Poorly treated domestic wastewater and diffuse nutrient loading from agriculture and informal human settlements greatly threaten water resources due to the alteration of ecosystem function and the reduction of the water’s fitness for use. Thus, the aim of the study was to assess nutrient loading in the eutrophic Roodeplaat Dam (RD) to inform water resource management as a foundation for the rehabilitation of the dam. The objectives were to determine Total Phosphorus (TP) and Nitrogen (N) loading capacity into the RD as well as to propose a total mean annual nutrient reduction for the sustainability of the dam. Flow Duration Curves and Load Duration Curves were employed as analytical tools. It was observed from the study N and TP varied significantly among the sites investigated and the actual TP loads were significantly higher than the allowable load throughout the study area. The study further observed that nutrient loading was more prominent during low flows due to the reduced dilution effect. Thus, this study recommended the application of nature-based solutions to control pollution and reduce stormwater and runoff input, as well as employing low-cost green treatment technology options to reduce nutrient loads from domestic wastewater effluent in municipal wastewater treatment pond systems, which requires less energy. Stringent point source pollution control measures were further recommended, and that water quality planning should consider the desired beneficial water use per catchment, considering the impactors.
Research Article
Estimation of Nutrient Load for Effective Water Resource Management in Dams: A case study of the Roodeplaat Dam, Southern Africa
https://doi.org/10.21203/rs.3.rs-3200126/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Roodeplaat Dam
Nutrients Loads
Flow Duration Curve
Load Duration Curve
Total Mean Annual Nutrient reduction
- Wastewater Treatment Works were observed to be the measure contributors to deteriorating water quality in the Roodeplaat Dam.
- Pollution should be controlled at the source to minimize the impacts on water resources.
- Consideration of nature-based solutions was recommended to control pollution from both point and non-point sources.
Improving access to sanitation and drinking water services is one of the Sustainable Development Goals (SDGs) that was adopted by the United Nations (UN) General Assembly in 2015. The adoption of SDGs, particularly SDG 6.3, is aimed at reducing water pollution and eliminating hazardous substances in water. Since water resources are a key socioeconomic development resource, sustainable water resource management should be prioritized as a development cornerstone (Makanda et al. 2022). The African Union Agenda 2063 further identifies unsustainable water resource management and pollution as barriers to improving quality of life and climate change resilience (African Union 2020; Garfias Royo et al. 2022). Effective nutrient management is one of the key challenges on many countries’ development agendas (Bojanowski et al. 2022). For instance, the 2022 SDG extended progress report highlighted that N and phosphorus (P) commonly exceed water quality targets more than any other water quality parameter (UNEP 2022).
Poorly treated domestic wastewater and diffuse nutrient loading from agriculture and informal human settlements greatly threaten water resources due to the alteration of ecosystem function and the reduction of the water’s fitness for use. Nutritional loading and microbiological contamination are common in developing countries, whereas industry-derived pollution is typically localized and associated with land use. In Africa, hazardous and insufficient water supply is common and frequently contributes to the burden of communicable diseases (Lencha et al. 2021). For instance, major lakes in Africa (e.g., Lake Victoria, Lake Tanganyika, Lake Kariba, and Hartbeespoort Dam) and rivers (Nile, Zambezi, and Orange) experience excessive nutrient loading, which leads to eutrophication (Britton et al. 2019; Winton et al. 2020; Morsy et al. 2020; Day et al. 2021; Olokotum et al. 2021; Njagi et al. 2022). Eutrophication is naturally associated with the ageing of lakes (Greeson 1969); however, anthropogenic activities exacerbate this phenomenon.
The common global approaches to determining nutrient loads on a catchment level are: 1) modelling (dynamic), and 2) mass balance (load-oriented). Some of the modelling approaches include Generalised Additive Models (GAMs) that use surrogate data from sensors to predict nutrient loads and commonly use linear regression (Biagi et al. 2022). These models' limitations are that they require unlimited and extensive data, which is not always practical, especially in developing countries. There are also more complex models such as the Soil and Water Assessment Tool (SWAT) and the Agro-hydrological Catchment Research Model – Nitrogen and Phosphorus (ACRU-NP) which use rainfall data to simulate catchment processes and estimate nutrient loads but are not commonly practical due to their respective intensive data requirements (Rossouw et al. 2008). Therefore, considering the limited data and resources, the current study applied the mass balance approach.
