Auditing the accuracy of a volunteer-based surveillance program for an aquatic invader Bythotrephes

We tested the sampling methods of a volunteer-based monitoring program designed to detect the non-indigenous spiny water flea, Bythotrephes longimanus, and found that the program could detect the majority of Bythotrephes invasions. Volunteers take two vertical hauls with a 30 cm diameter net at each of three pelagic stations. To determine if the volunteers were using a large enough net at their three stations, we performed a 17-lake comparison of the volunteer's net with a 75 cm diameter, research-grade net. We found no difference in the number of stations at which Bythotrephes was detected (paired t-test, p = 0.155) with the two nets, because Bythotrephes densities were above the detection limits for both nets. To determine if three stations were sufficient to detect the invader with the volunteer's net, we deployed it at 30 stations in two lakes with average (Harp Lake, 4.17 Bythotrephes m(-3)) vs. low Bythotrephes densities (Sugar Lake, 0.92 m(-3)). In Harp Lake, repeated randomized sampling of the 30 sets of data indicated that only three stations were needed for 100% capture success. In Sugar Lake, seven stations were needed for 100% capture success, but three stations, the current program design, failed to detect the invasion only 14% of the time.