2000, Peirson and Frear, 2003 and Boswell et al., 2007). These angular measurements (or phase differences) also provide information about objects protruding from the seabed. Angular information has been applied to the acoustic 3D imaging of the deep sea-floor
(see Cutter & Demer 2010 and the references therein). Our objective here is to present a method for discriminating between surface and volume components in the acoustic signal in order to detect the presence and relative density of razor clams within the seabed. The challenge is to use the angular information provided by a split-beam echosounder in shallow waters to extract the relevant statistical features for discriminating among high-density, low-density and depleted razor clam beds. The article is organised as follows. In section 2, the study area, groundtruthing stations and sampling methodology in the acoustic survey are described. In section 3,
the LDK378 clinical trial statistical methods used to analyse the split-beam angular information are presented in detail. Section 4 presents the results obtained with the statistical unsupervised classification. In section 5, these results are discussed regarding their statistical significance MK0683 in vivo and the potential effects that other experimental and environmental factors could have on them. Section 6 presents the main conclusions of the work. The study was carried out in the Ría de Pontevedra (Galicia, NW Spain), an area fished by ten fishermens’ associations that harvest fish, crustaceans and molluscs (bivalves and cephalopods). One of the most economically important molluscs in this area is the razor clam, which includes three different species: Ensis ensis, E. siliqua and Solen marginatus. All of them are infaunal bivalves with an elongated and semi-rectangular shape, usually found in high-density patches (beds), surrounded by very low density areas. The fishermen of Ría de Pontevedra harvest 46 different razor clam beds characterised by continuous sandy areas with a homogeneous mollusc density. These
areas are distributed between 0 and 12 m below the sea surface, with an average size of 11.76 × 104 m2 (Fismare 2011). Three of these razor clam beds, regularly exploited by fishermen, were considered for Cyclic nucleotide phosphodiesterase this study: Raxó, Aguete and A Cova (Figure 1). These three beds are located in sandbars 5–11 m deep and have approximate areas of 9.3, 6.7 and 28.3 × 104 m2 respectively. Based on the razor clam harvesting density, the areas were qualitatively described as very productive (Raxó), productive (Aguete) or non-productive (A Cova) by local fishermen at the time of the survey; we hypothesised that productivity is directly related to density. Six sampling points, two per sandbar (see Figure 3, p. 507), were set up to measure the actual density of razor clams and other (epibenthic) bivalves, and the granulometric characteristics of the seabed.