The RD was selected for the current study due to its socioeconomic value and associated activities, including conservation area, tourism, commercial crop irrigation, water supply, and recreational purposes (e.g., fishing, and water sports). Moreover, water resources in this catchment support major economic activities and a population of approximately 5 million people (DWS 2017). Water hyacinth (Eichhornia crassipes) has been identified as a major problem in the RD consequently adversely affecting the surrounding landowners, businesses, recreational activities, crop irrigation, and abstraction for water purification (Ndlela et al. 2018; Batayi et al. 2020). This invasive species has been declared a Category 1 weed, for which planting is banned in South Africa (DWAF 2008). Water hyacinths cover the waterbody surface and exacerbate the effects of eutrophication, reducing oxygen levels and resulting in water taste, odour, colour, and turbidity issues (WRC 2010).
The aim of the current study was to assess nutrient loading in the eutrophic RD to inform water resource management as a foundation for the rehabilitation of the Southern African dams. The study's objectives were to determine the TP and N loading capacity into the RD and propose total mean annual nutrient reduction targets for sustainable RD management.
2.1 Description of the Study Area
The RD (25.6289° S, 28.3506° E) is located in the Capital City, Pretoria in the Gauteng Province (GP) of South Africa, in a temperate climatic zone (Fig. 1).
The catchment area is 668 km2 and the dam has a net capacity of 41.9 x 106 m3 of which 50% is from the return flows from the Baviaanspoort and Zeekoegat Wastewater Treatment Works (WWTWs), a mean depth of 10.6 m, and a maximum depth of 43 m at the dam wall (Pieterse and Toerien 1978; van Ginkel and Silberbauer 2007). The water surface temperature of the dam from 2001–2021 averaged between 16 and 25 °C, which is an indication of a monomictic dam with stable thermal stratification during the summer months (Mnyango et al. 2022).
The three major rivers feed into the dam, namely, the Pienaars River, Edendalespruit River, and Hartbeespruit (Morelete) River. These rivers are highly impacted by upstream anthropogenic activities (Modley et al. 2019). The two notable point sources of nutrients in the vicinity are the Baviaanspoort WWTW, which is approximately 10 km upstream of the dam on the eastern bank of the Pienaars River, and the Zeekoegat WWTW, which is located immediately to the west of the dam and discharges wastewater effluent directly into the dam at the Rowing/Canoe Club. There are also non-point sources of pollution surrounding the RD, which include formal and informal settlements, agricultural land use activities along the banks of these tributaries, urbanisation, and industrial effluent (DWS 2017).
2.2 Water Quality Data
The study followed a case-study design where a quantitative methodology was used. Secondary water quality data was sourced from the Water Management System (WMS) of the South African Department of Water and Sanitation (DWS) national database. The water quality samples had been collected at monitoring sites: A, B, C, and D and the discharging points of the two WWTWs, from October 2001 to September 2021 using an integrated sample technique with a 5 m hosepipe and a surface grab sample technique, respectively. All samples were collected in the phonic zone between intervals 0-5 m, every second week. The following water quality parameters were selected for the study: Nitrate and Nitrite (NO3 + NO2) as N and TP. These parameters were selected due to their direct influence on the trophic status of the RD.
The sample analysis followed by DWS was as follows: water samples for physico-chemical analysis for NO3 + NO2 were collected and transferred into full-capacity sterile 250 ml polyethylene sampling bottles to minimize headspace volume (to avoid loss of target compounds) and labelled accordingly. Samples were placed into a cooler box filled with ice packs (to maintain low temperatures) and transported to the Resource Quality Information Services (RQIS) National Laboratory of the DWS laboratory where they were stored under darkness at 4 °C until analysis. The chemical analyses were done by following the approved standard laboratory methods using the Flammable Atomic Absorption Spectrophotometry (FAAS) and Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES) techniques. In addition to the normal samples, the TP was measured using the digestion methods (Van Ginkel 2007). Thereafter, the results from the sample analysis were captured in the WMS.
2.3 Water Quantity Data
Secondary water quantity data used in the investigations were sourced from the DWS national hydrology database (https://www.dws.gov.za/Hydrology/Verified/hymain.aspx). Flow data had been measured from October 2001 to September 2021 (hydrological years) from five active gauging stations, namely, A2H054, A2H055 (tributaries of) situated in Hartbeesspruit River, A2H027 located in the Pienaars River, A2H029 situated in the Edendalespruit River, and A2R009 situated on the dam wall which captures all outflow of the RD.
2.4 Data Analysis
2.4.1 Statistical Analysis
Surface water quality raw data was populated and analysed in a Microsoft Excel spreadsheet using descriptive statistics. Analytical precision, as per the principle of electroneutrality (Younger 2007), ensured the reliability of the analysis. This was achieved by calculating the percentage of charge balance error (% CBE) using the following formula in Eqn. 1.
