Inferring the annual, seasonal, and spatial distributions of marine species from complementary research and commercial vessels’ catch rates

Abstract

The objective of this study is to analyse at fine scale the annual, seasonal and spatial distributions of several species in the Eastern English Channel (EEC). On the one hand, data obtained from scientific surveys are not available all year through, but are considered to provide consistent yearly and spatially resolved abundance indices. On the other hand, on-board commercial data do cover the whole year, but generally provide a biased perception of stock abundance. The combination of scientific and commercial catches per unit of effort (CPUEs), standardized using a delta-generalized linear model, allowed to infer spatial and monthly dynamics of fish distributions in the EEC, which could be compared with previous knowledge on their life cycles. Considering the scientific survey as a repository, the degree of reliability of commercial CPUEs was assessed with survey-based distribution using the Local Index of Collocation. Large scale information was in agreement with literature, especially for cuttlefish. Fine scale consistency between survey and commercial data was significant for half of the 19 tested species (e.g. whiting, cod). For the other species (e.g. plaice, thornback ray), the results were inconclusive, mainly owing to poor commercial data coverage and/or to particular aspects of the species biology.Ecosystem-Based Fisheries Management (EBFM) requires enhancing knowledge of 32 ecosystem functioning, therefore allowing forecasting the impact of fisheries on salient 33 ecosystem components (Long et al., 2015) and to design future management plans and tools 34 including Marine Protected Areas (Meyer et al., 2007) or fishing closures (Hunter et al., 35 2006). This necessitates a stepwise approach, the first tier of which, and one of the most 36 important, is to gain fine scale knowledge on the seasonal and geographic distribution of 37 marine organisms, in general, and fish stocks in particular (Booth, 2000). 38 Scientific surveys have been implemented for decades to derive spatially-and yearly-resolved 39 abundance indices of commercial fish and shellfish species (e.g. van Keeken et al., 2007). 40Surveys provide abundance indices, derived from standardized and controlled protocols, 41 which allow for a wide spatial coverage associated with a weak selectivity (Verdoit et al., 42 2003). Survey data, however, are costly to obtain and therefore rarely provide for adequate 43 seasonal coverage of the resource distribution. In contrast, information derived from 44 commercial fisheries are generally available all year through. Consequently, the catch per unit 45 of effort (CPUE), the most common and easily collected fishery-dependent index of 46 abundance (Maunder and Punt, 2004), has the potential to reflect fish distributions. However, 47 commercial CPUEs can generally not be used directly as abundance indicators. This is 48 because fishers target rather than sample fish densities, and continuously adapt their activities 49 to prevailing conditions, through technological development and tactical ...

Distribution and life history trait models indicate vulnerability of skates

“…To reduce counting false zeros, both landed and discarded individuals that were sampled were considered. To reduce zero inflation, spatio-temporal biases and presence over estimations, only gear types with an even spatio-temporal coverage (set gillnet, trammel net, otter beam trawl and otter twin trawl and Danish seine net for Raja clavata), and a skate bycatch of more than one percent of the total hauls were analysed, similar to Bourdaud et al, (2017). Data outside the known ranges of skate depth distributions (accessed from ObsMer and SBTS data, >100 m for R. undulata and >150 m for R. clavata and R. montagui) and ICES divisions in which almost no skates were identified within were removed to reduce zero inflation ( Table 2).…”

“…Fisheries-dependent and independent data comparative studies (e.g. Bourdaud et al, 2017 andPennino et al, 2016), have shown that distribution models using such data are complementary and coherent to independent models.…”

“…The use of dependent data can provide more insight into species' temporal behaviour as a result of the data collected throughout the year (Bourdaud et al, 2017;Pennino et al, 2016). Furthermore, fisheries-dependent data is a relatively abundant and cheap method to acquire data on fish and therefore ideal for data deficient species.…”

Unravelling the scientific potential of high resolution fishery data

“…Scientific survey data usually have a broader and more homogeneous geographical coverage than commercial fishery data, as fishers target certain species and areas. However, scientific survey data have less intensity and temporal coverage (Pennino et al 2016;Bourdaud et al 2017). While both commercial and scientific data are important sources of information, it is a challenge to link the two types of data and provide a coherent picture (Poos et al 2013;Bourdaud et al 2017).…”

“…However, scientific survey data have less intensity and temporal coverage (Pennino et al 2016;Bourdaud et al 2017). While both commercial and scientific data are important sources of information, it is a challenge to link the two types of data and provide a coherent picture (Poos et al 2013;Bourdaud et al 2017). Currently, integrated commercial datasets rely on coupling data from logbooks, sales slips and the Vessel Monitoring System (VMS) to allocate landings to vessels' hauls and fishing grounds (Hintzen et al 2012).…”