Fisheries Research 82 (2006) 235–245 Management strategies for short lived species: The case of Australia’s Northern Prawn Fishery 3. Factors affecting management and estimation performance Catherine M. Dichmont a,∗ , Aijun (Roy) Deng a , Andr´ e E. Punt b,c , William Venables a , Malcolm Haddon c,d a CSIRO, P.O. Box 76, Cleveland, Qld. 4163, Australia b School of Aquatic and Fishery Sciences, Box 355020, University of Washington, Seattle, WA 98195-5020, USA c CSIRO, GPO Box 1538, Hobart, Tasmania 7001, Australia d MRL/TAFI, University of Tasmania, Private Bag 49, Hobart, Tasmania 7001, Australia Received 4 August 2005; received in revised form 8 May 2006; accepted 4 June 2006 Abstract Management strategies for tiger prawns, Penaeus semisulcatus and P. esculentus, in Australia’s Northern Prawn Fishery (NPF), are evaluated in terms of conservation- and economic-related performance measures. A two-stage process is used to determine the factors to which these performance measures are most sensitive. The first stage involves identifying the possible factors and their interactions, constructing a partial factorial design to allow the impact of first- and second-order interactions on the performance measures to be identified, and analysing the resultant performance measures using generalised linear models. The second stage entails an experiment based on a balanced design of the possible combinations of the key factors. The factors found to have the greatest impact on the performance measures are: (a) how fishing efficiency has changed over time and whether or not the assessment is based on the correct trend in fishing efficiency, (b) the catchability coefficient used to convert from fishing effort to fishing mortality, (c) the difference between the intended fishing effort and the actual fishing effort expended (implementation error), and (d) whether recruitment is spatially correlated among stocks or not. © 2006 Elsevier B.V. All rights reserved. Keywords: Input controls; Management Strategy Evaluation; Monte Carlo simulation; Multi-species modeling; Northern Prawn Fishery; Spatial structure 1. Introduction Australia’s Northern Prawn Fishery (NPF) is an input- controlled multi-stock and -species fishery targeting predomi- nantly tiger and banana prawns. A fleet of less than 100 vessels trawls an area of about 200,000 km 2 (Mick Haywood, CSIRO, personnel communication). Management of the target species is through limits on the length of the fishing season (includ- ing a temporal closure that divides the year into two periods) and a system of individual tradeable gear (fishing effort) units that controls the total amount of headrope length in the fleet. At present, the status of the two species of tiger prawns (Penaeus semisulcatus and P.esculentus) is assessed using a weekly delay- difference model that treats each species separately (Dichmont et ∗ Corresponding author. Tel.: +61 738267219; fax: +61 738267222. E-mail address: cathy.dichmont@csiro.au (C.M. Dichmont). al., 2003a). The assessment involves estimating annual recruit- ment and indices of spawning stock size from catch and fishing effort data, and using these estimates to estimate the parameters of a stock–recruitment relationship. In common with many fisheries worldwide, the NPF does not have a long-term fishery-independent index of abundance so management is based on inferences based on fitting models to catch and effort data. This, combined with the longevity of the target species and the input control nature of the fishery, makes estimation of the values for some of parameter of the assessment model difficult. For example, it has proved impossible to esti- mate the extent to which fishing power has changed over time (Dichmont et al., 2003b). As a result, stock assessments are conducted for several scenarios related to this (bounded by the “Base Case High” (H) and “Base Case Low” (L) scenarios). A second complication is that the data are not informative about the catchability coefficient (the constant of proportionality between fishing effort and fishing mortality) so assessments are based on 0165-7836/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.fishres.2006.06.008