Assessment of fish populations in still waters using hydroacoustics and survey gill netting: Experiences with Arctic charr ( Salvelinus alpinus) in the UK

Abstract

A diverse range of sampling techniques has been developed for the assessment of fish populations in lakes and reservoirs around the world, none of which is suitable for all types of fish in all types of standing water habitat. Consequently, a common contemporary approach is the combination of several sampling techniques depending on the habitats and species under study, logistical issues and, in some cases, ethical issues. In accordance with this approach, a protocol for the assessment of Arctic charr ( Salvelinus alpinus) populations has been developed to meet the assessment requirements of the European Union (EU) Habitats Directive. The protocol makes extensive use of hydroacoustics (to determine population density by area, hereafter referred to as abundance) and more limited survey gill netting (to determine other biological characteristics while minimising mortalities) to sample this salmonid inhabitant of large and often deep lakes and reservoirs. The protocol is applied for the first time here for Arctic charr in five effectively unfished Scottish lakes and reservoirs (Loch Builg, Loch Doon, Lock Eck, Loch Insh and Loch of Girlsta) between July 2003 and June 2005. Geometric mean abundances of fish of all species ranged from 5.5 fish ha −1 (lower and upper 95% confidence limits of 1.9–15.8 fish ha −1) to 476.9 fish ha −1 (366.5–620.6 fish ha −1), with Arctic charr varying between 1.6 fish ha −1 (0.7–3.6 fish ha −1) and 457.8 fish ha −1 (351.8–595.8 fish ha −1). A total of 405 fish of seven species, i.e. Arctic charr, brown trout ( Salmo trutta), eel ( Anguilla anguilla), minnow ( Phoxinus phoxinus), perch ( Perca fluviatilis), whitefish ( Coregonus lavaretus) and three-spined stickleback ( Gasterosteus aculeatus), was recorded from the five sites. Arctic charr dominated the overall fish communities at three sites, were important at a fourth, but were very rare at a fifth. Age–frequency distributions of Arctic charr encompassed individuals between 1 and 10 years old, although most sites did not show this full range. Assessment outcomes for this diverse group of water bodies ranged from three of favourable status, through one of unfavourable but maintained status, to one of unfavourable and declining status. Following this successful first practical application, comments are made on the protocol in the light of its performance in the field and subsequent data analysis. Wider implications for other assessment activities are also considered, including sampling standardisation and the more demanding community-based, rather than simply population-based, EU Water Framework Directive.