This study considered only samples with % CBE falling within ±5 % for analysis. Water quality parameters considered for analysis included common major cations (NH3+), and anions (NO3-, NO2-). There were data gaps in 2001-2021 due to poor sampling regimes. However, the current study used the long-term data set analysis which proved to be sufficient to provide insights into the water quality of the dam.
Variation in N and P loads were assessed using repeated measures mixed models in the Variance Estimation and Precision (VEPAC) package of Statistica V14 (Tibco Software, USA). The calendar year was considered as a repeated measure, whereas "site" and "load identity" (i.e., actual versus allowable TP or N load) were considered as fixed effects. Pairwise differences were assessed using Fisher's LSD test. The correlation between flow rate and nutrient loads was assessed using Spearman’s rank order correlation test. The normality of the data was evaluated using normal probability plots of residuals. All non-normal data were rank transformed prior to mixed models or regression analyses. P-values below 0.05 were considered significant.
2.4.2 Loading Capacity
Nutrient loading is the mass of constituents transported in a water stream. Whereas nutrient loading capacity is the greatest amount of loading a waterbody can receive without violating water quality standards (Harding 2015). Nutrient load analysis is based on two principles: (i) loads are directly proportional to flows, and (ii) water quality parameters are often related to flow rates (US-EPA 2007). The tool that was used to determine the relationship between water quality and water flows is the LDC. LDCs are plots of actual pollutant loading to a waterbody superimposed on the allowable loading to the waterbody (Teague et al. 2011). The first step in generating the LDC is evaluating the FDC.
FDC is a cumulative frequency curve that shows the percentage of time during which specified discharges were equalled or exceeded in a given period (Searcy 1969). Thus, the FDCs for the current study were generated using flow data from five gauging stations. Microsoft Excel ranked daily flows according to magnitude (in descending order). Then the exceedance probability for the daily flow was generated using Eqn. 2. Using this equation, flow duration intervals are expressed in percentages, with zero corresponding to the highest discharge (flooding conditions) and 100 to the lowest (drought conditions) (US-EPA 2007).
Where P is the probability that a given flow will be equaled or exceeded; M is the assigned rank calculated using the Excel formula: M = RANK.AVG(number;ref; [order]); and n represents the total number of flow measurements. Then the duration curve was divided into five hydrological classes: high flows (0–10 %), moist conditions (10–40%), mid-range conditions (40–60%), dry conditions (60–90%), and low flows (90–100%) (US-EPA 2007).
2.4.3 Estimation of Nutrient Loads
Generally, the nutrient load is calculated by multiplying the daily flow by the corresponding nutrient concentration. Thus, to meet this requirement, in cases where the flow gauging station occurs within a reasonable distance upstream or downstream of the water quality sampling site, or vice versa, the simultaneous daily flow matching the concentration date is used. Therefore, for the purpose of this study, flow data and corresponding water quality data were ignored when the flow was zero, as the aim was to relate changes in water quality to changes in flow.
Loads were categorized into two groups: the actual (or current) load and the allowable load. Actual loads were calculated by multiplying the actual concentrations of water quality variables by the corresponding flows and the conversion factor (Eqn. 3), whereas allowable loads were determined using the LDC (Eqn. 5) and represent the mass loading rate of pollutants that the waterbody can receive and remain in compliance. Nutrient load calculations were done for monitoring sites A–D and WWTW discharge points.
3.1 Statistical Analysis of Nutrient Concentration and Flow
Nitrogen and TP loads varied significantly among sampling locations over the 20-year study period (N: F3,53 = 179.85, P < 0.001; TP: F3,53 = 28.48, P < 0.001). Post-hoc analysis indicated significant pairwise differences in N loads among all the sites investigated (i.e., Sites A and C; A and D; B and C; B and D; C and D [P < 0.001]), but not between sites A and B (P = 0.46). In addition, TP was significantly lower at site D than at sites A, B and C (P < 0.001). No further significant differences among sites were apparent in TP loads (P > 0.05).
