SEARCH

SEARCH BY CITATION

Keywords:

  • annual species;
  • exotic species;
  • perennial species;
  • plant composition;
  • plant species richness;
  • playa wetlands;
  • sediment;
  • southern high plains

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

1. Hydroperiod, wetland size and land use of watersheds surrounding wetlands have important individual influences on plant communities in wetlands. Our objectives were to determine the effect and relative importance of local and landscape factors on plant species richness, diversity and composition of different functional groups (i.e. total, wetland-dependent, perennial, annual and exotic species) in recently inundated playa wetlands.

2. We surveyed plant communities in 80 wet playas in the Southern High Plains, USA, and measured local factors: water depth, playa volume loss, sediment depth and playa area. We included landscape variables within 3 km: number of playas, edge density, percentage urban area and percentage Conservation Reserve Program area (CRP; conversion from highly erodible cropland to mostly introduced perennial grassland). We also recorded dominant land use as native grassland or cropland.

3. Water depth negatively influenced all plant community metrics (i.e. richness, diversity and cover) while playa volume loss (sediment eroded from watershed filling the basin) had a negative influence on total, wetland-dependent and perennial richness and cover. Playas with more cropland within their watersheds had greater annual and exotic richness and cover, suggesting that agricultural activities within playa watersheds have changed plant composition and facilitated biological invasion.

4. Playa area was less important in predicting plant community metrics in playas. Although not as dominant as local variables, edge density had a positive influence on species richness. Other landscape factors such as number of playas, percentage urban area and percentage CRP area were less important and consistent among different plant community metrics.

5.Synthesis and applications. Our results show that continued unsustainable sedimentation will result in loss of perennial species and promotion of exotic and annual species in playas. Watershed management to limit unsustainable sedimentation has the potential to maintain original playa plant communities dominated by perennial/native species and should also reduce the loss of playa functionality.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Wetland plant community composition can be influenced by factors such as hydrology, surrounding land use, propagule sources and dispersal capabilities (e.g. Casanova & Brock 2000; Smith & Haukos 2002). Wetland vegetation composition and cover responds to hydrology and concomitant changes through time (Euliss et al. 2004). In addition, the surrounding vegetation matrix may influence the plant community of a habitat patch (Wiser & Buxton 2008). For example, land use adjacent to wetlands (e.g. 250–300 m) influences plant diversity and species richness via the abundance and distribution of propagules and the route by which propagules disperse (Houlahan et al. 2006). Surrounding habitat patches with more edges (e.g. roads and ditches) may increase the likelihood of introducing invasive species by providing corridors (Wilcox 1989; Parendes & Jones 2000). However, wetland seed banks are important for species that do not disperse readily. At a landscape scale, wetland area, wetland isolation and surrounding land use may influence plant species composition via the rate at which propagules are generated (Matthews et al. 2009a) and dispersed (Boughton et al. 2010). Wetland area also has been shown to be somewhat predictive of wetland plant richness as larger wetlands may hold water longer (Smith & Haukos 2002). Determining the strength of these factors in dictating wetland plant communities is important in selecting potential future restoration endpoints. If vegetation (e.g. richness, composition) is to be used as a restoration endpoint, an understanding of how wetlands and their surrounding landscape attributes influence the plant community is necessary in decision making.

Loss of wetlands reduces landscape connectivity for wetland-dependent species relying on these habitats for reproduction, productivity and dispersal. This, in turn, can decrease landscape level biodiversity (Semlitsch & Bodie 1998; Gibbs 2000; Houlahan & Findlay 2003). In the conterminous United States, more than 53% of historical wetland area has been lost to filling, draining and modification of wetlands (Dahl 1990). Degradation has resulted further in functional and physical loss of wetland area and species (Holland et al. 1995; Davis & Froend 1999). Species will succeed or fail in impacted wetlands as dictated by local and landscape factors coupled with species dispersal and tolerance of environmental gradients (van der Valk 1981; Wright, Flecker & Jones 2003). Conservation and restoration efforts require that we understand how wetland degradation impacts biota and the function and longevity of wetlands (Kentula 2000).

Approximately 25 000–30 000 playa wetlands in the Southern High Plains (SHP, c. 82 000 km2), USA, serve as focal points of biodiversity for plants, invertebrates, birds and amphibians (Haukos & Smith 1994). Playas are shallow, circular, depressional wetlands that formed and are maintained through combined processes of wind, waves and dissolution (Smith 2003). Playas average 6·3 ha (Guthery & Bryant 1982) and comprise c. 2% of the SHP landscape (Haukos & Smith 1994). Native prairie surrounding playa wetlands has mostly been converted to row-crop agriculture. Over the past 80 years (Smith 2003), intensive cultivation of the SHP has resulted in significant erosion of playa watersheds and unsustainable sediment accumulation in playas. Through enrolment in the Conservation Reserve Program (CRP; the dominant conservation program in the SHP, designed in part to conserve highly erodible soils throughout the US), unsustainable sedimentation from cropland surrounding playas should be curtailed. Playas embedded in the cropland landscape have lost more than 100% of their hydric-soil defined volume (Luo et al. 1997; Tsai et al. 2007). Playas with more cropland within their watersheds have shorter hydroperiods and higher water loss rates (Tsai et al. 2007, 2010). Playas surrounded by cropland also have fewer perennials and more exotic plant species than playas embedded in native grassland (Smith & Haukos 2002). However, influences of hydrology and landscape variables on vegetation communities have not been studied.

