surface temperature (Table 3). The significance of the upland model was primarily due to the large sample size (n = 1045). On uplands, bare ground was the best predictive variable of soil-surface temperature, but accounted for only 12% of the variation (Table 3), where increases in bare ground and leaf litter were associated positively with temperature (Fig. 3). Grass cover and vegetation height were correlated negatively with soil-surface temperature (Fig. 3). Although significant correlations existed between community parameters (bare ground, cover of leaf litter, cover of grass and vegetation height) and soil-surface temperature, all correlation coefficients were small (r < [+ or -] 0.34, Fig. 3).

[FIGURE 2 OMITTED]

Standard deviations in soil-surface temperatures were greater between quadrat averages than within quadrats for both upland and riparian sites. Upland standard deviations between quadrat averages ranged from 7.1 to 8.0 C across all grazing treatments. Within the riparian zone, the highest variability occurred in the moderately grazed treatment followed by ungrazed and heavily grazed treatments (Table 4). Standard deviations of soil-surface temperatures within 0.25-[m.sup.2] quadrats were similar across all upland grazing treatments and averaged about 4 C, while riparian standard deviations within 0.25-[m.sup.2] quadrats ranged from 1 to 2 C (Table 4). However, some 0.25-[m.sup.2] quadrats had individual temperatures that varied by as much as 34 C. Standard deviations in soil-surface temperature within quadrats were 1.9 to 3.5 times greater on uplands than within riparian zones (Table 4).

Mean widths of the riparian zone was similar for heavy (33 m), moderate (36 m), and ungrazed (28 m) treatments and allowed us to make direct semi-variance comparisons between treatments. The form of the semi-variogram based on soil-surface temperature was rather unique compared with semi-variograms based on vegetation characteristics. Semi-variance of soil-surface temperature was spatially dependent and rose continuously with lag-distance in heavily and ungrazed pastures, indicating a high degree of continuity (i.e., exponential model; Fig. 4). However, moderately grazed treatments also were spatially dependent, but the relationship was spherical, indicating that moderately grazed sites had a higher degree of landscape variability than other grazing treatments (i.e., points > 16 m are unrelated or independent of one another; Fig. 4). For example, semi-variance of 2503, occurred at 8 m in moderately grazed sites vs. 16 m in heavily and un-grazed sites, indicating that moderately grazed treatments exhibited a higher degree of thermal heterogeneity at the pasture level (Fig. 4).

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

For all variables other than soil-surface temperature and forb cover, semi-variances of moderately grazed treatments lie intermediate between heavily grazed and ungrazed sites (Fig. 4). Semi-variance of heavy grazed sites were primarily lower for most vegetation characteristics, while ungrazed sites were typically higher than moderately grazed sites. However, these semi-variance relationships between heavy and ungrazed sites were inversed for cover of bare ground and angle of obstruction (Fig. 4). Nearly all ungrazed vegetation characteristics, except leaf litter, fit a spherical model that reached a sill at a lag distance [less than or equal to] 20 m and became spatially independent thereafter, while heavily and moderately grazed sites typically fit an exponential model that never reached a sill, indicating that ungrazed treatments were more heterogeneous than heavily or moderately grazed treatments at the pasture level.

