On 'variability' as a sampling artefact: the case of Sardinella in north-western Africa

Objective evaluation of the global impact of fisheries on ocean ecosystems may be hampered by various biases suggesting natural variability of exploited species to be stronger and more widespread than is really the case. One of these is reporting biases: papers are usually not published which show that nothing has changed. Another such bias is that much variability is fishery-induced, i.e. due to the truncation by fishing of the age composition of exploited populations. A third source of bias, emphasized here, is that resulting from sampling a migrating population with a fixed device. This bias is illustrated by contrasting the relatively stable echo-acoustic estimates of biomass of Sardinella spp. along the north-west African coast, i.e. from Morocco and Mauritania to Senegal (data from 1992–98), with the more variable estimates of biomass in the waters of each of these countries. We conclude that published reports of 'variability' in exploited species should explicitly account for the effect of migrations and other movements, especially when such reports are to be used for contrasting fisheries-induced with environmental impacts on biomass.