Occupancy of aquatic macroinvertebrates in seasonal wetlands

Background/Question/Methods

The Clean Water Act requires the maintenance and restoration of biological integrity in the nation’s waters. Seasonal wetlands receive little attention by policy makers because of their size and semi-permanent hydrology. The ecological importance of wetlands spans across trophic levels and contributes to improved water quality through nutrient cycling and groundwater infiltration. Aquatic macroinvertebrates are highly diverse, have varying sensitivity to habitat and water characteristics, and are an important food source for many fish, bird, and amphibian species. While a great deal of attention has been paid to macroinvertebrates in stream systems, there is a lack of information regarding these organisms in lentic systems; however, there have been attempts to assess wetland health based, in part, on macroinvertebrate presence. In order to create a method to rapidly assess wetland health on public lands in Illinois, we sampled 78 ponds for macroinvertebrates using an occupancy and detectability framework. We conducted repeated samples over four rounds, each containing three sampling events. Detection and occupancy probabilities were estimated in Program PRESENCE 5.3 and an information theoretic approach was used to examine the effects of seasonality and site specific habitat variables on occupancy and detectability.

Results/Conclusions

The detection probabilities of macroinvertebrate families ranged from 0.0994 to 0.7144.  Whereas some macroinvertebrate families exhibited constant detection, the detection probability of others was survey specific. Both temperature and precipitation had constant and seasonal effects on detectability. Generally, detectability was low in the early spring and summer and high in mid to late spring. Naïve occupancy ranged from 0.0128 to 0.9359. Occupancy of macroinvertebrates was influenced by fish presence, canopy cover, slope, surface area, and drying of the wetlands but differed between families. The site specific variables had a good fit for hemipteran and odonate occupancy models. Seasonal effects on detection coupled with low detection rates for some families despite repeated intensive samples, suggests the need for caution when using macroinvertebrates as indicators of wetland quality based solely on the assessment of presence/absence from a single sample.