Recurrent destructive grazing of successionally immature kelp forests by green sea urchins in Vestfjorden, Northern Norway

Outbreak populations of the green sea urchin Strongylocentrotus droebachiensis Miiller caused widespread decimation of the original Laminaria hyperborea (Gunn.) Fosl. kelp forest in Vestflorden, Northern Norway, during the early 1980s. In the follow~ng decade, some of the resulting urch~n-dominated barren grounds reverted to kelp forest, which, after persisting for more than 5 yr, began to be eliminated yet again. Age analysis of annual growth rings in kelp stipes from field sites at V ~r a y Island i n l c a t e that the time of initial kelp recovery varied from site to site but took place between 1984 and 1987. The earliest independent observations of localized macrophyte recovery were made in 1984 at the Harbour Pier site, where, by August 1992, the new kelp forest had already been eliminated by recurrent destructive grazing. Sea urchin density inside re-established kelp forests at 3 other sites on Vaer~y Island was 45 to 75 ind. m-2 in 1992. These urchin populations had significantly aggregated spatial patterns, and recurrent destructive grazing appeared to be mrmnent. This prediction was verified in 1993 when barren grounds reappeared at all study sltes. The threshold conditions for initiation of destructive grazing have been approximated by a curve in the aggregation-density plane. Sea urchins in the Vestfjorden area are infected by the recently discovered epizootic endoparasitic nematode Echinornermella matsi. In 1992, the prevalence of E. matsiat Varay Island ranged from 8.8% in the barren ground at the Harbour Pier site, to between 13.6 and 21.6% in the successionally immature kelp forest at the 3 other sites. The observed kelp forest recovery at Vceray Island was predicted by the macroparasite hypothesis which states that E. matsi may function as a terminator of sea urchin outbreaks in Northern Norway. However, the succession towards an ecologically mature kelp forest community has been interrupted by the unexpected recurrence of destructive grazing, and the macroparasite hypothesis must therefore be rejected in its present form. Furthermore, these local events may, on the larger time and spatial scales of the current outbreak phenomenon, indicate that the euphotic hard bottom component of the coastal ecosystem in Northern Norway has entered a cyclical domain.