Our objectives were to test the influence of local and landscape variables on plant species richness, diversity and composition to determine their relative influence on plant communities in wet playas. We included local factors such as water depth, playa area, sediment depth and percentage playa volume loss that were hypothesized to impact plant presence and cover. While other studies have simply discussed land use as an influencing factor on wetland plant communities, we investigated more specific landscape scale factors such as percentage urban area, percentage CRP area and tilled index (ratio of tilled to untilled land within each watershed) to determine the impact of various anthropogenic land disturbances. We also incorporated edge density (i.e. total length of edges per area) to represent potential exotic source locations or dispersal routes (e.g. ditches along roads) and number of playas as potential source populations. We hypothesized that land use and surrounding playas are the dominant landscape factors determining plant species richness, diversity and composition (e.g. Smith & Haukos 2002; Houlahan et al. 2006; Matthews et al. 2009a).

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Study area

We conducted this study in the Texas portion of the SHP (N 30°37·9′–N 35°44·2′, W 100° 2·1′–W 103° 6·4′; Fig. 1), which is the largest plateau in the USA (Sabin & Holliday 1995). Precipitation falls irregularly, annually averaging 450 mm in the northeast to 330 mm in the southwest portion of the SHP, mainly occurring in spring and early autumn (Bolen, Smith & Schramm 1989). Playas are isolated from groundwater flow, thus precipitation and runoff are their only sources of inundation (Wood & Osterkamp 1987). Because of the localized nature of precipitation, not all playas hold water in the same year. As evapotranspiration (2000–2500 mm year−1; Bolen, Smith & Schramm 1989) and infiltration (Zartman, Evans & Ramsey 1994) exceed precipitation and runoff, most playas dry annually (Smith 2003).

image

Figure 1.  Location of study playas (n = 80) in the Southern High Plains, USA, in 2003 and 2004, with county names listed.

Download figure to PowerPoint

Data collection

Forty playas containing surface water were selected in each year (80 total), after inundation precipitation events in late May and early June in 2003 and 2004. Selected playas were split evenly between cropland and grassland using the predominant surrounding land use for categorization. We determined vegetation composition in playa wetlands in early July, August and September in 2003 and 2004 (total of 240 surveys) using step-point sampling (Bonham 1989) along two parallel transects of approximately equal length. Each transect ran along a 45° angle from the southwest to the northeast edge of the playa. We also recorded water depth concurrent with vegetation surveys at three equidistant locations across the diameter of the playa basin.

We measured playa characteristics when playas were dry. We measured playa area by walking along the visual edge of each playa with a Global Positioning System unit (Applied Field Data Systems, Houston, TX, USA) and Fieldworker software (FieldWorker Products Ltd., Toronto, ON, Canada). The visual edge is distinguished by obvious changes of slope, soil colour and vegetation type (i.e. upland to wetland; Luo et al. 1997). We used a soil auger to determine sediment depth from the top of the sediment to the Randall clay (i.e. hydric soil; Allen et al. 1972) at 5–6 points in the playa basin. We obtained basin elevation with a level using the centre of each playa and each of eight equally spaced points along the visual edge. We determined playa soil edge (change from Randall clay to upland soil) by taking a series of sediment cores perpendicular to the visual edge. We then used mean sediment depth, playa area, location of soil edge, distance from the visual edge to the edge of playa basin and mean basin elevation to calculate sediment volume and original playa volume (Luo et al. 1997; Tsai et al. 2010). Finally, playa volume loss was calculated as sediment volume divided by original playa volume. Three assumptions were made for volumetric calculations: (i) Sediment was evenly distributed across the playa basin, (ii) Playa is elevated after sedimentation with the same shape and (iii) The shape of playa was a truncated cone (Tsai et al. 2010).

We obtained Digital Orthophoto Quarter Quadrangle aerial photos from the Texas Natural Resources Information System website (TNRIS 2006) and digitized a 3-km radius plot (i.e. 2827 ha) from the centre of each playa. We chose a 3-km radius to study the effect of landscape level variables as suggested by previous studies (e.g. Houlahan & Findlay 2004; Declerck et al. 2006). Land uses were classified as playa, grassland, cropland, CRP, urban and other (e.g. reservoir). We used farm folders from the Farm Service Agency of U. S. Department of Agriculture in Deaf Smith and Floyd counties to verify our land use data layers. We used FRAGSTATS*ARC® (The Sanborn Map Company Inc. Colorado Springs, CO, USA) to calculate landscape variables within the 3-km buffer, including number of playas, Shannon diversity index of land use, percentage urban and CRP area and edge density (McGarigal & Marks 1995). Shannon diversity index of land use is calculated based on the number and evenness of land use types, which uses the equation for Shannon index of diversity (Magurran 1988). Edge density is the total length of edges (e.g. roads, field edges) of all patches within a given area (m ha−1). We also obtained digitized U. S. Geological Survey contour maps (TNRIS 2006) and estimated the watershed of each study playa (Ekanayake et al. 2009). We followed Tsai et al. (2007) to calculate tilled index [amount of tilled (i.e. cropland and CRP) vs. untilled landscape (i.e. native grass) within the watershed]. The values for tilled index range from −1 (100% native grass watershed) to 1 (100% cropland/former cropland watershed).

