The thesis presents a series of principles, similar to those developed by the Organisation for Economic Co-operation and Development in relation to policy oriented indicators, intended to codify best-practice with regards to the design of pesticide risk indicators for use by farmers. Development of these principles was based upon extensive reviews of the literature concerning the potential exposure to and impacts of pesticides on non-target organisms and the relative merits and limitations of different methodological approaches, discussions with farmers concerning the suitability of different approaches to risk indicators and experimental assessments of indicator performance. These assessments took the form of a three-year field trial examining the impact of different levels of pesticide input upon non-target arthropods and several smaller trials that compared the performance of different insecticides and investigated the effect of differences in pesticide dose on non-target arthropods. The data were then compared with the theoretical results generated by a number of different risk indicators. The thesis also identifies a number of barriers to the implementation of these principles in the form of new pesticide risk indicators. These include restrictions on the range of non-target organisms and potential effects for which high quality data is available and the methodological difficulties associated with the incorporation of sublethal, indirect and ecosystem-level effects into risk indicators. Improvements in the validation of pesticide risk indicators that enable objective evaluations of indicator performance to be made are also required.

Thesis (Ph.D.)--Aberdeen University, 2009. / Title from web page (viewed on Dec. 8, 2009). Includes bibliographical references.

M.Sc. (Zoology) / Viticulture is an important agricultural practice in many countries. The long term use of pesticides in vineyards has resulted in increased concentrations of such pollutants in sediments, water and other environmental compartments. Mitigation measures in agriculture, especially vineyard agriculture, are a prerequisite to the sustained integrity of the natural environment, and specifically the aquatic environment. Mitigation presents itself in the form of water bodies, riparian buffer strips, ground vegetation cover, grassed field paths, natural or artificial wetlands and modification of pesticide application rate. However, to our knowledge no studies exist which highlight the state of dams related to viticulture or the use of mitigation measures, in South Africa. The study aimed to quantify the cumulative risk posed to dams on wine farms and the effect of pesticides (fungicides, herbicides and insecticides) on aquatic ecosystems while evaluating the use of mitigation measures in South African viticulture. Epilithic diatom community structure assessment and frustule abnormalities; zooplankton and macroinvertebrate community structure assessment, and characterization of mitigation measures were carried out at nine representative study sites on wine farms in the Western Cape, South Africa. Pesticide risk assessment models PRIMET and PERPEST were employed to assess the risk that pesticides pose at each site from available data. The Relative Risk Model was used to assess the risk posed to identified Risk Regions. Different risk categories were observed across the sites from PRIMET and PERPEST, ranging from no risk to high risk for specific pesticides and relative risk among the risk regions was assessed. Diatom community structure displayed spatial and temporal variability between sites as well as observable diatom frustule abnormalities due to pesticide input. Zooplankton and macroinvertebrate communities displayed variability spatially as well as temporally and this was linked to the presence of pesticides confirming the predictions from PRIMET, PERPEST and the RRM.

