Urbanisation of catchment areas is a major cause of freshwater ecosystem degradation worldwide. As catchments become more developed and river ecosystems become increasingly engulfed in various land use activities, there is a growing need to understand these impacts on freshwater ecosystems. Benthic macroinvertebrates are extensively used as indicators of ecosystem health and have been an instrumental tool in ecosystem monitoring and management. The effects of changing land use on macroinvertebrates at a fine scale however, have not been extensively investigated.
Therefore an investigation was conducted to compare chemical, physical and biological surface water quality parameters and aquatic macroinvertebrate community composition along the first 8 km of the Hartbeesspruit, which contains multiple land use types, in the upper Apies-Pienaar catchment in Gauteng, South Africa.
Five sampling sites corresponding to changes in land use were sampled four times at six-week intervals from September 2013 to February 2014. Influential variables that were recorded included in-stream habitat, riparian cover, flow regime and surface water quality parameters. Physical surface water parameters that were tested in situ included pH, salinity, total dissolved solutes, temperature, clarity and conductivity. Ex situ surface water parameters that were tested included physical parameters (alkalinity and turbidity), chemical parameters (major ions, metal ions and nutrients), and biological parameters (bacteria, coliforms and Escherichia coli). Macroinvertebrates were sampled using Hester-Dendy artificial samplers, which, following a 6 week exposure period, were sampled three times from November 2013 to February 2014. Macroinvertebrates were identified to family level and counted.
Macroinvertebrate community composition across sites was assessed through macroinvertebrate abundance, family richness, SASS score, ASPT, Shannon-Wiener index, Pielou s evenness, non-metric multidimensional scaling and Indval analyses. Nineteen families were collected, of which only three made up 80% of macroinvertebrates sampled. These families were Hirudinea, Chironomidae and Oligochaeta.
Indices of macroinvertebrate community composition indicated a general increase in value from upstream to downstream which showed similar comparative variation between sites to physical water quality parameters (except temperature and clarity), major ions (except arsenic), the metal ion magnesium and nutrient sulphate. Surface water parameters showed patterns indicative of effects due to evaporation, dilution and connectivity of water flow along the stream due to the presence of dams and wetlands. Temperature was an important influence on macroinvertebrate abundance and family richness at a temporal scale.
On a spatial scale the most influential parameters on macroinvertebrate composition were seen to be depth, turbidity and conductivity, and temperature to a lesser extent. The land use types that showed the greatest association with various assemblages were the urban, recreational and least transformed wetland land uses. Although major influential factors, this pattern was not seen to be strictly due to the input of contaminants arising from associated activities, nor the variation in physical characteristics, but rather the discontinuity in flow regime.
It was concluded that at a fine scale, the strongest factors that influenced macroinvertebrate community composition along the Hartbeesspruit, was not land use type but rather the hydrological pathways of connectivity and stream flow that exist within the system. The hydrological pathways influenced values and concentrations of chemical and physical surface water parameters which in turn further influenced macroinvertebrate assemblages present.
Mini Dissertation (MSc)--University of Pretoria, 2015.