Actual N load varied significantly from the allowable load when the sampling sites were considered collectively (F1,2088 = 10.15, P = 0.001) (Fig. 2). Similarly, TP actual load (Fig. 3) was significantly higher than the allowable load in the study area (F1,1953 = 137.12, P < 0.001). N loads varied significantly among the WWTWs over the study period (F1,19 = 8.43, P < 0.05), whereas TP did not vary significantly among WWTWs (F1,19 = 3.63, P = 0.07). Moreover, the actual N loads were significantly higher than allowable loads at both the Baviaanspoort (F1,767 = 12.42, P < 0.001) and Zeekoegat (F1,423 = 566.25, P < 0.001) WWTWs. A similar trend was apparent for TP, with actual TP being significantly higher at both the Baviaanspoort (F1,669 = 1800.57, P < 0.001) and Zeekoegat (F1,385 = 401.24, P < 0.001) WWTWs.
Tab. 2–6 present correlations between the nutrient concentrations and flow using Spearman’s rank test. With linear regression/correlation data, the p- value is good to consider, but it is more important to assess the R value. An R below 0.4 is considered a weak association. Notable, there was a weak, but significant positive correlation between N concentration and flow when the study area was considered collectively (rs= 0.11, n = 2162, p < 0.001). Similarly, TP correlated weakly, but significantly with flow rate estimates (rs= 0.20, n = 982, p < 0.001).
3.2 Flows into the RD System
Fig. 4 depicts the FDC of the cumulative frequency of 2001–2021 flow data recorded at five gauging stations: A2H027, A2H029, A2H054, A2H055, and A2R009. At A2H027, 90% of flow was ≥ 0.44 m3.s-1 and 10% was ≥ 1.76 m3.s-1. At A2H029, 90% of flow was ≥ 0.01 m3.s-1 whereas 10% was ≥ 0.34 m3.s-1. Relatively similar conditions were observed at A2H054 and A2H055, where 90% and 10% flows were ≥ 0.11 m3.s-1 and ≥ 1.14 m3.s-1, respectively. These findings suggested the dominance of low flows in the tributaries, with the Pienaars River (A2H027) being the major flow contributor.
A large volume of water from the Pienaars River into RD is probably effluent from the Baviaanspoort WWTW (30 Mℓ/day full operating capacity) into the river in the immediate upstream reach of A2H027 (DWAF 2008). The second largest contributor of flow to the RD is the Hartbeespruit River (A2H054), which drains the suburban western part of the RD and an industrial area through its tributary, the Moreleta River (A2H055). The third largest contributor is the Edendalespruit River (A2H029), which mainly flows through the agricultural area on the east side of the RD (Bosman and Kempster 1985). In addition, Zeekoegat WWTW, which discharges into the central point of the RD (monitoring site B), is operating at 50Mℓ/day (58% of full capacity) (Burkard and Van der Merwe-Botha 2017).
At A2R009, there was no flow 50% of the time, whereas for 10% of the time the flow was ≥ 4.71 m3.s-1; the site is at the dam wall, and it is associated with periodic dam spills and releases. As a result, high-flow (flooding) events downstream occur when the RD is filled to capacity. This condition is only likely to occur following substantial rainfall in the upstream reaches of the RD catchment during the wet seasons, whereas, in the dry season, waterbodies are susceptible to water quality impacts. This usually negatively impacts ecosystem health and may have significant cost implications as water may require additional treatment for beneficial use.
3.3 Nutrient Loading Capacity
In the early 1980s, the South African Department of Water Affairs and Forestry (now DWS) promulgated special standards (for the purification of effluent) of 1 mg/ℓ for P and 1.5 mg/ℓ for N as a long-term eutrophication control program (DWAF 1998; DWS 2023). However, this approach has been criticized on the grounds that it provides a blanket approach and ignores that there are differences in the nutrient-receiving capacity of waterbodies and that in some catchments the contribution from non-point sources was high enough that the removal of point sources would have ignored the effects on the trophic status of waterbodies (Pretorius 1983; Griffin 2017).
In the 1990s, DWAF adopted the approach of water quality objectives (as a management tool) that focused on the cumulative impacts on the receiving water resources. Such an approach informed the development of RQOs. In cases where catchment nutrient numerical targets are not yet established, national water quality standards may be used. The RQOs are numerical and narrative descriptors of conditions that need to be met in order to achieve the required management scenario as provided during the resource classification and are required for both the quality and flow of the water resource (Odume et al. 2018). Therefore, the results from monitoring sites A–D were compared against the RQOs instead of the specialized standards. Loads above the LDC illustrated the exceedance of RQOs, and those below the curve indicated compliance.
Fig. 5 illustrates the loading capacities for NO3 + NO2 and TP using the RQO targets of 1.0 mg/ℓ and 0.13 mg/ℓ, respectively. Site A depicted an increase in concentration with a decrease in flows for both NO3 + NO2 and TP, which resulted in frequent exceedances of RQOs under dry conditions. This suggested a constant contribution from point source pollution. Frequent exceedances of allowable concentration were noted in Site C between the 10th and 90th percentiles, which could be an indication that the site receives both point and non-point loads. In site D, water quality exceeded allowable concentrations for both NO3 + NO2 and TP occurred mostly under wet conditions, below the 60th percentile.