Data analyses

Species richness included species encountered on both step-point transects, including points with no vegetation (i.e. bare soil or water). We calculated percentage composition (cover) by dividing number of points with vegetation by total points. We calculated Shannon index of diversity (hereafter diversity) using the number of points at which a plant species was encountered to represent the relative abundance of individuals (Magurran 1988). We categorized plant species into functional groups as follows: perennial or annual, native or exotic and wetland-dependent (i.e. facultative wetland and obligate wetland plants) or non-wetland (i.e. facultative, facultative upland and obligate upland plants) following the U.S. National Wetlands Inventory (1996). We calculated species richness and cover in functional groups.

We tested variance inflation factor for all variables to assess collinearity and exclude highly correlated variables (Kutner et al. 2004) to avoid decreasing statistical power and parameter accuracy of models (Graham 2003). After testing multicollinearity, we used the remaining eight variables (i.e. water depth, tilled index, playa volume loss, playa area, edge density, number of playas, percentage urban area and percentage CRP area) to build candidate model sets. Although sediment depth is an easier measurement to obtain than playa volume loss, we chose playa volume loss instead of sediment depth for models because playa volume loss is a standardized way to evaluate the influence of sedimentation on playa function. We constructed 46 a priori generalized models based on biological relevance and field observations using the eight variables to describe species richness, diversity and cover of different plant functional groups. To ensure a balanced model set, candidate model sets were built considering that each variable appeared approximately an equal number of times. We treated playa as the experimental unit and three vegetation surveys for each playa as repeated measures. We tested normality and homogeneity of variance and used Poisson distribution with log-link function for species richness and cover of different plant functional groups and normal distribution with identity-link function for plant diversity. We performed the analyses using generalized linear mixed models (lmer function in lmer4 package) in R (version 2.9.2; R Development Core Team 2009). Additionally, we used Student’s t-test to compare sediment depth and playa volume loss between playas in cropland (tilled index > 0)- and grassland (tilled index < 0)- dominated watersheds.

We selected models using corrected Akaike’s Information Criterion (AICc) (Burnham & Anderson 2002). We considered models with ΔAICc < 2 as best fit models for each response variable because they had support given the data. We used the concept of multimodel inference to calculate relative importance of each variable (Burnham & Anderson 2002:149). We also calculated the direction and effect size of variables in the models with ΔAICc value <2 to determine the relationship between response and explanatory variables.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

We recorded a total of 85 plant species during the survey. Mean tilled index was 0·51, indicating that watersheds of our study playas were dominated by cropland. Mean water depth was 32·3 cm, and maximum water depth was c. 150 cm (Table 1). Mean sediment depth for all playas was 27·1 cm. However, mean sediment depth for playas within grassland (9·5 cm, SE = 1·2) was smaller than cropland-dominated watersheds (44·8 cm, SE = 4·2; = 6·68, < 0·001). Mean playa volume loss for all playas was 109·7%, indicating that playas we sampled have lost more than their original hydric-soil defined volume. Mean playa volume loss for playas within grassland-dominated watersheds (32·1%, SE = 5·1%) was smaller than cropland-dominated watersheds (187·4%, SE = 30·7%; = 4·00, < 0·001). Playas are situated at the lowest point of closed watersheds hence impacted playas still occur in depressions that hold water even when the associated playas have lost their original volume.

Table 1.   Mean, SE, minimum and maximum of explanatory variables measured in playas (n = 80) in the Southern High Plains, USA, from June 2003 to May 2005, to evaluate their influence on plant communities
Explanatory variablesMeanSEMinimumMaximum
  1. *Tilled index = (tilled landscape − untilled landscape)/(tilled landscape + untilled landscape).

Water depth (WD, cm)32·33·70·0149·7
Tilled index (TI)*0·510·07−1·001·00
Playa volume loss (VL, %)109·717·71·0951·0
Sediment depth (SD, cm)27·12·90·9104·9
Area (AR, ha)11·10·91·447·2
Shannon diversity index of land use within 3 km [SHDI(3)]0·530·170·080·96
Edge density within 3 km [ED(3); m ha−1]24·20·99·355·7
Number of playas within 3 km [NP(3)]16·91·5067
Conservation Reserve Program area within 3 km [PR(3), %]23·01·70·061·4
Urban area within 3 km [PU(3), %]1·40·30·014·9

Species richness

Water depth and playa volume loss were the two most important variables with the greatest effect sizes in best models of total, wetland-dependent and perennial species richness (negative influence; Tables 2 and 3). Edge density also had some support given the data on total, wetland-dependent and perennial species richness in playas (positive influence; Table 3). The Akaike’s weights of the best models for total, wetland-dependent and perennial richness were 0·60, 0·63 and 0·78, respectively (Table 2).

Table 2.   Multiple regression models predicting plant species richness and diversity in playas (n = 80) in the Southern High Plains, USA, from June 2003 to May 2005, with number of parameters (k), Δ AICc and Akaike’s weights (wi). Only models with ΔAICc < 2 were presented for each response variable
Response variableModel*kΔAICcwi
  1. *WD, water depth; TI, tilled index; VL, playa volume loss; AR, playa area; ED(3), edge density; NP(3), number of playas; PR(3), Conservation Reserve Program area; PU(3), urban area; coefficients have been scaled.

  2. †Number of parameters including the intercept.