DISCUSSION

RELATIONSHIPS BETWEEN VEGETATION AND TEMPERATURE

Significant variations in soil heat flux can occur in regions where vegetation cover is clumped with large areas of exposed soil (conditions that were common on our upland study sites) and temperature variability is highest during the mid-afternoon hours (Saunders et al., 1998; Kustas et al., 2000). Therefore, it was reasonable to expect landscape soil-surface temperatures to be highly variable across multiple spatial scales on our study site. Our study suggests that within a very small area (< 0.25-[m.sup.2]) temperatures could vary from biologically tolerable (typically < 40 C) to extremely hot and potentially lethal ([greater than or equal to] 50 C) as temperatures [greater than or equal to] 50 C are capable of destroying vital proteins that are essential for life support (Calder and King, 1974; Daubenmire, 1974; Larcher, 1991). Hence, thermal patterns of landscape variability may be a critical habitat feature capable of limiting the distribution and abundance of plant and animal species throughout the day, especially in hot arid environments. Further, it is likely that available habitat determined by the thermal environment will vary temporally throughout the day as the angle of sun rays change with respect to the landscape variables (i.e., vegetation structure, slope, aspects, etc. that intercept sun rays) and seasonally throughout the year. Thus, consideration of landscape-level thermal patterns may be a critical component to understanding ecological processes associated with biodiversity and conservation at multiple spatial and temporal scales.

On our study site, riparian drainages were typically covered (> 50%) with an overstory of deciduous trees. Because canopy shading from trees reduces the amount of solar radiation at the soil-surface (Larcher, 1991), we expected landscape position to have an influence on soil-surface temperatures. On our study site, topographic position may be the most important factor moderating mean soil-surface temperature. For instance, soil-surface temperatures in wooded riparian areas were on average 20 C below that of upland temperatures. In fact, 96% of all riparian soil-surface temperatures were [less than or equal to] 39 C, while 94% of upland soil-surface temperatures were [greater than or equal to] 40 C during the heat of the day. This temperature pattern between open and closed canopies also has been documented along forest edges (Didham and Lawton, 1999). Under a canopy of trees, the flux of thermal energy is greatly reduced, so that within the riparian zone soil-surface temperatures are less variable over space during extremely hot conditions (Larcher, 1991). As a result, the degree to which the riparian-upland edge influenced soil-surface temperatures indicates that the abiotic component may ultimately reduce habitat availability during the heat of the day. Because upland sites are primarily associated with thermally intolerable temperatures > 40 C during the heat of the day, riparian habitats, although a minor component of these landscapes (<5%), may serve as a thermal refuge for many animals during the heat of the day.

Vegetation cover of < 0.5 m tall is generally a poor predictor of soil-surface temperature; vegetation heights > 0.5 m are most important (Saunders et al., 1998). To document relationships between structural differences in vegetation and mean soil-surface temperature throughout the landscape, we chose only to analyze upland sites (vegetation cover typically < 0.5 m tall) because of the obvious overriding effect of canopy shading (i.e., shrubs and trees > 0.5 m tall) in the riparian areas. In general, vegetation characteristics on uplands were poor predictors of soil-surface temperature. Bare ground was the best predictive variable but accounted for only 12% of the variation in soil-surface temperature. On the uplands, grass cover and vegetation height were related inversely to soil-surface temperatures, while bare ground and leaf litter were associated positively with soil-surface temperature. The positive association between leaf litter and soil-surface temperatures may seem counterintuitive but likely is related to the proportion of cover that is transpiring plant material. For instance, in heavily grazed sites the interstitial spaces between clumps of vegetation cover is primarily composed of bare ground, which can increase temperatures. However, as grazing intensity decreases, bare soil is replaced progressively by leaf litter. So depending on grazing intensity, bare ground or litter is an indication of the absence of live transpiring plants. Because controlling the rate of transpiration through stomatal regulation is an efficient technique that acts to cool the plant and the surrounding environment (Daubenmire, 1974; Larcher, 1991), the temperature response to cover of leaf litter and bare ground was similar, presumably because the cover of transpiring plants (grasses, forbs and shrubs) generally did not differ between treatments (Table 1). Further, the semivariances of temperature (350[gamma]) and leaf litter (600[gamma]) within moderately grazed sites support our conclusions because they both correspond to a patch size of ca. 16 m, indicating that they may be spatially related.

