NEKTON IN GULF COAST WETLANDS: FINE-SCALE DISTRIBUTIONS, LANDSCAPE PATTERNS, AND RESTORATION IMPLICATIONS

Abstract

Many decapod crustaceans and fishes are common inhabitants of flooded salt marshes in the northwestern Gulf of Mexico, but spatial distributions are uneven, and population sizes are difficult to estimate. We measured fine-scale (1–10 m) distributions of nekton on the vegetated marsh surface using enclosure samplers in Galveston Bay, Texas, and used these patterns to estimate population size. Natant decapod crustaceans were abundant in the marsh; densities of juvenile brown shrimp Farfantepenaeus aztecus, white shrimp Litopenaeus setiferus, and blue crabs Callinectes sapidus were highest 1 m from the water's edge and declined rapidly to 10 m from the edge. We developed regression models to describe these fine-scale density patterns and validated the models with independent data on density distributions from two other marsh systems. We used a Geographic Information System to transfer the density models to a natural marsh landscape; the highly fragmented Elmgrove Point marsh was composed of shallow nonvegetated bottom (37.4% of the area) and Spartina alterniflora vegetation (62.6%) with ∼15% of the vegetated area within 1 m of the marsh–water interface. We estimated that this 437-ha salt marsh complex supported populations of 16.2 million brown shrimp, 15.5 million white shrimp, and 11.3 million blue crabs. We divided the marsh complex into 39 sectors and examined relationships between nekton populations and landscape-scale patterns of marsh fragmentation. The amount of edge and the population estimates in a sector increased consistently with the amount of water up to ∼20–25%. Nekton population declines were not apparent until the late stages of marsh disintegration (>70% open water). We also used our fine-scale density models to simulate the effects of adding creeks to a 1-ha created salt marsh of solid vegetation. For shrimp, 1–2 creeks/ha (or a comparable amount of edge) were required to reach populations equivalent to shallow nonvegetated bottom, and up to 6 creeks/ha were required to reach populations similar to those estimated for the natural marsh complex. Simulated populations of blue crabs reached levels in the natural marsh complex with the addition of fewer creeks (2–3 creeks/ha).