An Experimental Analysis of Patch Size, Habitat Subdivision, and Extinction in a Marine Intertidal Snail

Both local and regional extinction processes are central to the population biology of spatially structured populations They are also of critical importance in the management of threatened species and the design of nature reserves We conducted a field experiment to examine the effects of patch size and degree of overall population fragmentation on extinction rates of the marine snail, Nucella emarginata. An equal area (1 m 2 ) was subdivided to different degrees One plate of 1 m 2 , two of 1/2 m 2 , four of 1/4 m 2 , eight of 1/.8 m 2 , sixteen of 1/16 m 2 , thirty‐two of 1/32 m 2 , and sixty four of 1/64 m 2 , were positioned randomly within an 11 × 12 grid (minimum distance between plates of 3 m) above a sandflat in Bodega Harbor, California Plates were colonized with equal densities (128/ m 2 ) of Nucella. There is no evidence for a distinct threshold size for population persistence. This pattern is compatible with models of extinction based on environmental stochasticity, but not with those incorporating pure demographic stochasticity. The observed variability in death rates was also significantly greater than that predicted under models of demographic stochasticity. The effect of fragmentation (the degree of subdivision of the same total area) on extinction rates was inconclusive. After two years of observation none of the degree of subdivision (plate size) treatments has had extinctions on all plates