Unfortunately, the DWS flow data does not include the effluent flows from the WWTW. Therefore, the effluent load from Zeekoegat WWTW was estimated by adding data from gauging stations A2H027, A2H029, A2H054, and A2H055 and subtracting it from A2R009 data. The estimated effluent flow was plotted against the corresponding water quality concentrations collected at the discharge point (Fig. 6A), and the same approach was applied to site B (Fig. 7). In contrast, the effluent from the Baviaanspoort WWTW (Fig. 6B) was calculated by multiplying the flow from the nearest gauging station (A2H027) by the concentration collected at the discharge point. Importantly, Fig. 6 illustrates a random relationship between nutrient concentrations in WWTW effluent loads into the RD system. Notably, loads increased with discharge before the dilution effect occurred.
Zeekoegat WWTW recorded the highest loads of 3341.8 kg/d at 5 m3.s-1 for N and 3864.9 kg/d at 15 m3.s-1 for TP, and Baviaanspoort WWTW recorded the highest loads of 9193.30 kg/d at 3 m3.s-1 for N and 6487.46 kg/d at 8 m3.s-1 for TP. High effluent loads could be associated with the malfunction of the two WWTWs.
It was observed that high effluent loads were experienced before the system could assimilate the pollution load. In this regard, self-purification, also known as assimilative capacity, is a natural biogeochemical process occurring in any ecosystem and leading to the elimination or assimilation of nutrients or other pollutants by the natural activity of its resident biological communities (Breil et al. 2022). Generally, river flow facilitates assimilative capacity and mitigates the impact of pollution (Farhadian et al. 2015).
The US-EPA (2007) indicated that streams receiving effluent from WWTWs exert a significant influence on water quality at low flows. This was the case at sites A (Fig. 6) and B (Fig. 7), which receive effluent from WWTWs. For instance, it was observed in Fig. 7 that load exceedance for both N and TP occurred during low flows, with N reaching a high level of 1399.52 kg/d at approximately 5 m3.s-1 compared to a level of 225.30 kg/d observed at 15 m3.s-1 and TP reaching a level of 530.57 kg/d at 6 m3.s-1. Whereas the highest TP load of 637.29 kg/l was observed during high flows (15 m3.s-1). These results could be attributed to the discharge from the WWTW; however, a further contribution could be from non-point sources in the catchment.
When the actual load exceeds the allowable loading under dry or low flow conditions, a contravention of the water quality standard occurs (Teague et al. 2011). In this context, water quality standard exceedance near high flows (flow exceedance range <40%) was associated with rainfall events and pollution sources were classified as non-point sources; conversely, water quality standard exceedance near low flows (flow exceedance range > 60%) was associated with dry weather conditions, and pollution sources were classified as point sources. Thus, this study concurred with other studies (Bowes et al. 2008; Slaughter 2011; Slaughter and Hughes 2013) that water quality for sites where there are known point sources is rarely associated with a clear asymptotic decrease in P concentrations with increasing flows as point source load decreases with increasing flow due to dilution of constant input, whereas non-point load usually increases with overland flows.
Dlamini et al. (2019) assessed the compliance of RQOs at a wastewater facility by investigating the effects of nutrient (NO3- and P) loading capacity at the catchment level. The authors revealed that the management of nutrient loads during low flows is crucial, and that compliance status should be based on the flow regime for each season. The findings of the current study concurred with Dlamini et al. (2019) and, in addition, recommended that pollution be controlled at the source to minimize impacts on the water resource.
N and P are major nutrient parameters in water resources, and TP, TN, and biodegradable organic matter are considered critical pollution indicators and the best estimates for pollution loading (Cheruiyot and Muhandiki 2014). Nutrients can be transported to waterbodies through landscapes (Mockler and Bruen 2018). For instance, nitrate (NO3-) is transported to streams via sub-surface pathways, whereas P from diffuse sources is driven by storm events and transported through overland flows (Jordan et al. 2005; Kröger et al. 2007; Tesoriero et al. 2009; Mockler and Bruen 2018). Mining and utilization of P minerals in fertilizers and detergents have caused major changes to the global biochemical P cycle in water resources (DWS 2023). Rapid aquatic plant growth is a common effect of nutrient enrichment in freshwater ecosystems (Yao et al. 2018) and can lead to toxic algal blooms, which constrict water security (Wei et al. 2022).