D.Phil (Aquatic Health) / Pesticides have been used to control and eliminate agricultural pests for many years thereby increasing crop yield in agriculture; however, the increasing human population worldwide has put increased pressure on agriculture to produce food. This is especially true in developing countries where the population growth has exceeded the predicted growth. Agriculture has therefore become more reliant on pesticides to increase the crop yield to meet the food demand. Many of the pesticides that are used in developing countries are not well studied in these countries and only limited information on their effects on the local environment is known. The information that is available is often from temperate regions and the effects of the pesticides in more arid, subtropical areas are not known. This is especially true in South Africa where a limited amount of studies on pesticides and their effects on the environment have been completed. The problem developing countries (including South Africa) face is the lack of funding which limit the studies on pesticides. Thus, a method was developed in Sri Lanka and Thailand whereby the risk and effects of pesticides to the environment, specifically non-target organisms, can be estimated using an international database which contains predominantly temperate data of pesticides. The first time this model was tested in South Africa was on the Crocodile River (West) Irrigation Scheme in the North West province. This model is initially dependant on the input of information from local farmers on the usage of pesticides. The data were then combined with secondary data on the pesticide characteristics to establish a preliminary risk posed by the pesticides. This risk could then be validated through biological monitoring and the actual pesticide values measured in the environment. This current study was initiated to test these probabilistic models in another part of South Africa with a more arid climate. The Vaalharts Irrigation Scheme (VHIS) comprises an area of 40 000ha where intensive agriculture activities take place and pesticide use is evident. The water for the irrigation scheme is transferred from a weir in the Vaal River to various irrigation canals and dams downstream. Water is then pumped out of the dams to irrigate the various crops. The runoff and drainage water is routed to drainage canals that return to the Harts River. The main crops in the area are wheat, maize and groundnuts, and a wide range of pesticides are used from the time the crops are planted until they are ready to be harvested. The current study was based on the Crocodile (West) River study in terms of the methodology followed, to further validate the use of the risk assessment methodology in southern Africa. Based on a situation analysis of the study area the following two hypotheses were formulated: the predictive and integrated risk models can be used to assess the ecological risk of pesticides when applied on the VHIS; and the risk information generated with these models can be communicated in an effective manner by using a Decision Support System (DSS).

Ph.D. / The toxicant dichlorodiphenyl-trichloroethane, commonly known as DDT, is a broad spectrum insecticide and is currently banned in most countries due to its toxic effects. However, in some countries restricted use of DDT has been authorized as an effective vector control within malarial control programmes. South Africa is one such country, where spraying of DDT occurs in three provinces including the Limpopo Province, KwaZulu Natal and Mpumalanga. Specifically in the Limpopo Province, spraying of DDT has been ongoing for almost 56 years within the eastern malaria belt of the province. Despite this long term spraying there is still a scarcity of data regarding DDT and its effects on indigenous aquatic organisms in South Africa. Any research regarding DDT will therefore be of the utmost value. It was in this context that the present study was initiated, which primarily aimed to assess the extent of contamination within DDT sprayed areas in South Africa and the associated effects on indigenous species, whilst identifying techniques that could be used in future monitoring of these areas. This assessment was done in the Luvuvhu River catchment at three reference sites and four exposure sites situated within the areas where indoor residual spraying of DDT is done annually. At these sites the extent of DDT contamination within the water, sediment and biota (using the bioindicator pecies C. gariepinus from only the lentic sites) in the Luvuvhu river was evaluated. The results showed that DDT concentrations were well above recommended levels in all three of the measured phases, with the highest concentrations predominantly observed at the Xikundu weir. This site was particularly impacted by DDT due to a combination of its close proximity to the DDT sprayed areas, concentration accumulation from upstream sources and environmental conditions that accentuated contamination. These elevated levels of DDT did, however, not induce significant quantifiable effects in the bioindicator C. gariepinus or in the fish and macro-invertebrate community structures. Specifically, the effects in the catfish, C. gariepinus, were assessed using a range of biomarkers specific to the endocrine disrupting effects of DDT, including indirect measures of vitellogenin (calcium, zinc, magnesium and alkali-labile phosphate (ALP) that are all present on the VTG molecule in high abundances), gonad-somatic index (GSI), condition factor (CF), analysis of covariance (ANCOVA) manipulated gonads, protein carbonyls (PC) and intersex. Although none of these biomarkers could be significantly correlated with the DDT contaminations, DDT was shown to induce a slight sub-organismal effect by slightly inducing the synthesis of ALP and Ca as well as reducing the gonad mass (shown by GSI and adjusted gonad mass biomarkers) and body condition. In contrast, the fish and macroinvertebrate communities showed no conclusive relationship with DDT contamination, using a variety of methodologies, including informal assessments, univariate diversity indices, multivariate statistics, abundance models, fish response assessment index (FRAI) as well as average score per taxon (ASPT) and Ephemeroptera, Plecoptera and Trichoptera (EPT) richness. In conclusion, it was shown that DDT concentrations within the Luvuvhu River only induced effects at the lower levels of complexity, which highlights the importance of the utilisation of biomarkers to measure more subtle long-term effects as compared to the usage of community level effects.