Ecosystem effects of a rock-lobster 'invasion': comparitive and modelling approaches

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

An eastward shift in the West Coast rock lobster Jasus lalandii took place in the early 1990s along the south-west coast of South Africa in an area known as East of Cape Hangklip (EOCH). Given the predatory capabilities of J. lalandii, an intricate relationship between the urchin Parechinus angulosus and juvenile abalone Haliotis midae, and an already over-exploited ecosystem, the lobster âinvasionâ is assumed to have had serious consequences on the benthic community and associated fisheries. To investigate these effects, I used both empirical and modelling approaches. Based on field studies, I first analysed temporal changes in rock lobsters and the benthic community at two lobster-invaded sites EOCH. Prior to 1990, rock lobsters were rare EOCH, but from the mid 1990s onwards they achieved densities of 0.4- 0.8 m-2. The pre-invaded benthic community was significantly different from the postinvaded community at both sites. Three major changes followed the lobster invasion: (1) a decline or even a disappearance of herbivores (a direct effect), (2) macroalgal proliferation (an indirect effect) and (3) increases of a range of sessile taxa (unknown effect). This was followed by a detailed spatial analysis of J. lalandii densities and the benthic community EOCH, in which I surveyed six sites (three invaded and three noninvaded) at three different depth zones (< 5 m, 6-12 m and 13-20 m). At all three depth zones Jasus lalandii was significantly more abundant in invaded areas than in noninvaded areas, and invaded and non-invaded benthic communities were significantly different. Invaded sites were characterized by higher densities of rock lobsters, macroalgae and sessile species, whereas non-invaded sites had greater amounts of herbivores and encrusting corallines. Abalone abundance reflected their previouslyrecorded dependency on urchins and the effects of rampant poaching. Floral species diversity was on average greater at invaded sites and increased with depth, whereas faunal species diversity was greater at non-invaded sites but also increased with depth. The depths in which strongest effects of J. lalandii were felt coincided with the depth of maximum abundance of the urchin Parechinus angulosus, the abalone Haliotis midae, the kelp Ecklonia maxima and encrusting corallines. In line with an ecosystem approach to fisheries management and to better understand the ecosystem dynamics EOCH, a lobster-urchin-abalone Minimally Realistic Model was developed for a lobster-invaded area, and an adjacent nonvi invaded area. A key feature of the model is that its focus was restricted to the critical interactions of interest and was fitted to all available data. An existing abalone stockassessment model formed the foundation of this multispecies model, to which rock lobsters and urchins were added. Abalone and rock lobsters were modelled using agestructured production models and urchins were modelled using a simpler surplus production model because of data limitations. The model estimated a lobster starting biomass (in 1985) of 314 tonnes (MT) and a carrying capacity of 1511 MT. Speciesinteraction parameters, particularly the lobster-abalone interaction, were difficult to estimate and the model was unable to estimate the urchin intrinsic growth rate parameter because the data had insufficient contrast. Results from the model suggest that the lobster invasion was probably caused by an influx of adult lobsters as opposed to increased larval settlement. Future projections suggest that given the virtual disappearance of urchins, complete removal of rock lobsters would be needed to allow the urchin population to re-establish itself. Recovery of urchins could take up to 50 years and recovery of abalone would take even longer. The model-predicted differences in lobsters, abalone and urchins between invaded and non-invaded areas paralleled empirical data. Further model explorations included (1) a hypothetical lobster invasion into a currently non-invaded zone EOCH and (2) the addition of a top fish predator into a lobster-invaded zone. Various hypothetical lobster invasions into the non-invaded zone all resulted in the eventual disappearance of urchins and, consequently, juvenile abalone. Available catch and effort data for fish indicated that a major decline in linefish has taken place, but that this occurred prior to the 1980s and was thus not the direct cause of the rock-lobster invasion. This was supported by outputs of a model incorporating fish predation, which demonstrated that the presently over-exploited fish biomass has very little effect on the rock lobster population, but that at historical pristine levels, fish would have been capable of preventing the establishment of a dense rock-lobster population and the consequent disappearance of urchins and abalone. These results indicate that the over-fishing of top-predators would have had massive ramifications for the rest of the ecosystem. Through dual empirical and modelling approaches, my study highlights the complexity of ecosystem interactions and the need for multispecies models in developing an ecosystem approach to fisheries management, and adds to the understanding of the causes and implications of human- and environmentally-induced shifts in community structure.