The average value of turbidity for the dry and wet periods was 12.52 and 3.39 NTU, respectively. The results show that all the water quality parameters analyzed fall under the permissible limit of the World Health Organization (WHO) for drinking water, except turbidity. Here, an integrated approach using both in-situ measurements of water quality parameters and remote sensing data was used to derive the water quality index (WQI) and inherent optical properties of water to deduce the factors governing seasonal and annual variability. This study aims to quantify the water quality on a temporal scale in the Doorndraai dam site in sub-Saharan Africa to design possible management options. The problem of water scarcity and clean water in sub-Saharan Africa is a growing concern. The highest predictive performance, reported in terms of R2, was 97% in the rural catchment and 82% in an urban catchment. Results show that ET, MinMax scaler, and a multivariate imputer were the best predictive model, scaler, and imputer, respectively. A specification book, sensitivity analysis, and best practices for developing virtual sensors are discussed. The effect of data scaling and missing value imputation were also assessed, while the Shapley additive explanations were used to rank feature importance. In the present work, a random forest, extremely randomized trees (ET), extreme gradient boosting, k-nearest neighbors, a light gradient boosting machine, and bagging regressor-based virtual sensors were used to predict N and P in two catchments with contrasting land uses. In these cases, machine learning techniques may serve as viable alternatives since they can learn directly from the available surrogate data. However, the high-frequency monitoring of these water quality indicators is not economical. To better control eutrophication and HCB, catchment managers need to continuously keep track of nitrogen (N) and phosphorus (P) in the water bodies. Harmful cyanobacterial bloom (HCB) is problematic for drinking water treatment, and some of its strains can produce toxins that significantly affect human health. Thus, this study recommends the implementation of an integrated management approach, which needs to prioritize nutrient management to retain societal resource value.
The selected indices were found to be effective for water resource management and could be applied to dams impacted by point-source pollution in Southern Africa. However, strategic goals should involve widening fitness for use. As such, continued nutrient enrichment must be mitigated to sustain fitness for irrigation, at least. Moreover, the WQI calculated for the dam with an average of 93.94 demonstrated very poor water quality that could be used for crop irrigation purposes only. Although the dam was classified as being eutro-hypertrophic, it was evident that water clarity was not a limiting factor but was P-limited, which was an indication of limiting conditions on primary production. The Roodeplaat Dam exhibited the spatial variation of physicochemical characteristics, indicative of influence by point-source pollution. The selection of indices for the study was based on the impacts of anthropogenic activities on the dam. This study employed different indices, namely the weighted arithmetic water quality index (WQI), Carlson Trophic State Index (TSI), van Ginkel TSI, and Trophic Level Index (TLI) to determine the water quality status of a man-made dam for the needs of sustainable water resource management in Southern Africa.
0 kommentar(er)
