Influence of Agriculture on Aquatic Invertebrate Communities of Temporary Wetlands in the Prairie Pothole Region of North Dakota, USA
Results
We identified resting eggs and remains of 6 invertebrate taxa in our hand-sorted soil samples (i.e., cladoceran ephippia, planorbid snail shells, lymnaeid snail shells, physid snail shells, ostracod shells, and trichopteran cases). Number of invertebrate taxa in soil samples differed (F1,18 = 14.00, p = 0.0015) between grassland (
= 2.947, SE = 0.388) and cropland ( = 0.895, SE = 0.388) wetlands. We found more cladoceran ephippia in soil samples from grassland wetlands than in samples collected from cropland wetlands (F1,18 = 27.45, p < 0.0001) (Figure 2). In fact, ephippia were entirely absent from the soils of 14 of the 19 cropland wetlands. All 19 grassland wetlands except one contained cladoceran ephippia. Both planorbid snail shells and physid snail shells occurred in greater numbers (F1,18 = 13.31, p = 0.0018 and F1,18 = 6.93, p = 0.0169, respectively) in soil samples from grassland wetlands than in those from cropland wetlands (Figure 3). We also found greater numbers of ostracod shells in samples from grassland versus cropland wetlands (F1,18 = 4.71, p = 0.0436). Although there appeared to be more lymnaeid snail shells in samples collected from wetlands in grassland, we were unable to separate the two condition classes (F1,18 = 1.23, p = 0.2823). Number of trichopteran cases also did not differ (F1,18 = 0.01, p = 0.9051) among the condition classes, but these occurred only rarely in our study wetlands (Figure 3).
| Figure 2. Back-trensformed mean abundances and 95% confidence intervals of Cladocera resting eggs (ephippia) and individuals found in and incubated out of soil samples collected from wetlands within grassland and cropland watersheds. All differences observed were significant (p < 0.05). |
| Figure 3. Back-transformed mean abundances and 95% confidence intervals of planorbid snail, lymnaeid snail, physid snail, an ostracod shells and trichopteran cases found in soil samples from wetlands within grassland and cropland watersheds. An * indicates that differences were significant (p < 0.05). |
Cladocerans, anostracans, and ostracods were the only invertebrates successfully incubated in our aquaria. However, we found more invertebrate taxa in our incubated soil samples from grassland wetlands (F1,18 = 19.00, p = 0.0004) than in soil samples from cropland wetlands ( = 1.632, SE = 0.162 versus
= 0.632, SE = 0.162). Number of cladocerans incubated from soil samples also reflected differences we observed in hand-sorted samples (Figure 2) with greater numbers incubated from the grassland samples (F1,18 = 28.67, p < 0.0001). Number of incubated anostracans (Figure 4) did not differ between wetland condition classes (F1,18 = 1.21, p = 0.2855); however like trichopteran cases in our hand-sorting analysis, anostracans were rare and were incubated from only 4 wetland samples. Reflecting the trend observed in our hand-sorting data, incubated ostracod numbers differed by condition class (F1,18 = 4.69, p = 0.0440), with greater numbers being incubated from the grassland samples (Figure 4).
| Figure 4. Back-transformed mean abundances and 95% confidence intervals of fairy shrimp (Anostraca) and seed shrimp (Ostracoda) incubated from soil samples from wetlands within grassland and cropland watersheds. An * indicates that differences were significant (p < 0.05). |
We were unable to detect any differences in viability of cladoceran ephippia (F1,4 = 0.23, p = 0.6578) among grassland (back-transformed mean = 0.380) and cropland (back-transformed mean = 0.498) wetlands; however, our sample was restricted to only those 5 wetland pairs that contained ephippia in both condition categories. Extreme variability and wide confidence intervals (-0.023 to 0.905 and -0.012 to 1.189, respectively) could have easily obscured any differences that may have existed.
Previous Section -- Methods
Return to Contents
Next Section -- Discussion