Total richness−0·32 (0·03) WD − 0·32 (0·05) VL + 0·11 (0·04) ED(3)40·000·60
Wetland-dependent richness−0·39 (0·04) WD − 0·34 (0·07) VL + 0·13 (0·06) ED(3)40·000·35
−0·40 (0·04) WD − 0·37 (0·08) VL + 0·13 (0·06) TI + 0·10 (0·07) ED(3)  + 0·10 (0·06) NP(3)  + 0·10 (0·05) PR(3)  + 0·06 (0·05) AR − 0·03 (0·06) PU(3)90·490·28
Perennial richness−0·49 (0·07) VL − 0·29 (0·04) WD + 0·12 (0·04) ED(3)40·000·78
Annual richness−0·35 (0·05) WD + 0·16 (0·07) TI30·000·18
−0·35 (0·05) WD + 0·17 (0·07) TI + 0·11 (0·07) ED(3)  + 0·10 (0·06) AR50·050·17
−0·35 (0·05) WD + 0·17 (0·07) TI + 0·09 (0·06) AR40·260·16
−0·35 (0·05) WD + 0·16 (0·07) TI + 0·09 (0·06) AR + 0·08 (0·06) PR(3)50·800·12
−0·36 (0·05) WD + 0·24 (0·07) TI − 0·17 (0·08) VL + 0·10 (0·06) PR(3)  + 0·10 (0·07) NP(3)  + 0·05 (0·06) AR + 0·04 (0·07) PU(3)  + 0·03 (0·08) ED(3)91·020·11
Exotic richness+ 0·38 (0·10) TI − 0·15 (0·08) WD30·000·29
+ 0·37 (0·10) TI21·370·15
Diversity−0·18 (0·04) VL − 0·18 (0·03) WD + 0·10 (0·04) PR(3)40·000·50
−0·21 (0·04) VL − 0·18 (0·03) WD + 0·09 (0·05) ED(3)  + 0·08 (0·04) TI + 0·08 (0·04) PR(3)  − 0·05 (0·05) NP(3)  + 0·04 (0·04) AR − 0·00 (0·05) PU(3)91·850·20
Table 3.   Relative importance of explanatory variables from multiple regression models to predict plant species richness, diversity and composition in playas (n = 80) in the Southern High Plains, USA, from June 2003 to May 2005
Explanatory variable*RichnessDiversityComposition
TotalWetland-dependentPerennialAnnualExoticTotalWetland-dependentPerennialAnnualExotic
  1. *WD, water depth; VL, playa volume loss; TI, tilled index; AR, playa area; ED(3), edge density; NP(3), number of playas; PR(3), Conservation Reserve Program are; PU(3), urban area.

WD1·0001·0001·0001·0000·4971·0001·0001·0001·0001·0001·000
VL1·0001·0001·0000·2230·0031·0001·0001·0001·0000·0290·100
TI0·1180·2770·0190·7280·9930·1970·0030·0040·0020·9200·581
AR0·2420·4410·1250·6020·2360·2770·5640·5230·3440·6800·338
NP(3)0·1430·3640·0420·1270·0860·2140·0830·1090·0610·0100·036
PR(3)0·2040·4030·0550·2820·1610·6940·0680·0630·0930·1200·319
ED(3)0·7180·6300·7990·3380·1900·3430·1850·2710·2960·2070·078
PU(3)0·1570·2960·0460·1390·0830·2230·2430·1420·1160·0160·034

For annual species richness, water depth had the greatest relative importance value and effect size (negative influence), followed by tilled index and playa area (positive influence; Tables 2 and 3). Number of playas, percentage urban area and percentage CRP area also appeared in the top models but those models had low weights and effect sizes, indicating less empirical support (Table 2). Akaike’s weight of best fit models for annual species was 0·74.

In the models of exotic species richness, tilled index had the strongest effect (positive influence; Tables 2 and 3). Water depth had the next strongest effect (negative influence). Playa area, playa volume loss and all the landscape variables were the least important (Table 3). Akaike’s weight of best fit models for exotic species was 0·44.

Total plant diversity

In models of diversity, water depth and playa volume loss had the strongest effect (negative influence; Tables 2 and 3). Percentage CRP area was the third most important variable (positive influence). Tilled index, playa area, number of playas, edge density and percentage urban area had less support given the data (Table 3). The combined Akaike’s weight of best fit models for total plant diversity was 0·70.

Plant cover

Water depth and playa volume loss had the strongest effect in best models of total, wetland-dependent and perennial plant cover (negative influence; Tables 3 and 4). Playa area (negative influence) and edge density (positive influence) also appeared in the best models of total, wetland-dependent and perennial cover. However, the effect sizes of playa area and edge density were small compared with water depth and playa volume loss. Akaike’s weight of best fit models for total and perennial cover was essentially the same at 0·85 and 0·84, respectively (Table 4), while Akaike’s weight of best fit models for wetland-dependent species was 0·94.

Table 4.   Multiple regression models predicting percentage composition (cover) of plant species in playas (n = 80) in the Southern High Plains, USA, from June 2003 to May 2005, with number of parameters (k), ΔAICc and Akaike’s weights (wi). Only models with Δ AICc < 2 were presented for each response variable
Response variableModel*kΔAICcwi
  1. *WD, water depth; TI, tilled index; VL, playa volume loss; AR, playa area; ED(3), edge density; NP(3), number of playas; PR(3), Conservation Reserve Program area and PU(3), urban area; coefficients have been scaled

  2. †Number of parameters including the intercept.

