Lehodey et al.: Modelling the distribution of Beryx splendens 



757 



since the standard deviations of the parameters are 

 high (Table 5), spatial extrapolation of the model to 

 seamount B is rather crude as demonstrated by com- 

 parison of actual and predicted CPUE (Fig. 4). 



Temporal validation was carried out on data from 

 the Fukuju Maru and RV Alis fishing on seamounts 

 B and J (Table 6). It is unsatisfactory because the 

 mean values of the residuals are not centered on zero 

 and the residuals are correlated with the length vari- 

 able for the RV Alis data and with the depth vari- 

 able for the Hokko Maru data (Fig. 6). As with the 

 bivariate normal model, this suggests the existence 

 of factors not accounted for by the model. 



Discussion 



Alfonsino length structure variation observed over 

 the seamounts of New Caledonia is similar to that 

 noted in Japan 2 and in New Zealand 10 (Massey and 

 Horn, 1990) where it was assigned to age-specific 

 migrations. In Japan, it was noted that alfonsino 

 move south as they grow, 2 young fish predominate 

 over some seamounts and old fish predominate over 

 other seamounts. In New Caledonia, age segregation 

 over the seamounts is so marked that it has been 

 possible to describe it mathematically. 



The bivariate normal and recursive models appear 

 to be complementary. The bivariate normal model 

 provides an instantaneous picture of alfonsino popu- 

 lation distribution on a given seamount; it provides 

 good CPUE estimates provided a sufficient amount 



of length and depth data are available. The recur- 

 sive model takes into account the dynamic nature of 

 the population's distribution as it allows the extrapo- 

 lation of CPUE obtained for one seamount to sea- 

 mounts that were not sampled. It allows preliminary 

 population estimation of unexploited stocks. Depend- 

 ing on current economic parameters, the model might 

 be used to indicate the depths at which fishing is 

 most economic. Once a fishery is operational, more 

 refined data will be available, which will enable the 

 bivariate normal model to be applied and stock man- 

 agement parameters defined for each of the sea- 

 mounts fished. 



The poor results obtained for the temporal valida- 

 tion could be due to poor precision of the depth data 

 collected from the longliners Hokko Maru and Fukuju 

 Maru. Also, neither of the models incorporate a time 

 factor. The data were collected from cruises carried 

 out in different years and in different seasons. Hence, 

 it is unlikely that conditions remained stable, par- 

 ticularly with regard to exploitation history, repro- 

 ductive behavior, or long-term climatic variations. 



Fishing methods and strategies were not modified 

 during the fishing period considered. Therefore, the 

 catches are probably representative of the standing 

 stock of alfonsino within the size limits determined 

 by the selectivity of the fishing gear. Since the daily 

 observation window did not change, vertical trophic 

 migrations would seem unlikely to contribute to the 

 observed variability. With regard to sex as a source 

 of variability, although the mean length of females 

 exceeds that of males 11 (Kotlyar, 1987; Massey and 



10 Horn, P. L., and B. R. Massey. 1989. Biology and abundance of 

 alfonsino and bluenose off the lower east coast. North Island, 

 New Zealand. N. Z. Fish. Tech. Rep. 15, 31 p. 



Lehodey, P. 1994. Les monts sous-marins de Nouvelle-Caledonie 

 et leurs ressources halieutiques. These de doctorat de 

 l'Universite Francaise du Pacifique, 398 p. 



