Spatial modelling for optimisation of bivalve culture: a case study for green mussel (Perna viridis) and blood cockle (Anadara granosa) culture in Pattani Bay, Thailand .

Abstract

This study focused on (1) Green mussel (Perna viridis) growth determination and harvestable area models and (2) optimisation model for cockle (Anadara granosa) culture in Pattani Bay, Thailand. The main objectives of this study are: (a) Development of a mussel growth determination model, (b) Development of a harvestable area model for mussels, (c) Development of a suitable area model for cockle farming. Four main, geo-referenced input data sets were developed, comprising (1) water quality parameters, (2) sediment quality parameters, (3) empirical mussel growth and (4) secondary data related to the target species and the bay ecology including satellite images of study area were employed. Data on twelve water quality parameters and four sediment parameters were collected in the field between June 2009 and February 2010 while mussel growth data was collected between June and September 2009. Different combinations of data from these groups were managed and manipulated within the IDRISI™ environment to address the different study objectives and to formulate databases, analyse data, and develop and illustrate spatial models. The main output models of the study were (1) mussel growth determination models, (2) suitable mussel culture and harvestable area models and (3) models for optimisation of suitable cockle culture area. Principal Component Analysis (PCA) and Multiple Linear Regressions (MLR) were employed to determine optimised conditions for mussel growth determination. The main determinants for mussel growth were particulate organic matter (POM), chlorophyll-a and salinity. All three together can predict 85.2% of mussel growth while chlorophyll-a and salinity could predict 84.9% and salinity alone could predict 83.4% of observed mussel growth. The optimum suitable area for mussel farming was estimated using overlaying techniques to combine three main components comprising salinity, water depth and shipping route. Of the 58 km2 of the total water space in the bay, a potential optimised area for mussel farming of 13 km2 was found. The remaining area was limited for culture due to low salinity, inadequate water depth and presence of shipping routes. Based on the empirical growth function of mussel, the mussel harvestable areas were projected and illustrated periodically. The harvest period of mussel within the potential culture site extends over 2 months. The spatial analysis showed that <50% of the current mussel culture operations within the bay are located within the optimal area and the models also identified > 12.5 km2 that could be developed further. The suitable cockle farming area was estimated using a Multi-Criteria Evaluation (MCE) approach based on sediment and water factors reclassified as a fuzzy data set. The maximum water surface area suitable for cockle culture was 48 km2, but this decreased to around 13 km2 when water depth, water current and sediment suitability were taken into account. The overall suitable cockle culture area was found in the middle part of the bay and approximately 75% (10.4 km2) of existing cockle farms were locate in this area. The remaining existing farms were in more vulnerable areas which experienced low salinity, low pH and high water current. It was noted that relocation of cockle culture areas in the bay is technically feasible and could be considered in future management plans. Suitable site selection and management is very important for an unfed culture system. This study indicates the usefulness of GIS technology for spatial planning and area optimization for mollusc culture in the study area. The study results also provide supportive information for spatial management of the bay aimed at sustainable resource use. Not only in the Pattani Bay, the same study protocols could be applied for the others bays, potential coastal culture sites in Thailand and other tropical countries.