Fundamental to trophic status assessment is the determination of the relationship between the level of nutrient loading (P and N in particular) and the in-lake conditions (Harding 2008). Thus, the attainment of water quality management objectives is key to avoiding or reducing water quality problems such as eutrophication. This study further recommended that P and N-special standards be updated according to the characteristics of water resources and that water quality be evaluated in terms of desired beneficial water use and per system or catchment, considering the impactors.
3.4 Estimated Nutrient Load Reduction
Tab.7 presents the estimated annual mean loads measured across the RD from 2001–2021. The growing trend on N was noted between 2015 and 2021, with the highest load of 414.74 t/a recorded in 2021. On the other hand, the TP load was observed to be fluctuating through the period of 2001–2021, with the highest load of 173.18 t/a being recorded in 2020.
Notably, the annual TN load exceeded the allowable load 38% of the time, whereas the annual TP load exceeded the allowable limit by 95%. Therefore, the RD was characterized by long-term variation in nutrient load, which could be related to the fact that the catchment was impacted by both point and non-point sources. However, the major contributors were point sources, as depicted in Fig. 6 and 7.
The findings in Tab. 7 were used to estimate the nutrient load reduction. Tab. 8 below presents the proposed total, median annual nutrient load limits (TMANL) for which the RD could reduce nutrient loads in order to meet the desired trophic state. Thus, instream nutrient concentrations must be improved to maintain aquatic ecosystem health, ensure ecological sustainability, and meet all water quality requirements for water users. This can be done by reducing N by 15.3% in order to meet the numerical limit of 1 mg/ℓ and TP by 66.27% to fulfil the 0.130 mg/ℓ requirements.
According to DWAF (1998), approximately 35–50% of the P in domestic wastewater and most of the P in industrial wastes comes from synthetic detergents, and 32% ends up in reservoirs (Quayle et al. 2010). This makes P the most manageable nutrient. While N removal is more complex, for instance, the conventional wastewater N removal process is an energy-intensive process due to the aeration demand (Jia and Yuan 2016). On the other hand, (Oberholster et al. 2019a) suggested a specific algae treatment for domestic wastewater effluent from pond systems as an alternative practice to improve the quality of the effluent. A reduction of 43.1% and 35.1% for TN in the two respective ponds was observed, indicating other possible effective and sustainable methods to be used to manage N loading in African countries. Thus, this current study recommended that each WWTW in the vicinity of the RD be upgraded to remove at least 8% of the N load and 30% of the TP load per facility per year. This can be achieved by the application of nature-based solutions to control pollution and reduce stormwater and runoff input, as well as employing low-cost green treatment technology options to reduce nutrient loads from domestic wastewater effluent, such as using a consortium of microalgae in municipal wastewater treatment pond systems, which requires less energy. Oberholster et al. (2021) presented compelling evidence in terms of the feasibility of using this technology in developing countries to reduce nutrient loads from domestic wastewater effluent.
In addition, rehabilitation programs for dams in developing countries and dams impacted by different sources of pollution, such as RD, would need to focus on the management of both point and non-point sources, taking into consideration the following actions:
- Enforcement of stringent regulatory actions to ensure that the surrounding WWTWs are complying with discharge permits.
- Adoption of the best management practices for non-point pollution sources such as nutrient tracing programs, nutrient and path management programs for fertilizers, and construction erosion and sediment control ordinances.
- The introduction of zero-phosphate detergents in African countries linked to awareness and education campaigns through citizen science.
The eutrophication of dams resulting from nutrient enrichment is globally considered a tremendous threat to freshwater ecosystems and the health of ecological systems. The study employed the FDC and LDC to estimate the effect of nutrient load on the water quality of the dam. It was observed from the study that N and TP varied significantly among the sites investigated and the actual TP loads were significantly higher than the allowable load throughout the study area which indicated that the study was impacted by both point and non-point sources, with the major impactor being point source pollution. For instance, sites A and B were associated with point source loading, site D was associated with non-point source loading, and site C was associated with both point and non-point source loading.
The study further proposed TMANL for which the RD could reduce nutrient loads in order to meet the desired trophic state. Thus, in order to better manage water resources in Southern Africa and strengthen water security against the backdrop of water quality deterioration, countries need institutional tools such as regulatory frameworks and the implementation of Integrated Water Resource Management tools. Such can be achieved when there are information systems and strong compliance and enforcement regulatory frameworks for water resources in place. Additionally, water quality must be evaluated in terms of desired beneficial water use and be evaluated per system or catchment considering the impactors.
