Influence of land use on invertebrate assemblages in southern Appalachian agricultural streams.

We investigated the influence of land-use on benthic invertebrate assemblages in agricultural streams. We hypothesized that agricultural streams with forested headwaters would have different invertebrate assemblages than similar streams that were cleared to the headwaters. To test this hypothesis we compared benthic invertebrate assemblages in six agricultural streams (3 with cleared headwaters and 3 with forested headwaters) in the French Broad River drainage in western North Carolina. Benthic invertebrate samples were collected at three sites along a land-use gradient on each stream. Physical and chemical parameters, including stream discharge, substrate size and heterogeneity, nutrient levels, periphyton biomass and temperature, were also measured along the land-use gradient in order to identify processes which may influence invertebrate assemblages. Land use was assessed using GIS data at varying spatial and temporal scales. The forested agricultural streams have had more riparian vegetation than the cleared agricultural streams since the 1950s, and this trend continued into the 1990s. In general riparian area in forest cover increased from 1950s to the 1970s, although no consistent changes have occurred from 1970s to 1990s. Environmental variables differed between these two stream types, often in response to riparian vegetation but also in response to specific land use. The environmental differences generally coincide with differences often found between forested and agricultural streams. The two stream types also supported different invertebrate assemblages. The forested agricultural stream invertebrates were generally more diverse and more represented by sensitive Plecoptera taxa. Within each season, the invertebrate assemblages at each site were grouped primarily according to present degree of agriculture. The influence of current agricultural practices, particularly heavy cattle impact, interacts with historical forest cover to explain current invertebrate diversity. This may indicate a long term legacy of agriculture, i.e., influences of past agriculture on the stream may persist and influence invertebrate assemblages for up to 25 to 50 years. In addition, the land use immediately upstream explains more variation in diversity than the land use of the entire watershed. The importance of the upstream segment in explaining variation in diversity also indicate that the immediately upstream influences may override the conditions which may be present in the headwaters.1.The study sites were three second- to third-order agricultural streams with forested headwaters and three similar agricultural streams which were cleared to the headwaters in the French Broad River Basin near Asheville, North Carolina. The six streams are located within Buncombe and Madison counties in the Blue Ridge Province of the southern Appalachian mountains. The three forested headwater agricultural (FA) streams originated in forests which gradually opened to cleared, agricultural areas: Heck Creek (a tributary of Walnut Creek), North Fork of Ivy Creek, and Upper Flat Creek. The three cleared headwater agricultural (CA) streams drained areas which have largely been cleared to the headwaters: North Turkey Creek, Pawpaw Creek, and Willow Creek (a tributary of Sandymush Creek). Sampling regime. The six streams were sampled at three sites (headwaters, intermediate, and a downstream site). Invertebrates were sampled (using a Surber sampler fitted with 500 um mesh) in October 1996 and April 1997. A range of physical and chemical variables were measured up to six times during the study. These variables included stream discharge, substrate size and heterogeneity, benthic organic matter, nutrient concentrations, and periphyton biomass. Temperature was recorded continuously by dataloggers at each intermediate site (Site 2) and by maximum-minimum thermometers at each upstream (Site 1) and downstream (Site 3) site