INFLUENCE OF GRAZING ON HETEROGENEITY

Many studies have documented the effects of grazing intensity on vegetation structure (Dyksterhuis, 1949; Ellison, 1960; Fuhlendorf and Smeins, 1997), and variables associated with vegetation or ground cover can explain differences in soil-surface temperatures in some ecosystems (Saunders et al., 1998), so we predicted that vegetation patterns and mean soil-surface temperatures would be influenced by grazing intensity. Surprisingly, the variability in soil-surface temperature, on our study site, was similar across all grazing treatments, implying that grazing intensity had little influence on moderating soil-surface temperatures under these severe conditions.

Because animals selectively graze at multiple scales and because they do not graze uniformly (Senft et al., 1987), grazing is likely to have an influence on the spatial arrangement of vegetation (Fuhlendorf and Smeins, 1997, 1998). On our study site, grazing appeared to have a predictable influence on the pattern of nearly all vegetation variables throughout the landscape. For example, semi-variances of variables within moderatly grazed sites, with the exception of forb cover, consistently lie intermediate between heavily and ungrazed sites, suggesting that moderately grazed sites exhibit patterns associated with moderate levels of semi-variance or exhibit moderate levels of heterogeneity at the pasture level. Our results demonstrate that thermal heterogeneity was greatest within moderately grazed treatments, yet heterogeneity of vegetation (i.e., litter, grass, forb, shrub cover, and vegetation height) was greater in ungrazed treatments than in treatments that were heavily or moderately grazed. However, semi-variances of bare ground and angle of obstruction were greatest within heavily grazed and lowest in ungrazed sites, with a range of [less than or equal to] 15 m, suggesting that these patterns of variability changed ca. every 15 m across the landscape. The average range (lag distance) and semi-variance associated with all vegetation characteristics differed with grazing treatment. Most vegetation characteristics within moderate and heavy grazed sites typically fit an exponential model that never reached an obvious sill within the scale of our study (Fig. 3), indicating that patterns of heterogeneity are either not apparent or occur on those sites at larger spatial scales (i.e., [greater than or equal to] 50 m).

Grazing can increase, decrease or have no effect on spatial heterogeneity (McNanghton, 1984; Fuhlendorf and Smeins, 1999; Adler and Lauenroth, 2000). Our results clearly supported these conclusions and demonstrated that grazing can impose distinct patterns of heterogeneity on both biotic and abiotic ecosystem components, but patterns of heterogeneity were not consistent and depended on the variable of interest and spatial scale. For instance, variability of ungrazed sites were highly dependent on the variable of interest, where variability of biotic components (i.e., cover of leaf litter, forbs and shrubs) was highest across all spatial scales, but variability of bare ground and angles of obstruction was lowest at all scales of observation. In contrast, influences of grazing on thermal heterogeneity (in moderately grazed treatments) were highest at smaller (i.e., lag distances [less than or equal to] 20 m) and larger (i.e., lag distances [greater than or equal to] 48 m) spatial scales and were lowest at moderate scales (i.e., lag distances 22-45 m). Thus, our data leaves little doubt that grazing can reduce and/or increase the inherent variability in vegetation that is created by environmental variables (e.g., soil type, moisture, topography, etc.) and that the range of influence across ecological systems is highly scale dependent (Glenn et al., 1992; Adler and Lauenroth, 2000). These findings are consistent with previous reports from the shortgrass steppe (Milchunas et al., 1989; Milchunas and Lauenroth, 1989), tallgrass prairie (Glenn et al., 1992) and mixed grass savanna (Fuhlendorf and Smeins, 1998, 1999), where increases in grazing intensity were associated positively and/or negatively with heterogeneity.