Acknowledgements This article resulted from the master’s research study of the corresponding Author. The authors gratefully acknowledge Stanley Nzama, Aphelele Mgabisa and Kgotso Mahlahlane from the DWS for their comments and guidance, the DWS for assistance with the secondary data, and the Centre for Environmental Management in the University of Free State (UFS) for support throughout the duration of this study.
Funding: This study was funded by the National Department of Water and Sanitation, South Africa.
Financial Interest: The authors have no financial or non-financial interests to disclose.
Author Contributions: Methodology, formal analysis, and investigation were performed by Samkele Mnyango. Statistical Analysis was performed by Samkele Mnyango and Christoff Truter. The first draft of the manuscript was written by Samkele Mnyango, and all authors commented on the previous versions of the manuscript. The manuscript was reviewed and edited by Melusi Thwala and Paul Oberholster. All authors reviewed and approved the final manuscript.
Data Availability: Data is available for sharing from the corresponding author upon reasonable request.
Ethical Approval: The study was conducted according to the guidelines of the UFS and approved by the Environment and Biosafety Research Ethics Committee of the UFS under Project No: UFS-ESD2021/0245 on 16 September 2021
Consent to Participate: Authors have consented to participate in the publication process.
Consent to Publish: Authors agreed to publish this manuscript.
Conflict of Interest: Authors do not have any competing interests.
- African U (2020) Agenda 2063 and SDGs Implementation in Africa. African Union Headquarters, Addis Ababa, Ethiopia, South Africa
- Batayi B, Okonkwo JO, Daso PA, Rimayi CC (2020) Poly- and perfluoroalkyl substances (PFASs) in sediment samples from Roodeplaat and Hartbeespoort Dams, South Africa. Emerg Contam 6:367–375. https://doi.org/10.1016/j.emcon.2020.09.001
- Biagi KM, Ross CA, Oswald CJ et al (2022) Novel predictors related to hysteresis and baseflow improve predictions of watershed nutrient loads: An example from Ontario’s lower Great Lakes basin. Sci Total Environ 826:154023. https://doi.org/10.1016/j.scitotenv.2022.154023
- Bojanowski D, Orlińska-Woźniak P, Wilk P, Szalińska E (2022) Estimation of nutrient loads with the use of mass-balance and modelling approaches on the Wełna River catchment example (central Poland). Sci Rep 12:13052. https://doi.org/10.1038/s41598-022-17270-4
- Bosman H, Kempster P (1985) Precipitation Chemistry of Roodeplaat Dam. Water SA 11:157–168
- Bowes MJ, Smith JT, Jarvie HP, Neal C (2008) Modelling of phosphorus inputs to rivers from diffuse and point sources. Sci Total Environ 395:125–138. https://doi.org/10.1016/j.scitotenv.2008.01.054
- Breil P, Pons M-N, Armani G et al (2022) Sustainable Solutions for Environmental Pollution. Wiley 2:1–93. https://doi.org/10.1002/9781119827665
- Britton AW, Murrell DJ, McGill RAR et al (2019) The effects of land use disturbance vary with trophic position in littoral cichlid fish communities from Lake Tanganyika. Freshw Biol 64:1114–1130. https://doi.org/10.1111/fwb.13287
- Burkard T, van der Merwe-Botha M (2017) Biogas Feasibility Study: Zeekoegat Wastewater Treatment Works - Executive Summary: Report for: Zeekoegat WWTW, City of Tshwane. Deutsche Gesellschaft für Internationale. Berlin, Germany, Commissioned by
- Cheruiyot C, Muhandiki V (2014) Review of Estimation of Pollution Load to Lake Victoria. Eur Sci J 10:24–37. https://doi.org/https://doi.org/10.19044/esj.2014.v10n5p%25p
- Day J, Goodman R, Chen Z et al (2021) Deltas in arid environments. Water (Switzerland) 13:1677. https://doi.org/10.3390/w13121677
- Dlamini S, Gyedu-Ababio TK, Slaughter A (2019) The Loading Capacity of the Elands River: A Case Study of the Waterval Boven Wastewater Treatment Works, Mpumalanga Province, South Africa. J Water Resour Prot 11:1049–1063. https://doi.org/10.4236/jwarp.2019.118062
