Optimization of a survey using spatially balanced sampling: a single-year application of clam monitoring in the Arcachon Bay (SW France)

Type Article
Date 2017-10
Language English
Author(s) Kermorvant Claire1, 2, Caill-Milly NathalieORCID1, D'Amico Frank2, Bru Noelle2, Sanchez FlorenceORCID1, Lissardy MurielORCID1, Brown Jennifer3
Affiliation(s) 1 : IFREMER, Lab Environm Ressources Arcachon, FED MIRA 4155, Issy Les Moulineaux, France.
2 : Univ Pau & Pays Adour, CNRS, Lab Math & Leurs Applicat Pau Federat MIRA, UMR5142, F-64000 Pau, France.
3 : Univ Canterbury, Sch Math & Stat, Christchurch, New Zealand.
Source Aquatic Living Resources (0990-7440) (Edp Sciences S A), 2017-10 , Vol. 30 , N. 11 , P. 37 (1-11)
DOI 10.1051/alr/2017036
WOS© Times Cited 8
Keyword(s) Venerupis philippinarum, bivalve, simulation, survey, Arcachon Bay, GRTS
Abstract

Bivalves are important components of benthic marine and freshwater ecosystems throughout the world. One of the most exploited bivalves used for human consumption is manila clam (Venerupis philippinarum). In Arcachon Bay (SW France), commercial fishers and scientists have developed a monitoring survey to estimate clam stocks to assist in implementing a sustainable management strategy. The survey design that is currently used is based on standard stratified random sampling (StRS). The survey has been undertaken every 2 years since 2006. Each survey costs approximately (sic) 50 000, with funding provided by similar to 20% of the commercial fishers. The survey is quite expensive, given that this resource is managed mostly at a regional level. In 2016 for instance, the survey was not done because of a shortfall in funds to support it. Recent studies on survey designs have focused on new developments that allow for higher statistical efficiency (lower sampling error) coupled with lower survey effort. Among these is the spatially balanced generalized random tessellation stratified (GRTS) design. The aim of this study is to compare the performance of the common StRS method with the GRTS design. To do this, we created a semi-virtual clam population by extrapolating the 2012 field survey results in the whole bay and simulated survey events with the two designs. We then assessed the two survey designs using three threshold precision levels (5%, 10% and 20% precision) for the two estimators of interest (biomass and abundance). We recommend the use of the GRTS design for clam surveys in Arcachon Bay. To achieve the same level of precision, GRTS requires less survey effort than StRS.

Full Text
File Pages Size Access
Publisher's official version 11 1 MB Open access
Top of the page

How to cite 

Kermorvant Claire, Caill-Milly Nathalie, D'Amico Frank, Bru Noelle, Sanchez Florence, Lissardy Muriel, Brown Jennifer (2017). Optimization of a survey using spatially balanced sampling: a single-year application of clam monitoring in the Arcachon Bay (SW France). Aquatic Living Resources, 30(11), 37 (1-11). Publisher's official version : https://doi.org/10.1051/alr/2017036 , Open Access version : https://archimer.ifremer.fr/doc/00410/52102/