CONCLUSION

Most previous studies of landscape heterogeneity have focused on variability of vegetation components with clear discontinuities that are often created by soil, topographic variation or disturbances associated with vegetation. But relationships between vegetation structure, topographic position, grazing and microclimate act in concert to shape grassland ecosystems and habitats. On our study site, effects of grazing on landscape heterogeneity were highly variable and depended primarily on the variable of interest and scale of observation. Further, presumably because soil-surface temperatures were highly variable within a relatively small area (i.e., [less than or equal to] 0.25-[m.sup.2]), grazing did not influence mean soil- surface temperature and results suggested that landscape position was the most important component moderating mean soil-surface temperatures. Lower soil-surface temperatures (associated with riparian areas) may provide a critical thermal refuge for many animals on hot summer days when air temperatures can exceed 37 C. Patterns of thermal heterogeneity were not directly related to any one vegetation variable, hence, landscape patterns based on vegetation parameters alone are limited in their use since patterns of thermal variability are likely influenced by the integration of vegetation and environmental variables. As a result, ecologists interested in the structure and function of ecosystem management should not discount the complex-feedback relationships between the biotic and abiotic components of landscape-level processes. For purposes of conservation and restoration, it is important to consider both biotic and abiotic ecosystem components.

SUBMITTED 9 FEBRUARY 2009

ACCEPTED 1 JUNE 2009

LITERATURE CITED

ADLER, P. B. AND W. K. LAUENROTH. 2000. Livestock exclusion increase the spatial heterogeneity of vegetation in the shortgrass steppe, Colorado. App. Veg. Sci., 3:213-222.

BOCK, C. E., J. H. BOCK, W. R. KENNEL AND V. M. HAWTHORNE. 1984. Responses of birds, rodents, and vegetation to livestock exclosure in a semidesert grassland site. J. Range Manage., 37:239-242.

CALDER, W. A. AND J. R. KING. 1974. Thermal and caloric relations of birds, p. 259-415. In: D. S. Farrier, J. R. King and K. C. Parkes (eds.). Avian biology: volume IV. Academic Press, New York, New York.

CHEN, J., J. F. FRANKLIN AND J. S. LOWE. 1996. Comparison of abiotic and structurally defined patch patterns in a hypothetical forest landscape. Conserv. Biol., 10:854-862.

CHRISTENSEN, N. L. 1997. Managing for heterogeneity and complexity on dynamic landscapes, p. 167-186. In: S. T. A. Pickett, R. S. Ostfeld, M. Shachak and G. E. Likens (eds.). The ecological basis for conservation: heterogeneity, ecosystems, and biodiversity. New York, New York.

CLARK, I. 1980. Part I: The semivariogram, p. 17-40. In: Geostatistics. McGraw-Hill, Inc., New York, New York.

CODY, R. P. AND J. K. SMITH. 1997. Applied statistics and the SAS programming language. Prentice Hall, Upper Saddle River, New Jersey.

DANIEL, T. W., J. A. HELMS AND F. S. BAKER. 1979. Principles of silviculture. McGraw-Hill Book Company, New York, New York.

DAUBENMIRE, R. F. 1959. A canopy-coverage method of vegetational analysis. Northwest Sci., 33:43-64.

--. 1974. Plant and environment: a textbook of autecology. John Wiley & Sons, New York, New York.

DENNIS, P., M. R. YOUNG AND I. J. GORDON. 1998. Distribution and abundance of small insects and arachnids in relation to structural heterogeneity of grazed, indigenous grasslands. Ecol. Ento., 3:253-264.

DIDHAM, R. K. AND J. H. LAWTON. 1999. Edge structure determines the magnitude of changes in microclimate and vegetation structure in tropical forest fragments. Biotropica, 31:17-30.

DYKSTERHUIS, E. J. 1949. Condition and management of rangeland based on quantitative ecology. J. Range Manage., 2:104-105.

ELLISON, L. 1960. Influence of grazing on plant succession of rangelands. Bot. Rev., 26:1-78.

FUHLENDORF, S. D. AND D. M. ENGLE. 2001. Restoring heterogeneity on rangelands: ecosystem management based on evolutionary grazing patterns. Bioscience, 51:625-632.

