Consequences of habitat disturbance and recovery to recruitment and the abundance of kelp forest fishes

Abstract

The complete or partial removal of kelp forests due to disturbance can alter species composition and cause large reductions in the abundance of fishes in temperate zones. The effects of a gradient in disturbance that results in the thinning of kelp, however, have not been explored. We conducted manipulations corresponding to 0%, 33%, 66%, and 100% removal of Macrocystis pyrifera (L.) C. Ag at three sites along Santa Catalina Island, California. Surveys of fishes conducted 1–3 months after the removal of M. pyrifera revealed differences in recruitment for Brachyistius frenatus (Gill). Additionally, a trend was seen for Oxyjulis californica (Günther) showing an increase in recruit fish densities in treatments with higher amounts of M. pyrifera. For older life stages, there also were differences among treatments in the numerical densities of Chromis punctipinnis (Cooper) , Paralabrax clathratus (Girard), and B. frenatus and in the biomass densities of P. clathratus and B. frenatus. Algae surveys conducted annually in the 2 years following disturbance showed the stipe densities of M. pyrifera among all treatments were at least as high as at the initiation of the experiment, indicating that treatments of M. pyrifera removal had recovered quickly to pre-disturbance levels. With one exception, there were no differences in recruitment or in the numerical or biomass densities of older juvenile and adult fishes among treatments in either of the two years following the initial disturbance. M. pyrifera in mature vs. recovering forests differs in its distribution and morphology, which might influence recruitment of kelp-associated fishes. To investigate these differences we conducted an experiment manipulating the distribution of adult and juvenile fronds of M. pyrifera. Recruitment of fishes was significantly higher on adult M. pyrifera and bundled fronds of juvenile M. pyrifera than on a dispersed distribution of juvenile M. pyrifera, indicating that the distribution of fronds primarily influences recruitment of kelp-associated fishes. Our results indicate that a gradient in disturbance and variation in the density of fronds of M. pyrifera influences recruitment and the distribution of older life stages of fishes, revealing some patterns at intermediate levels of removal that would not be observed by the simple presence or absence of this habitat. Although recovery of M. pyrifera to pre-manipulation levels was achieved over a relatively short period time in this study, variation in recruitment and the abundance of kelp forest fishes as a result of disturbance will depend on the intensity of the disturbance, the conditions that promote or delay habitat recovery, and the life history characteristics of individual species.