Total composition−0·73 (0·02) WD − 0·52 (0·10) VL − 0·16 (0·10) AR40·000·24
−0·73 (0·02) WD − 0·51 (0·10) VL − 0·17 (0·10) AR + 0·14 (0·09) PU(3)50·000·24
−0·73 (0·02) WD − 0·49 (0·10) VL30·490·19
−0·73 (0·02) WD − 0·48 (0·10) VL + 0·14 (0·09) ED(3)40·550·18
Wetland-dependent composition−0·72 (0·02) WD − 0·70 (0·14) VL − 0·24 (0·13) AR40·000·28
−0·72 (0·02) WD − 0·64 (0·13) VL + 0·23 (0·12) ED(3)40·070·27
−0·72 (0·02) WD − 0·64 (0·14) VL31·220·15
−0·72 (0·02) WD − 0·70 (0·14) VL − 0·24 (0·13) AR + 0·11 (0·12) PU(3)51·400·14
−0·72 (0·02) WD − 0·69 (0·14) VL − 0·25 (0·13) AR + 0·07 (0·13) NP(3)51·950·10
Perennial composition−1·04 (0·18) VL − 0·68 (0·03) WD + 0·19 (0·13) ED(3)40·000·29
−1·03 (0·17) VL − 0·68 (0·03) WD30·160·27
−1·08 (0·18) VL − 0·68 (0·03) WD − 0·15 (0·13) AR41·100·17
−1·07 (0·18) VL − 0·68 (0·03) WD − 0·16 (0·13) AR + 0·15 (0·12) PU(3)51·900·11
Annual composition−0·83 (0·04) WD + 0·48 (0·17) TI − 0·28 (0·17) AR40·000·34
−0·84 (0·04) WD + 0·50 (0·17) TI30·430·27
−0·83 (0·04) WD + 0·48 (0·17) TI − 0·27 (0·17) AR + 0·18 (0·17) ED(3)51·130·19
Exotic composition−0·94 (0·10) WD + 0·56 (0·26) TI30·000·29
−0·94 (0·10) WD + 0·57 (0·26) TI − 0·45 (0·25) PR(3)  + 0·09 (0·25) AR51·200·16
−0·94 (0·10) WD − 0·43 (0·26) PR(3)31·800·12

Water depth was the most important variable predicting annual cover (negative influence) with the greatest effect sizes (Tables 3 and 4). Tilled index (positive influence) and playa area (negative influence) also had high relative importance values. Akaike’s weight of best fit models for annual cover was 0·80.

Similar to annual cover, water depth (negative influence) was the most important variable in best models of exotic species, followed by tilled index (positive influence) (Tables 3 and 4). Percentage CRP area (negative influence) and playa area (positive influence) also appeared in the best fit model but effect sizes were small. The combined Akaike’s weight for the best fit models was 0·57.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Water depth

As expected, water depth was the most important variable predicting plant communities after inundation in wet playas for nearly all response variables. Water depth is one of the most dominant factors influencing plant community in wetlands (Casanova & Brock 2000). The prevailing negative relationships we found between water depth and plant variables showed that when water was too deep, it was a limiting factor for plants. Most emergent playa plant species require shallow or no standing water to germinate (Haukos & Smith 1993b). We only studied wet playas after inundation in summer so water depth can only be interpreted as a seasonal predictor in our case. Nevertheless, controlling water levels is an important tool when implementing specific management plans for individual wetlands (e.g. moist-soil management; Haukos & Smith 1993a).

Conversion of prairie to cropland

The amount of watershed disturbed by cultivation (tilled index) showed strong positive effects in models of annual and exotic richness and cover. These results corroborated the earlier findings in playas by Smith & Haukos (2002). Farming of the watershed may increase opportunities for biological invasion, particularly exotics (Matthews et al. 2009b) through increased nutrient input (Houlahan & Findlay 2004; Tyler, Lambrinos & Grosholz 2007) and exposed soil (Burke & Grime 1996). Assuming that edges are sources or dispersal routes of annuals, the positive influence of playa area and edge density on annual (tendency to be cropland-associated) richness provides further support. Erosion of upland areas during heavy rainfall events may be one mechanism for seeds of annuals and exotics to be transported into playas. However, we did not measure the mechanism of biological invasion or determine seed bank composition. More information is needed to be able to identify specific life-history and dispersal traits for individual species in playas.

Sedimentation resulting from cultivation

For total, wetland-dependent and perennial (generally wetland-associated) species richness and cover, playa volume loss has a similar negative effect as water depth. In the SHP, erosion occurs naturally, but historically, rates of sedimentation were sufficiently low to allow natural formation and deepening processes in playas (e.g. dissolution) to maintain playa shape and function (Luo et al. 1997). However, intensive agricultural activities have caused higher sedimentation rates since the 1930s, and sedimentation rates in playas with cropland watersheds (4·8–9·7 mm year−1) are much higher compared with playas with grassland watersheds (0·67–0·85 mm year−1) (Luo et al. 1997). Laboratory experiments have shown that even a minimal amount of sediment (i.e. 5 mm) burying the seed bank can reduce seedling emergence and species richness (Jurik, Wang & van der Valk 1994; Gleason et al. 2003). Moreover, playas which have lost their original volume also tend to have shorter hydroperiods (Tsai et al. 2007, 2010). This may restrict plant species that require more time to establish. In a 50-year simulation model, Smith et al. (2011) estimated that playas with cropland watersheds may have 50% shorter hydroperiods compared with playas with grassland watersheds because of sedimentation. If sedimentation in playas is sufficient to reduce seedling emergence and species richness, current unsustainable rates of sedimentation in the SHP could adversely affect biodiversity, especially in playas within cropland watersheds.