-- AND F. E. SMEINS. 1996. Spatial scale influence on longterm temporal patterns of a semi-arid grassland. Landsc. Ecol., 11:107-113.

-- AND --. 1997. Long-term vegetation dynamics mediated by herbivores, weather and fire in a Juniperus-Quercus savanna. J. Veg. Sci., 8:819-828.

-- AND --. 1998. Influence of soil depth on plant species response to grazing within a semi-arid savanna. Plant Ecol., 138:89-96.

-- AND --. 1999. Scaling effects of grazing in a semi-arid grassland. J. Veg. Sci., 10:731-738.

--, H. ZHANG, T. R. TUNNEL, D. M. ENGLE AND A. F. CROSS. 2002. Effects of grazing on restoration of southern mixed grass prairie soils. Restor. Ecol., 10:401-407.

GEIGER, R. 1965. The climate near the ground. Harvard University Press, Cambridge, Massachusetts.

GLENN, S. M., S. L. COLLINS AND D.J. GIBSON. 1992. Disturbances in tallgrass prairie: local and regional effects on community heterogeneity. Landsc. Ecol., 7:243-251.

HARRELL, W. C. AND S. D. FUHLENDORF. 2002. Evaluation of habitat structural measures in a shrubland community. J. Range Manage., 55:488-493.

HATCH, S. L., K. N. GANDHI AND L. E. BROWN. 1990. Checklist of the vascular plants of Texas. Texas Agricultural Experiment Station MP-1655, College Station, Texas.

HOLECHEK, J. L., R. P. PIEPER AND C. H. HERBEL. 1989. Range management: principles and practices. Prentice-Hall, Englewood Cliffs, New Jersey.

HOSMER, D. W., JR. AND S. LEMESHOW. 1989. Applied logistic regression. John Wiley and Sons, New York, New York.

IPCC. 2001. Climate change 2001: the scientific basis. Inter-government panel on climate change. Cambridge University Press, Cambridge, UK.

--. 2007. Climate change 2007. Climate change impacts, adaptation and vulnerability. Working Group II. Inter-government panel on climate change. Cambridge University Press, Cambridge, UK.

ISAAKS, E. H. AND R. M. SRIVASTAVA. 1989. Applied Geostatistics. Oxford University Press, Inc., New York, New York.

JONES, H. G. 1983. Plants and microclimate: a quantitative approach to environmental plant physiology. Cambridge University Press.

KARR, J. R. AND K. E. FREEMARK. 1983. Habitat selection and environmental gradients: dynamics in the "stable" tropics. Ecology, 64:1481-1494.

KAY, C. E. 1998. Are ecosystems structured from the top-down or bottom-up? A new look at an old debate. Wildl. Soc. Bull., 26:484-498.

KENDEIGH, S. C. AND B. J. FAWVER. 1981. Breeding bird populations in the Great Smokey Mountains, Tennessee and North Carolina. Wilson Bull., 93:218-242.

KOPP, S. D., F. S. GUTHERY, N. D. FORRESTER AND W. E. COHEN. 1998. Habitat selection modeling for northern bobwhite subtropical rangeland. J. Wildl. Manage., 62:884-895.

KUSTAS, W. P., J. H. PRUEGER, J. L. HATFIELD, K. RAMALINGAM AND L. E. HIPPS. 2000. Variability in soil heat flux from a mesquite dune site. Agri. Forest. Metero., 103:249-264.

LAL, R. 2008. Carbon sequestration. Phil. Trans. Royal Society Bull., 363:815-830.

--, R. F. FOLLETT AND J. M. KIMBLE. 2003. Achieving soil carbon sequestration in the United States: a challenge to the policy makers. Soil Sci., 168:827-825.

LARCHER, W. 1991. Physiological plant ecology, second edition. Springer-Verlag, Berlin, Germany.

LEVIN, S. A. 1992. The problem of pattern and scale in ecology. Ecology, 73:1943-1967.

