Survey design and analysis procedures: a comprehensive review of good practice

Abstract

This paper provides a major review of acoustic survey and-data analysis procedures, it has been prepared as a result of discussions in the FAST working group of ICES. A review of published literature is combined with an assessment of methodology. A consideration of information available a priori and its impact on the choice of survey area and survey design is presented. The subject of predetermined survey tracks using both systematic and random strategies is addressed, indicating the assumptions implied by these choices. Adaptive strategies that might be required for mobile or less predictable spatia1 distributions are discussed along with the advantages and the costs of an adaptive approach. The methods of determining of stock species composition are examined. The methods for assessing the degree of homogeneity species size and proportion are presented. The important choice of averaging method i.e. how the samples provide information on the true density within an area is examined. Sources of error within the estimate are discussed. Firstly, methods for computing the spatial sampling error are examined, and secondly, a summary of other sources of error is presented. A appraisal of these errors is presented and provides an intrinsic error analysis. Finally a brief comparison of the results of acoustic surveys with the results of other techniques is presented. Throughout the paper the assumptions implicit in each choice are discussed, and appropriate selections of survey design and analysis methods are presented in tabular form. The paper concludes with a summary of recommended procedures.