Structure, modeling, and ecological characterization of Bocozelle mangrove habitat in the coastal zone of Saint Marc Bay, Haiti

Very little comprehensive research has been done to characterize the structure of the mangroves in Haiti despite acceleration in their level of degradation. This research aimed to (1) determine the structure and extent of the mangrove forest of Bocozelle, (2) detect any changes that occurred during the past few decades, (3) determine the effects of anthropogenic activities on this coastal forest, and (4) assess the usefulness of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data for extracting mangrove biophysical variables. Data from 32 plots along 9 transects were collected to determine the complexity index and the importance value for this coastal ecosystem. Aerial photographs from 1978 and 2002 were compared to detect change in the mangrove community, and interviews were completed to assess the impacts of anthropogenic activities on this coastal habitat. Then, the in situ biophysical data were correlated and regressed with remote sensing-derived vegetation indices extracted from ASTER imagery acquired at the same time the field data were collected. The results indicated that the Bocozelle mangroves had a very low mean complexity index of 0.43. Among the 4 mangrove species observed, Laguncularia racemosa had the highest mean importance value (131). The forest had a tree height that varied from 2 to 5 meters with a mean of 4 meters and a mean basal area ranging from 0.058 to 0.26 m2/0.1 ha with a mean of 0.15 m 2/0.1 ha. The mangrove area cover diminished 8.89% from 1978 to 2002. From 69 interviews conducted, 86% of the respondents thought that the percent mangrove area cover had diminished with time and 90% pointed out that harvesting was the main cause. Those results indicate that the mangrove of Bocozelle has low structural values.; The vegetation parameters were found to have strong correlations with the derived-remote sensing indices. In particular, the percent canopy closure and leaf area index (LAI) had correlations of 0.851 and 0.908 with the normalized difference vegetation index (NDVI) and the soil adjusted vegetation index (SAVI), respectively. This research supports the utility of remote sensing data for mangrove biophysical investigations.