LOIK, M. E., S. P. REDAR AND J. HARTE. 2000. Photosynthetic responses to a climate-warming manipulation for contrasting meadow species in the Rocky Mountains, Colorado, USA. Funct. Ecol., 14:166-175.

MAcARTHUR, R. H. AND E. R. PIANKA. 1966. On optimal use of a patchy environment. Am. Nat., 100:603-609.

McNAUGHTON, S. J. 1984. Grazing lawns: animals in herds, plant form, and coevolution. Am. Nat., 124:863-886.

MILCHUNAS, D. G. AND W. K. LAUENROTH. 1989. Three-dimensional distribution of plant biomass in relation to grazing and topography in the shortgrass steppe. Oikos, 55:82-86.

-- AND --. 1993. Quantitative effects of grazing on vegetation and soils over a global range of environments. Ecol. Mono., 63:327-366.

--, P. L. CHAPMAN AND M. K. KAZEMPOUR. 1989. Effects of grazing, topography, and precipitation on the structure of a semiarid grassland. Vegetatio, 80:11-23.

MOFFATT, H. H. AND A. J. CONRADII. 1979. Soil survey of Washita County Oklahoma. USDA Soil Conservation Service, Stillwater, Oklahoma.

OSTFELD, R. S. 1985. Limiting resources and territoriality in microtine rodents. Am. Nat., 126:1-15.

--, S. T. A. PICKETT, M. SHACHAK AND G. E. LIKENS. 1997. Defining scientific issues, p. 3-10. In: S. T. A. Pickett, R. S. Ostfeld, M. Shachak and G. E. Likens (eds.). The ecological basis for conservation: heterogeneity, ecosystems, and biodiversity. New York, New York.

PERRY, D. A. 1994. Forest ecosystems. Johns Hopkins University Press, Baltimore, Maryland.

PIERSON, F. B. AND J. R. WIGHT. 1991. Variability of near-surface soil temperature on sagebrush rangeland. J. Range Manage., 44:491-497.

RITCHIE, M. E. 2000. Nitrogen limitation and trophic vs. abiotic influences on insect herbivores in a temperate grassland. Ecology, 81:1601-1612.

ROADS, J. O., S. C. CHEN, A. K. GUETTER AND K. P. GEORGAKAKOS. 1994. Large-scale aspects of the United States hydrologic cycle. Bull. Am. Meteor. Soc., 75:1589-1610.

SAMS, M. S., R. L. LOCHMILLER, C. W. QUALLS, JR., D. M. LESLIE, JR. AND M. W. PAYTON. 1996. Physiological correlates of neonatal mortality in an overpopulated herd of white-tailed deer. J. Mamm., 77:179-190.

SANDERSON, J. AND L. D. HARMS. 2000. Landscape ecology: a top-down approach. Landscape Ecology series. CRC Press, Boca Raton, Florida.

SARNELLE, O., K. W. KRATZ AND S. D. COOPER. 1993. Effects of an invertebrate grazer on the spatial arrangement of a benthic microhabitat. Oecologia, 96:208-218.

SAS. 2001. Proprietary Software Release 6.08 ts407. SAS Institute Incorporated, Cary, North Carolina.

SAUNDERS, S. C., J. CHEN, T. R. CROW AND K. D. BROSOFSKE. 1998. Hierarchical relationships between landscape structure and temperature in a managed forest landscape. Landscape Ecology, 13:381-395.

SCHLEUCHER, E. 1999. Energetics and body temperature regulation in two convergent dove species from extreme habitats. Ornis Fenn., 76:199-210.

SEDGELEY, J. A. 2001. Quality of cavity microclimate as a factor influencing selection of maternity roosts by a tree-dwelling bat, Chalinolobus tuberculatus, in New Zealand. J. App. Ecol., 38:425-438.

