Underwater video analysis as a non‐destructive alternative to electrofishing for sampling imperilled headwater stream fishes

Abstract

ABSTRACT1. Traditionally headwater streams are surveyed using electrofishing, a potentially harmful sampling method. Increasingly scientists are seeking out the least destructive stream sampling methods, especially when sampling imperilled fishes. It is therefore imperative that the efficacy of alternative, non-destructive sampling methods be investigated.2. This study investigated the potential for using underwater video analysis (UWVA) as an alternative to electrofishing for assessing the diversity and relative abundance of imperilled fishes in clear, relatively shallow (<1.5 m) headwater streams.3. Detection rates using UWVA were higher than with electrofishing, relative abundance estimates between the two methods were significantly correlated, the attractive effect of the camera was negligible and UWVA was logistically feasible for sampling small headwater streams.4. These results have clear conservation implications as it was demonstrated that UWVA is a suitable sampling method for estimating relative abundance of fishes in small, clear headwater streams.

SmeltCam: Underwater Video Codend for Trawled Nets with an Application to the Distribution of the Imperiled Delta Smelt

“…In particular, underwater video systems are becoming increasingly popular for studying fishes in marine [7], [8], estuarine [9], [10], [11] and freshwater habitats [12], [13]. They are especially desirable when a key objective is to minimize or avoid the adverse effects of handling stress associated with traditional sampling methods [14], [15].…”

Use of underwater video to assess freshwater fish populations in dense submersed aquatic vegetation

“…Fish arrival rates and time of MaxN can be different among fish species and can be a predictor variable in fish density estimates influencing the duration of sampling events for many fishes, particularly slow-moving or cryptic species (Priede and Merrett 1998;Cappo et al 2004). Ebner and Morgan (2013) found most freshwater fish species in Australian waterholes were detected within 10 min of underwater camera deployments while Ellender et al (2012) found that two imperilled freshwater fish in African streams were observed within 15-22 min. This supports our assumption that any sampling or behavioural biases within Lepomis spp.…”

“…To evaluate specific taxa, including endangered and threatened species, a program that takes into consideration the specific habitats and behaviours of those fish will need to be designed before applying UVC methodology (e.g. Ellender et al 2012). Any application of our boat-based UVC methodology to other environments should consider the influence of depth of deployment and habitat complexity on fish detection probability.…”

“…Sampling designs could be focussed on areas, depths and seasons to evaluate a particular species of concern for conservation (e.g. Ellender et al 2012). The use of the UVC requires water to have adequate transparency, which limits its utility in highly turbid systems.…”

Non-native rainbow trout change the structure of benthic communities in headwater streams of the Cape Floristic Region, South Africa

“…One site was sampled per day and sites were sampled in a random order. Snorkel surveys provide reliable estimates of fish abundance in small, clear streams in the CFR (Ellender et al, 2012;Weyl et al, 2013), and we used three-pass snorkel surveys (Thurow, 1994) to estimate fish densities and population size structures at each site (see Shelton et al, 2014 for further details). The same diver conducted all snorkel censuses so that sampling effort among sites was constant (Hankin & Reeves, 1988).…”