Macroinvertebrate communities in the Wilge River, including seasonality and preferences

Ferreira, M., Dr.; Van Vuren, J.H.J., Prof.The Wilge River is one of many adjoining tributaries of the Olifants River located in the Olifants Water Management Area (WMA4) within the Highveld (11) – Lower Level 1 Ecoregion (Dallas, 2007). These river systems experience extreme demand for natural resources, as they flow through heavily utilised economic hubs. They are closely associated with land modification and pollution, primarily mining and industrial-related disturbances and extensive agricultural activities, all of which are the primary cause of impairment to river health. The primary aim of this study was to study the aquatic macroinvertebrate communities at six monitoring sites along the Wilge River, coupled with two monitoring sites on adjoining tributaries, and to further identify the driving variables that influence these communities both spatially and temporally. Functional Feeding Groups (FFGs) of the aquatic macroinvertebrates and the surrounding land use in the project area was taken into account. The analysis of in situ water quality measured during the period March 2010 to May 2013, clearly illustrated high levels of variation both spatially and temporally. In situ water quality was a limiting factor to the aquatic ecosystem from a Dissolved Oxygen (DO) and Percentage Saturation (DO%) perspective. The remaining parameters were within the South African guideline for aquatic ecosystems (DWAF, 1996). Habitat availability illustrated clear seasonal variation as well, of which the wet season indicated better habitat availability compared to the dry season. The dominant biotopes in the study area were vegetation (VEG) and gravel, sand and mud (GSM). Site WIL04 illustrated the poorest habitat integrity overall primarily attributed to the site’s steep incised banks and deep channel which lacks the stones biotope. The South African Scoring System, Version 5 (SASS5) results indicated that there was a change in the integrity of the aquatic macroinvertebrate community’s in the study area and further illustrated variability both spatially and temporally. It was evident that the aquatic macroinvertebrate communities within the Wilge River and two adjoining tributaries sampled, were generally in a slightly to modified state with moderate variations. The lowest number of taxa, SASS5 and average score per taxon (ASPT) values was recorded at site WIL04 and this was mainly brought about due to changes in flow and habitat availability. The ASPT score ranged from 3.8 at site WIL04 to 7.7 at site WIL02, indicating that the aquatic macroinvertebrate communities were primarily composed of tolerant and moderately tolerant taxa. Of these mostly tolerant taxa, predators and gathering collector populations were the most dominant FFG, with the shredders being the least abundant within the study area. The Bray-Curtis cluster analysis of the aquatic macroinvertebrate communities clearly illustrated a high level of similarity and seasonal variation among the communities. The high similarity was an indication that similar taxa occurred at the sites within the groups identified. However, in accordance with the Similarity Percentages (SIMPER) analysis, there was no clear indication of dominant taxa. There was however a separation of sites TRI01 and WIL04. This was expected due to differences in the physical stream condition (flow) and other habitats / general biotopes, primarily at site WIL04. Stream bed composition is one of the most important physical factors controlling the structure of freshwater invertebrate communities (Mackay and Eastburn, 1990). The separation and similarity of these two sites were not a consequence of dominant taxa, but rather a consequence of differing water quality, habitat availability and common tolerant taxa driving the system. Inclusive, the seasonal variation illustrated was contributing to the changes in the in situ water quality and habitat availability, thus making seasonal variation also a driving variable, in the differences between the sites. The Redundancy Analysis (RDA) bi-plots indicated, as with the Bray-Curtis similarity matrices and related NMDS plot, that there was a distinct seasonal separation. It further illustrated a clear separation of site TRI01 and WIL04 due to reasons mentioned above. All the environmental variables, with the exception of pH, was identified as significant drivers in the river systems (p<0.05). This however varied seasonally. During the wet season, clarity, DO and pH were the significant drivers, while clarity, TDS/EC, percentage saturation and pH were the significant drivers during the dry season (p<0.05). These drivers were expected due to possible sources namely intensive agriculture in the project area. The RDA tri-plots further indicated the significant role that the ASPT, SASS5 score and the IHAS played within the aquatic macroinvertebrate community (p