Diatoms as water quality indicators in the upper reaches of the Great Fish River, Eastern Cape

Abstract

The Karoo has been exposed to some form of agricultural activity since the early 1900s. This has resulted in the modification of the rivers in this region. Modifications include weirs, dams and removal of riparian vegetation either for planting crops or livestock grazing. These modifications, together with other agricultural effects such as overgrazing, have impacted the quality of the water in the rivers. Chemical analysis is expensive and does not integrate all the biotic factors affecting aquatic ecosystems. For farmers to even consider monitoring their effects on the water quality a less expensive and more integrated form of monitoring needs to be considered. For most farmers the absolute values offered by the traditional chemical analysis are meaningless. Diatoms indices, on the other hand, can provide a more meaningful analysis of their water quality. With this in mind the main aim of this study was to determine if diatoms could be used for biomonitoring in the semi arid conditions in the Karoo. The section of the Great Fish River which is spring fed was chosen for monthly monitoring from 2010 to 2012. Biological samples, and water samples for physical and chemical analysis were collected simultaneously from 5 sites. A total of 51 diatom genera with 269 taxa were identified. Dominant taxa (>5% relative abundance at all sites over all samples) were Amphora pediculus (Kützing) Grunow, Craticula buderi (Hustedt) Lange-Bertalot, Fragilaria biceps (Kützing) Lange-Bertalot, Nitzschia frustulum (Kützing) Grunow, Nitzschia paleacea (Grunow) Grunow in van Heurck, Planothidium lanceolatum (Brébisson ex Kützing) Lange-Bertalot and Rhopalodia gibba (Ehrenberg) O.Müller. These species are predominantly pollution tolerant diatoms. The Generic Diatom Index (GDI), the Specific Pollution sensitivity Index (SPI), the Biological Diatom Index (BDI) and the % Pollution Tolerant Valves (%PTV – part of the UK Trophic Diatom Index TDI) were used to infer that the river water was impacted by agricultural activity. These European diatom indices (GDI, BDI and %PTV) were used. The SPI with the South African Diatom Index (SADI) database was evaluated. Significant correlations were established between the indices and pH, NO3-N and NH4-N. Canonical Correspondence Analysis (CCA) was performed to determine the most important environmental variables. These were Electrical Conductivity (EC) and NO3-N with the same gradient, followed by pH and PO4 again with an almost identical gradient. It was concluded that decades of agricultural activity has had a negative impact on the water quality and the main drivers for the diatom community composition was EC and NO3-N. This result is validated by the abundance of Nitzschia species. A large number of deformed valves were encountered. Neither the chemical analysis nor the diatom indices were able to explain the high number of deformed valves and the severity of these deformities observed.