Other factors

Unlike other studies of species-area relationships in wetlands (e.g. Matthews et al. 2005; Houlahan et al. 2006), area was not a dominant factor for total species richness in our study and only had low relative importance values. Smith & Haukos (2002) suggested that area could be a valuable factor when considering regional floral diversity because playas are the only habitats for wetland-dependent species to survive in the semi-arid SHP. Playas only have two habitats, basin floor and edge (Luo et al. 1997); therefore, the weak species-area relationship in playas found by Smith & Haukos (2002) suggests that habitat diversity is likely to be more important in influencing richness than population size. While larger playas included more sampling points, more individuals in larger playas did not contribute much to total species richness.

Landscape factors can strongly influence wetland plant communities at a local scale (Declerck et al. 2006; Matthews et al. 2009a), and we have shown how land use surrounding playas can influence playa plant communities. While most landscape variables within a 3-km radius appeared in best models for all response variables, they had relatively less influence on plant communities than local factors (smaller relative importance values and effect sizes). Among these, edge density had a positive influence on total, wetland-dependent and perennial richness while percentage CRP area had a positive influence on diversity. Both edge density and percentage CRP area reflected human-related disturbance in the landscape with the potential for introducing non-native species to wetlands. Moreover, previous studies have suggested that waterbirds (Figuerola & Green 2002; Mueller & van der Valk 2002) may play important roles in seed dispersal among wetlands via external and internal transport. However, our results showed that the number of playas, an indirect measure of source populations, had little influence on plant richness and cover. Although we do not understand the mechanism by which these landscape factors influence overall richness and different species groups, our results suggest that it may be beneficial to incorporate variables beyond wetland boundaries when managing vegetation.

Disturbance within the wetland

Disturbance within wetlands such as tilling and grazing may also directly influence plant communities. For example, five of our study playas were tilled prior to initial inundation. Plant richness and cover was low in these playas (Tsai 2007). Tilling a playa removes all non-crop vegetation within the basin, resulting in low non-crop plant cover (generally weedy annuals). Tilling through the basin may also change the physical appearance of a playa and affect its ability to maintain a natural hydroperiod (Tsai et al. 2007).

Cattle grazing in wetlands may influence plant communities through trampling (Declerck et al. 2006), increased nutrient input (Proulx & Mazumder 1998) and direct consumption (Lee Foote & Rice Hornung 2005). However, the effects of grazing on plant communities vary based on cattle management strategies (Boughton et al. 2011), nutrient levels in the ecosystem (Proulx & Mazumder 1998) and traits of dominant plant species (Bullock et al. 2001). Although cattle were present in many of our study playas (cropland and grassland dominated watersheds combined; Tsai 2007), their cultural management was extremely variable making replication of grazing treatments to examine cattle effects impossible in this study. Therefore, future studies are needed to evaluate the effect of cattle on playa plant communities in a controlled setting. Additionally, while cattle may facilitate seed dispersal among wetlands (Cosyns et al. 2005), most livestock were fenced in around one playa, limiting the likelihood of seed dispersal.

