Using Occupancy Modeling to Improve the Accuracy of Electrofishing Abundance Estimates for Riverine Smallmouth Bass

Occupancy modeling increases the reliability of animal population estimates by explicitly accounting for imperfect and variable detection. We examined the use of occupancy modeling to improve the accuracy of electrofishing for estimating Smallmouth Bass Micropterus dolomieu abundance. Local Smallmouth Bass populations were sampled using tow-barge electrofishing across a range of environmental conditions in streams of the Ozark Highlands. Mark-recapture estimates of Smallmouth Bass abundance derived from generalized multinomial N-mixture occupancy models consistently agreed with baseline estimates obtained via snorkel surveys, even at sites where either capture probability or fish abundance was low. Estimated detection of Smallmouth Bass using mark-recapture at average sampling conditions was 0.26. Detection covariates that explained variation in capture probability of Smallmouth Bass were electrofishing intensity, water clarity, wetted width, and water depth. We identified a viable option for stream managers to improve the dependability of electrofishing for both local and regional monitoring of Smallmouth Bass populations. Additionally, the flexible framework of occupancy modeling could easily be extended to other sampling methodologies and stream fishes and provides an effective way to develop strategic monitoring protocols while maximizing resources. Our study also highlights the benefits of dual stream fish sampling approaches to validate the accuracy of methods.