326 



Fishery Bulletin 103(2) 



30,000 



25,000 



20.000 



15,000 



10,000 



5000 



g 140,000 

 CO 



120,000 



0-5 



5-10 



10-15 



15-40 



0-5 



5-10 10-15 



Depth (m) 



15-40 



Figure 5 



Total, fishable, and exploitable green sea urchin iStrongylocentrotus droe- 

 baclnensis) biomass estimates by depth zone. Top, area 1; bottom, area 2. 



from the population dynamics modeling approach were 

 0.38 and 0.57 (2000) for management areas 1 and 2, 

 respectively. 



Cross validation of sea urchin density surfaces yield- 

 ed a mean residual of 0.50 (median=0, standard de- 

 viation^. 86, skewness=2.80, ra = 60) (Fig. 7). Residuals 

 were greatest in regions with the highest spatial vari- 

 ability, such as sites within depth zones 1 and 2 and in 

 the eastern survey strata. 



Discussion 



Spatial variability and distribution 



The objective of this study was to investigate the spatial 

 variability in green sea urchin density to estimate the 

 biomass of the Maine stock. However, several factors 

 limited the choice of spatial statistical approaches that 



could be used to assess the fishery. In particular, the 

 physical structure of the study area, the dependence of 

 sea urchin variables upon the environment and a high 

 degree of small-scale spatial uncertainty make small- 

 scale approaches inappropriate. 



First, the study area was neither uniform nor con- 

 tinuous. Because the aim of the fishery-independent 

 survey program was to assess the whole population of 

 sea urchins in Maine, the study area had to span the 

 entire coastline. Consequently, the study area encom- 

 passed many features that create discontinuities in a 

 spatial model at varying, yet relatively small, spatial 

 scales. These features included the highly indented 

 coastline, the presence of several hundred islands and 

 the exclusion of regions because of environmental con- 

 straints. Second, green sea urchin variables were not 

 independent of the study area; rather, they were depen- 

 dent on several environmental, ecological, and anthro- 

 pogenic factors. In particular, depth, substrate type, 