SENFT, R. L., M. B. COUGHENOUR, D. W. BAILEY, L. B. RITTENHOUSE, O. E. SALA AND S. M. SWIFT. 1987. Large herbivores' foraging and ecological hierarchies. BioScience, 37:789-799.

SHARPE, P. B. AND B. VAN HORNE. 1999. Relationships between the thermal environment and activity of Piute ground squirrels (Spermophilus mollis). J. Thermal Biol., 24:265-278.

TRANGMAR, B. B., R. S. YOST AND G. UEHARA. 1985. Application of geostatistics to spatial studies of soil properties. Adv. Agron., 38:45-94.

TURNER, M. G. AND R. H. GARDNER. 1991. Quantitative methods in landscape ecology: an introduction, p. 1-14. In: M. G. Turner and R. H. Gardner (eds.). Quantitative methods in landscape ecology. Springer-Verlag, New York.

TURNER, S.J., R. V. O'NEILL, W. CONLEY, M. R. CONLEY AND H. C. HUMPHRIES. 1991. Pattern and scale: statistics for landscape ecology, p. 17-49. In: M. G. Turner and R. H. Gardner (eds.). Quantitative methods in landscape ecology. Springer-Verlag, New York.

WALSBERG, G. E. 1985. Physiological consequences of microhabitat selection, p. 289-314. In: M. L. Cody (ed.). Habitat selection in birds. Academic Press, New York, New York.

--. 1986. Thermal consequences of roost-site selection: The relative importance of three modes of heat conservation. Auk, 103:1-7.

--. 1993. Thermal consequences of diurnal microhabitat selection in a small bird. Ornis Scand., 24:174-182.

WEBB, D. R. AND C. M. ROGERS. 1988. Nocturnal energy expenditure of Dark-eyed Juncos roosting in Indiana during winter. Condor, 90:107-112.

WHITE, P. S. AND J. L. WALKER. 1997. Approximating nature's variation: selecting and using reference information in restoration ecology. Rest. Ecol., 5:338-349.

WIENS, J. A. 1997. The emerging role of patchiness in conservation biology, p. 93-107. In: S. T. A. Pickett, R. S. Osffeld, M. Shachak and G. E. Likens (eds.). The ecological basis for conservation: heterogeneity, ecosystems, and biodiversity. New York, New York.

--. 1976. Population responses to patchy environments. Ann. Rev. Ecol. Syst., 7:81-120.

DARRELL E. TOWNSEND II (1) AND SAMUEL D. FUHLENDORF

Department of Natural Resource Ecology and Management, Oklahoma State University, Stillwater 74078

(1) Corresponding author present address: Grand River Dam Authority, Office of Ecosystems Management, Administration Headquarters, 226 W. Dwain Willis Avenue, P.O. Box 409, Vinita, Oklahoma 74301; FAX: (918) 256-2333; e-mail: dtownsend@grda.com
TABLE 1.--A comparison of vegetation characteristics
(percent ground cover, vegetation height and angle of
obstruction) between grazing treatments on Marvin Klemme
Range Research Station, Bessie, Oklahoma, 2000, where n
denotes the number of measured data and SD denotes the
standard deviation

                       Heavy (a,b)               Moderate (c)
Vegetation
characteristic    n       Mean        SD     n     Mean      SD

Bare ground      398   36 ***, ***   32.4   400   23 ***    30.1
Leaf litter      398    2 ***, ***   10.4   400   10 ***    24.6
Grass            398   51 (ns), *    34.4   400   54 (ns)   34.4
Forb             398   24 (ns, ns)   22.8   400   25 (ns)   22.4
Shrub            398    3 (ns, ns)   14.5   400    3 (ns)   15.5
Vegetation       400   28 ***, ***   22.6   400   37 ***    31.9
  height (cm)
Angle of         400   46 ***, ***   37.3   400   53 ***    36.5
  obstruction
  [degrees]