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Playas are vulnerable wetland patches influenced by the surrounding landscape in the SHP. We demonstrated that in addition to water depth, overall playa plant species richness was negatively influenced by watershed disturbance (i.e. tilled index and playa volume loss) through unsustainable sedimentation. Cultivation practices also increased the opportunity for exotic and annual species to colonize playas. Watershed management is crucial to the maintenance of original plant communities dominated by perennial/native species (although some native annuals species are desirable for wildlife; e.g. Smith, Haukos & Prather 2004). Managers should consider landscape factors when implementing management strategies to maintain playa plant communities. Wetland management cannot be considered in a vacuum and must take system restoration into context. The primary native shortgrass species such as buffalo grass Bouteloua dactyloides, blue grama Bouteloua gracilis and sideoats grama Bouteloua curtipendula can be planted immediately adjacent to the playa basin. We recommended a minimum of 100-m buffer to minimize sediment deposition and to maintain hydrological function, which is the key to sustainability of ecosystem services (Smith et al. 2011). There are federal agricultural programmes available to restore watersheds and wetland basins. However, watershed programmes (i.e. CRP) have not often used native plant species and wetland basin programmes (i.e. Wetland Reserve Program) have seldom been applied (Smith et al. 2011). Modifying plantings by requiring native grasses and increasing use of these federal programmes will improve function of the entire wetland and upland ecosystem.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Funding for this project was provided by the U.S. Environmental Protection Agency (no. R–82964101–0). L.M. Smith was supported by the Caesar Kleberg Foundation for Wildlife Conservation. The Natural Resources Conservation Service, U.S. Department of Agriculture, assisted in locating landowners of study playas. We thank the numerous private landowners that allowed us access to their playas. Technicians and friends assisted with data collection and GIS. We thank D.A. Haukos, N. Jayasena and J. O’Connell for providing helpful comments.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  • Allen, B.L., Harris, B.L., Davis, K.R. & Miller, G.B. (1972) The Mineralogy and Chemistry of High Plains Playa Lake Soils and Sediments. Report 14-01-0001-1551. Texas Tech University, Lubbock.
  • Bolen, E.G., Smith, L.M. & Schramm, H.L. Jr (1989) Playa lakes: prairie wetlands of the Southern High Plains. BioScience, 39, 615623.
  • Bonham, C.D. (1989) Measurements for Terrestrial Vegetation. John Wiley & Sons, Inc, New York.
  • Boughton, E.H., Quintana-Ascencio, P.F., Bohlen, P.J., Jenkins, D.G. & Pickert, R. (2010) Land-use and isolation interact to affect wetland plant assemblages. Ecography, 33, 461470.
  • Boughton, E.H., Quintana Ascencio, P.F., Nickerson, D. & Bohlen, P.J. (2011) Management intensity affects the relationship between non native and native species in subtropical wetlands. Applied Vegetation Science, 14, 210220.
  • Bullock, J.M., Franklin, J., Stevenson, M.J., Silvertown, J., Coulson, S.J., Gregory, S.J. & Tofts, R. (2001) A plant trait analysis of responses to grazing in a long-term experiment. Journal of Applied Ecology, 38, 253267.
  • Burke, M.J.W. & Grime, J.P. (1996) An experimental study of plant community invasibility. Ecology, 77, 776790.
  • Burnham, K.P. & Anderson, D.R. (2002) Model Selection and Multimodel Inference: a Practical Information-Theoretic Approach, 2nd edn. Springer-Verlag, New York.
  • Casanova, M.T. & Brock, M.A. (2000) How do depth, duration and frequency of flooding influence the establishment of wetland plant communities? Plant Ecology, 147, 237250.
  • Cosyns, E., Claerbout, S., Lamoot, I. & Hoffmann, M. (2005) Endozoochorous seed dispersal by cattle and horse in a spatially heterogeneous landscape. Plant Ecology, 178, 149162.
  • Dahl, T.E. (1990) Wetlands Losses in the United States 1780’s to 1980’s. U.S. Department of the Interior, Fish and Wildlife Service, Washington, D.C.
  • Davis, J.A. & Froend, R. (1999) Loss and degradation of wetlands in southwestern Australia: underlying causes, consequences and solutions. Wetlands Ecology and Management, 7, 1323.
  • Declerck, S., De Bie, T., Ercken, D., Hampel, H., Schrijvers, S., Van Wichelen, J., Gillard, V., Mandiki, R., Losson, B. & Bauwens, D. (2006) Ecological characteristics of small farmland ponds: associations with land use practices at multiple spatial scales. Biological Conservation, 131, 523532.
  • Ekanayake, A.J., Tsai, J.-S., Allen, L.J.S., Smith, L.M., Surles, J.G. & Allen, E.J. (2009) Estimating watershed area for playas in the Southern High Plains, USA. Wetlands, 29, 387395.
  • Euliss, N.H. Jr, LaBaugh, J.W., Fredrickson, L.H., Mushet, D.M., Laubhan, M.K., Swanson, G.A., Winter, T.C., Rosenberry, D.O. & Nelson, R.D. (2004) The wetland continuum: a conceptual framework for interpreting biological studies. Wetlands, 24, 448458.
  • Figuerola, J. & Green, A.J. (2002) Dispersal of aquatic organisms by waterbirds: a review of past research and priorities for future studies. Freshwater Biology, 47, 483494.
  • Gibbs, J.P. (2000) Wetland loss and biodiversity conservation. Conservation Biology, 14, 314317.
  • Gleason, R.A., Euliss, N.H. Jr, Hubbard, D.E. & Duffy, W.G. (2003) Effects of sediment load on emergence of aquatic invertebrates and plants from wetland soil egg and seed banks. Wetlands, 23, 2634.
  • Graham, M.H. (2003) Confronting multicollinearity in ecological multiple regression. Ecology, 84, 28092815.
  • Guthery, F.S. & Bryant, F.C. (1982) Status of playas in the Southern Great Plains. Wildlife Society Bulletin, 10, 309317.
  • Haukos, D.A. & Smith, L.M. (1993a) Moist-soil management of playa lakes for migrating and wintering ducks. Wildlife Society Bulletin, 21, 288298.
  • Haukos, D.A. & Smith, L.M. (1993b) Seed-bank composition and predictive ability of field vegetation in playa lakes. Wetlands, 13, 3240.
