43 Towards a new management strategy for Pacific Island sea cucumber fisheries Marc Léopold, 1,2 * Jayven Ham, 2 Rocky Kaku, 2 Jeremie Kaltavara, 2 Jason Raubani, 2 Sompert Gereva, 2 Zacharie Moenteapo, 3 Serge Andréfouët 1 and Pascal Dumas 1,2 Introduction This article summarises the main conclusions of a paper published in 2013 in a special issue of Environmental Con- servation called “Politics, science and policy of reference points for resource management”, and provides an update on the data for sea cucumber fisheries in Vanuatu. The sharp decline in marketable sea cucumber resources in most Pacific Islands seems to indicate that a change is needed in current fisheries management approaches. If the sustainability of this fishery is to be ensured, it is vital to identify which biological data are actually useful for setting effective fishing regulations. Data uncertainty must also be taken into account, and precautionary management should be implemented to maintain sea cucumber resources above their biologi- cal recovery threshold. This article describes the new total allowable catch (TAC) management strategy tested by fisheries departments in New Caledonia and Vanuatu. This alternative strategy could well be tested in other countries in the region. Sandfish fisheries management in New Caledonia Method In New Caledonia, an adaptive TAC management sys- tem has been in use since 2008 for the Northern Prov- ince’s main fishery for sandfish, Holothuria scabra (Fig. 1). The resource had been showing signs of over- fishing, which fishers noted through the decrease in size of sea cucumbers in that area. Methodology for estimating stock biomass From June 2008 to April 2012, stock biomass in the study area (Fig. 2) was assessed eight times. An estima- tion method was designed using on-site censuses and detailed habitat mapping. The census protocol is rapid and cost-effective. Depend- ing on the stage of the tides, counting was done on foot on the reef flat or by free dives in areas less than 2 m deep. The survey teams consisted of two observers — one fisher and one fisheries department agent. Counting took place on permanent 100 m-long and 2 m-wide belt transects (i.e. 200 m 2 per transect). All specimens were counted and measured (length, L and width, W) to the nearest 5 mm, which made it possible to estimate their weights. Sampling covered 40–112 transects, depending on the size of the area being assessed and available logis- tical resources. A map of marine habitats was needed to give a spatial dimension to data collection and to extrapolate census observations to the entire fishing grounds (12–26 km 2 ). This map was created using high-resolution Quickbird satellite images (Fig. 3), because the resolution of large- scale reef geomorphology maps derived from Landsat 7 sensors was not fine enough for the purposes of this study. The image was imported into a geographic infor- mation system (GIS) and processed using a simplified, user-oriented protocol. Some 25 polygons were defined, reflecting different habitats. Habitat surface areas were calculated through GIS. This simple mapping process 1 IRD (French Institute of Research for Development), U227 COREUS2, BP A5, 98848 Nouméa Cedex, New Caledonia 2 Vanuatu Fisheries Department, Private Bag 9045, Port-Vila, Vanuatu 3 Northern Province Fisheries Department, BP 41, 98860 Kone, New Caledonia * Corresponding author: marc.leopold@ird.fr Figure 1. Sandfish, Holothuria scabra, one of the most valuable sea cucumber species on the world market (Image E. Tardy).