                       Ungrazed
Vegetation
characteristic    n    Mean     SD

Bare ground      399      12   23.7
Leaf litter      399      19   31.5
Grass            399      57   38.5
Forb             399      27   25.9
Shrub            399       2   13.9
Vegetation       400      53   36.0
  height (cm)
Angle of         400      73   29.7
  obstruction
  [degrees]

Significance indicated by: ns = nonsignificant,
* P < 0.05, ** P < 0.01, *** P < 0.001

(a) ANOVA test for heavy vs. moderate treatments

(b) ANOVA test for heavy vs. ungrazed treatments

(c) ANOVA test for moderate vs. ungrazed treatments

TABLE 2.--Frequency and percent of soil-surface temperatures
collected along 200 m transects on Marvin Klemme Range
Research Station, Bessie, Oklahoma during summer 2000, where
n denotes the number of measured data

                        Upland        Riparian      Pooled

Temperature C          n      %      n      %      n     %

0-39                   66    6.3    148   95.5   214   17.8
40-59                 859   82.2      6    3.9   865   72.1
[greater than         120   11.5      1    0.7   121   10.1
  or equal to] 60

TABLE 3.--Vegetation characteristics (cover of plant
functional groups, angle of obstruction and vegetation
height) selected by a stepwise multiple regression analysis
to predict soil-surface temperatures on Marvin Klemme Range
Research Station, Bessie, Oklahoma, 2000

                                  Partial
Landscape                        regression            Partial
Position     Variable            Coefficient    SE    [R.sup.2]

Upland       Intercept              49.12      0.51
             Bare ground             0.10      0.01      0.12
             Litter                  0.13      0.01      0.09
             Vegetation height      -0.13      0.01      0.01

Riparian     Intercept              28.06      0.70
             Grass                   0.03      0.02      0.07
             Angle of                0.02      0.02      0.02
               obstruction

Landscape
Position     Variable               F        P

Upland       Intercept           9165.48   <0.001
             Bare ground          126.01   <0.001
             Litter               113.35   <0.001
             Vegetation height     12.27    0.001

Riparian     Intercept           1604.46   <0.001
             Grass                  3.79    0.054
             Angle of               2.43    0.121
               obstruction

TABLE 4.--Mean soil-surface temperature (C), standard
deviation (SD), and coefficient of variation (CV) for
different grazing intensities and topographic positions
(between quadrats and within quadrat), along 200 m transects
on Marvin Klemme Range Research Station, Bessie, Oklahoma,
summer 2000, where n denotes the number of measured data

Topographic                 Heavy
position
Scale                 n     Mean    SD     cv

Upland
  Within quadrat     1765    51    4.32    8.56
  Between quadrats    353    51    7.95   15.57

Riparian
  Within quadrat      235    29    1.23    4.12
  Between quadrats     47    29    3.91   13.27

Topographic                 Moderate
position
Scale                 n     Mean    SD     cv

Upland
  Within quadrat     1735    50    4.41    8.74
  Between quadrats    347    50    7.09   14.12

Riparian
  Within quadrat      265    29    2.21    7.52
  Between quadrats     53    29    8.59   30.02

Topographic                 Ungrazed
position
Scale                 n     Mean    SD     cv

Upland
  Within quadrat     1725    52    4.30    8.17
  Between quadrats    345    52    7.08   13.61

Riparian
  Within quadrat      275    31    1.42    4.61
  Between quadrats     55    31    6.62   21.28
COPYRIGHT 2010 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Townsend, Darrell E., II; Fuhlendorf, Samuel D.
Publication:The American Midland Naturalist
Article Type:Report
Geographic Code:1USA
Date:Apr 1, 2010
Words:6931
Previous Article:Life history and demographics of the endangered birdwing pearlymussel (Lemiox rimosus) (Bivalvia: Unionidae).
Next Article:Excursive behaviors by female white-tailed deer during estrus at two Mid-Atlantic sites.
Topics:

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