Environmental factors that influence the distribution of coral reef fishes: modeling occurrence data for broad-scale conservation and management

Marine Ecology Progress Series

Hugh P. Possingham¹  Maria Beger¹ 

关键词: Species distribution modeling;    Logistic regression;    Environmental parameters;    Coral reef fishes;    Coral reef conservation;    GLM;   

DOI  :  10.3354/meps07481

学科分类：海洋学与技术

来源: Inter-Research

【 摘 要 】

ABSTRACT: To manage coral reef species, it is important to identify the factors that determine their distribution inexpensively. We identified the remotely measured environmental factors that are most influential in determining the distributions of coral reef fish species on a regional scale. Logistic regression models for 227 fish species related presence/absence data to 4 remotely determined environmental predictor variables: depth, presence of a land–sea interface, exposure, and the distance to the nearest estuary. We compared modeled Akaike information criterion (AIC) values with AIC values of randomly distributed species with different numbers of occurrences and levels of habitat specificity to evaluate model significance. Efficient species distribution models were identified for 118 predominantly habitat-specific fishes of the 227 species for which we had data. All 4 predictor variables significantly influenced the distributions of at least some fish species. Depth was the most frequently efficient variable for single variable models. For combinations of 2 predictor variables, depth and exposure, as well as depth and distance from the nearest estuary, were the prevalent predictors of fish distributions. Several fish species responded to the combination of variables distance from an estuary and presence of the terrestrial–marine interface, indicating that these species depend on intact coastal reef habitat, which is in decline near the main sediment-laden rivers. Statistically significant models were predominantly developed for habitat-specific species. These habitat-specific species are of greater conservation concern than widespread species because threats affect them more severely if the threats are selectively affecting their habitat. For this reason, species distribution modeling using remotely determined environmental data may be an efficient method to build models for habitat-specific species and inform marine reserve design.

【 授权许可】

Unknown