| Summary: | Tropical freshwater ecosystems are understudied and not well understood relative to temperate systems; however, they are becoming increasingly imperiled by escalating anthropogenic impacts. The aim of this thesis was to investigate how tropical freshwater macroinvertebrate communities changed both structurally and functionally over a pristine-rural-urban land-use gradient, in relation to different spatial and temporal scales, as well as to the availability of potential food sources. Fifteen streams in the Panama Canal Watershed were sampled during the dry and wet seasons of 2007 and 2008, for macroinvertebrate communities (benthic and leaf litter), environmental variables, and potential food sources. Along the land-use gradient, in both habitat types, taxon richness, diversity, and evenness all decreased significantly; whereas, abundance increased significantly. For the benthic macroinvertebrate community, unique variation was explained equally well by local (water chemistry and sediment type) and landscape (riparian vegetation and watershed land use) characteristics in the dry season, and landscape characteristics explained slightly more variation in the wet season. Leaf-litter macroinvertebrate community unique variation was better explained by local variables than by landscape variables in both seasons. In terms of potential food resources, fine detritus and inorganic material were the most common across all streams (increased quantities in urban streams) and seasons; whereas, the availability of diatoms and leaf material increased in the dry season. Using gut content analyses, we found that collectors (gatherers and filterers) were by far the most common functional feeding group, increasing in abundance along the land-use gradient. Predators, shredders, and scrapers were all most abundant in pristine streams and decreased along the land-use gradient. Finally, using seven community metrices, a potential biomonitoring tool was developed, the Neotropical Low-land Stream Multimetric Index (NLSMI), which distinguished well among the different levels of stream impairment. This study demonstrated that tropical communities were negatively affected by human land alteration, but that community responses depended on the habitat sampled, the influence of different spatial scales varied between the seasons, and the effect of food resources appeared to be complex. These aspects must be taken into consideration for management decisions and restoration strategies to be effective.
|
|---|
