PubAg

Main content area

A Model Framework to Determine the Production Potential of Fish Derived from Coastal Habitats for Use in Habitat Restoration

Author:
Wong, Melisa C., Dowd, Michael
Source:
Estuaries and coasts 2016 v.39 no.6 pp. 1785-1800
ISSN:
1559-2723
Subject:
Decapoda, Zostera marina, coasts, ecosystems, fecundity, fish, fish production, habitat conservation, habitat preferences, habitats, longevity, mortality, seagrasses, sediments, uncertainty, Nova Scotia
Abstract:
Metrics of fish production are often used to guide habitat restoration in coastal ecosystems. In this study, we present a general model framework to estimate the absolute production potential of fish (i.e., fish and large decapods) derived from coastal habitats. Production potential represents lifetime production, whether or not the fish uses the habitat of interest for their entire lifespan. The framework uses an age-structured Leslie population matrix with length-dependent survival and fecundity, coupled with growth and length-weight functions. Uncertainty quantification was also included and accounted for parameter dependencies using copulas. Given the limited abundance data available, we made the simplifying assumptions of steady-state populations and a direct scaling of the resultant proportional stable age distribution with observed fish density (in at least one age class). Literature values for regional estimates of mortality and growth were used. We applied our model using data of fish density from seagrass (Zostera marina, eelgrass) beds and bare soft-sediment bottom on the Atlantic coast of Nova Scotia, Canada. A total of 22 species of fish was collected. Species-specific estimates of fish production potential from seagrass ranged from 8.6 × 10⁻³ to 50.0 g WW m⁻² year⁻¹, with uncertainty estimates being within the same order of magnitude as the median. Production potential of most fishes was enhanced by seagrass relative to adjacent bare sediment. The model framework can be adapted and extended to include increasing complexity (e.g., time dependencies) as more extensive data are acquired, and thus has application beyond that presented here.
Agid:
5558157