  • Haukos, D.A. & Smith, L.M. (1994) The importance of playa wetlands to biodiversity of Southern High Plains. Landscape and Urban Planning, 28, 8398.
  • Holland, C.C., Honea, J.E., Gwin, S.E. & Kentula, M.E. (1995) Wetland degradation and loss in the rapidly urbanizing area of Portland, Oregon. Wetlands, 15, 336345.
  • Houlahan, J.E. & Findlay, C.S. (2003) The effects of adjacent land use on wetland amphibian species richness and community composition. Canadian Journal of Fisheries and Aquatic Sciences, 60, 10781094.
  • Houlahan, J.E. & Findlay, C.S. (2004) Estimating the ‘critical’ distance at which adjacent land-use degrades wetland water and sediment quality. Landscape Ecology, 19, 677690.
  • Houlahan, J.E., Keddy, P.A., Makkay, K. & Findlay, C.S. (2006) The effects of adjacent land use on wetland species richness and community composition. Wetlands, 26, 7996.
  • Jurik, T.W., Wang, S.C. & van der Valk, A.G. (1994) Effects of sediment load on seedling emergence from wetland seed banks. Wetlands, 14, 159165.
  • Kentula, M.E. (2000) Perspectives on setting success criteria for wetland restoration. Ecological Engineering, 15, 199209.
  • Kutner, M.H., Neter, J., Nachtsheim, C.J. & Li, W. (2004) Applied Linear Statistical Models, 5th edn. McGraw-Hill Companies, Inc, New York.
  • Lee Foote, A. & Rice Hornung, C.L. (2005) Odonates as biological indicators of grazing effects on Canadian prairie wetlands. Ecological Entomology, 30, 273283.
  • Luo, H.-R., Smith, L.M., Allen, B.L. & Haukos, D.A. (1997) Effects of sedimentation on playa wetland volume. Ecological Applications, 7, 247252.
  • Magurran, A.E. (1988) Ecological Diversity and its Measurement. Princeton University Press, Princeton.
  • Matthews, J.W., Tessene, P.A., Wiesbrook, S.M. & Zercher, B.W. (2005) Effect of area and isolation on species richness and indices of floristic quality in Illinois, USA wetlands. Wetlands, 25, 607615.
  • Matthews, J.W., Peralta, A.L., Flanagan, D.N., Baldwin, P.M., Soni, A., Kent, A.D. & Endress, A.G. (2009a) Relative influence of landscape vs. local factors on plant community assembly in restored wetlands. Ecological Applications, 19, 21082123.
  • Matthews, J.W., Peralta, A.L., Soni, A., Baldwin, P., Kent, A.D. & Endress, A.G. (2009b) Local and landscape correlates of non-native species invasion in restored wetlands. Ecography, 32, 10311039.
  • McGarigal, K. & Marks, B.J. (1995) . FRAGSTATS: Spatial Pattern Analysis Program for Quantifying Landscape Structure. PNW-GTR-351. U. S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, Portland.
  • Mueller, M. & van der Valk, A. (2002) The potential role of ducks in wetland seed dispersal. Wetlands, 22, 170178.
  • Parendes, L.A. & Jones, J.A. (2000) Role of light availability and dispersal in exotic plant invasion along roads and streams in the HJ Andrews Experimental Forest, Oregon. Conservation Biology, 14, 6475.
  • Proulx, M. & Mazumder, A. (1998) Reversal of grazing impact on plant species richness in nutrient-poor vs. nutrient-rich ecosystems. Ecology, 79, 25812592.
  • R Development Core Team (2009) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3-900051-07-0, Available at: http://www.R-project.org (accessed April 2010) ).
  • Sabin, T.J. & Holliday, V.T. (1995) Playas and lunettes on the Southern High Plains: morphometric and spatial relationships. Annals of the Association of American Geographers, 85, 286305.
  • Semlitsch, R.D. & Bodie, J.R. (1998) Are small, isolated wetlands expendable? Conservation Biology, 12, 11291133.
  • Smith, L.M. (2003) Playas of the Great Plains. University of Texas Press, Austin.
  • Smith, L.M. & Haukos, D.A. (2002) Floral diversity in relation to playa wetland area and watershed disturbance. Conservation Biology, 16, 964974.
  • Smith, L.M., Haukos, D.A. & Prather, R.M. (2004) Avian response to vegetative pattern in playa wetlands during winter. Wildlife Society Bulletin, 32, 474480.
  • Smith, L.M., Haukos, D.A., McMurry, S.T., LaGrange, T. & Willis, D. (2011) Ecosystem services provided by playas in the High Plains: potential influences of USDA conservation programs. Ecological Applications, 21, S82S92.
  • Texas Natural Resources Information System (TNRIS) (2006) TNRIS data distribution system. Available at: http://www.tnris.state.tx.us/digital.htm (accessed September 2006).
  • Tsai, J.-S. (2007) Local and Landscape Factor Influences on Avian Community Composition in Playas of the Southern High Plains. PhD Thesis. Texas Tech University, Lubbock.
  • Tsai, J.-S., Venne, L.S., McMurry, S.T. & Smith, L.M. (2007) Influences of land use and wetland characteristics on water loss rates and hydroperiods of playas in the Southern High Plains, USA. Wetlands, 27, 683692.
  • Tsai, J.-S., Venne, L.S., McMurry, S.T. & Smith, L.M. (2010) Vegetation and land use impact on water loss rate in playas of the Southern High Plains, USA. Wetlands, 30, 11071116.
  • Tyler, A.C., Lambrinos, J.G. & Grosholz, E.D. (2007) Nitrogen inputs promote the spread of an invasive marsh grass. Ecological Applications, 17, 18861898.
  • U.S. National Wetlands Inventory (1996) National List of Plant Species that Occur in Wetlands. U.S. Fish and Wildlife Service, St. Petersburg.
  • van der Valk, A.G. (1981) Succession in wetlands: a gleasonian approach. Ecology, 62, 688696.
  • Wilcox, D.A. (1989) Migration and control of purple loosestrife (Lythrum salicaria L.) along highway corridors. Environmental Management, 13, 365370.
  • Wiser, S.K. & Buxton, R.P. (2008) Context matters: matrix vegetation influences native and exotic species composition on habitat islands. Ecology, 89, 380391.
  • Wood, W.W. & Osterkamp, W.R. (1987) Playa-lake basins on the Southern High Plains of Texas and New Mexico: part II. A hydrologic model and mass-balance arguments for their development. Geological Society of America Bulletin, 99, 224230.
  • Wright, J.P., Flecker, A.S. & Jones, C.G. (2003) Local vs. landscape controls on plant species richness in beaver meadows. Ecology, 84, 31623173.
  • Zartman, R.E., Evans, P.W. & Ramsey, R.H. (1994) Playa lakes on the Southern High Plains in Texas: reevaluating infiltration. Journal of Soil and Water Conservation, 49, 299301.