- Department of Water and Forestry (DWAF) (1998) Assessment of the Implementation of the Phosphate Standard at the Baviaanspoort and Zeekoegat Water Care Works. Report No. N/A230/01/DEQ0797. Pretoria, South Africa
- Department of Water and Forestry (DWAF) (2008) Resource Management Plan (RMP) for Roodeplaat Dam. Project 2006 – 304. Pretoria, Republic of South Africa. https://www.dws.gov.za/Documents/Other/RMP/Roodeplaat/RoodeplaatDam.pdf. Accessed 01 Jun 2021
- Department of Water and Sanitation (DWS) (2017) Determination of Resource Quality Objectives in the Mokolo, Matlabas, Crocodile (West) and Marico Catchments in the Limpopo North West Water Management Area (WMA 01): Resource Quality Objectives and Numerical Limits Report. Final. Report No: RDM/WMA01/00/CON/RQO/0516
- Department of Water and Sanitation (DWS), EMP&S/00/IHS/SDS/0520 (2023) Eutrophication Management Strategy for South Africa. Edition 2 (signed version), Project Report Number 4.2, Sources Directed Studies Report Number RDM/. Pretoria, South Africa. Pretoria, South Africa. https://www.dws.gov.za/RDM/WRCS/Doc/Draft%20RQOs%20Report%20(Mokolo,%20Matlabas,%20Croc%20West%20&%20Marico).pdf. Accessed on 14 Sep 2021
- Farhadian M, Haddad OB, Seifollahi-Aghmiuni S, Loáiciga HA (2015) Assimilative Capacity and Flow Dilution for Water Quality Protection in Rivers. J Hazard Toxic Radioact Waste 19. https://doi.org/10.1061/(ASCE)HZ.2153-5515.0000234
- Garfias Royo M, Diep L, Mulligan J et al (2022) Linking the UN Sustainable Development Goals and African Agenda 2063: Understanding overlaps and gaps between the global goals and continental priorities for Africa. World Dev Sustain 1:100010. https://doi.org/10.1016/j.wds.2022.100010
- Greeson PE (1969) Lake Eutrophication? A Natural Process. J Am Water Resour Assoc 5:16–30. https://doi.org/10.1111/j.1752-1688.1969.tb04920.x
- Griffin NJ (2017) Rise and fall of dissolved phosphate in South African rivers. S Afr J Sci 113:7. https://doi.org/10.17159/sajs.2017/20170020
- Harding WR (2008) The Determination of Annual Phosphorus Loading Limits for South African Dams. WRC Report No 1687/1/08. Private Bag X03, Gezina, 0031, Pretoria, South Africa
- Harding WR (2015) A Feasibility Evaluation of the Total Maximum Daily (Pollutant) Load (TMDL) Approach for Managing Eutrophication in South African Dams. WRC Report no.2245/1/15. Private Bag X03, Gezina, 0031, Pretoria, South Africa
- Jia H, Yuan Q (2016) Removal of nitrogen from wastewater using microalgae and microalgae–bacteria consortia. Cogent Environ Sci 2:1275089. https://doi.org/10.1080/23311843.2016.1275089
- Jordan P, Menary W, Daly K et al (2005) Patterns and processes of phosphorus transfer from Irish grassland soils to rivers—integration of laboratory and catchment studies. J Hydrol (Amst) 304:20–34. https://doi.org/10.1016/j.jhydrol.2004.07.021
- Kröger R, Holland MM, Moore MT, Cooper CM (2007) Hydrological Variability and Agricultural Drainage Ditch Inorganic Nitrogen Reduction Capacity. J Environ Qual 36:1646–1652. https://doi.org/10.2134/jeq2006.0506
- Lencha SM, Tränckner J, Dananto M (2021) Assessing the Water Quality of Lake Hawassa Ethiopia—Trophic State and Suitability for Anthropogenic Uses—Applying Common Water Quality Indices. Int J Environ Res Public Health 18:8904. https://doi.org/10.3390/ijerph18178904
- Makanda K, Nzama S, Kanyerere T (2022) Policy Implementation for Water Resources Protection: Assessing Spatio-Temporal Trends of Results from Process-Based Outcomes of Resource-Directed Measures Projects in South Africa. Water (Basel) 14:3322. https://doi.org/10.3390/w14203322
- Mnyango SS, Thwala M, Oberholster PJ, Truter CJ (2022) Using Multiple Indices for the Water Resource Management of a Monomictic Man-Made Dam in Southern Africa. Water 2022, 14:3366. https://doi.org/10.3390/W14213366
- Mockler E, Bruen M (2018) Catchment Management Support Tools for Characterisation and Evaluation of Programme of Measures. Wexford
- Modley LAS, Rampedi IT, Avenant-Oldewage A et al (2019) Microcystin concentrations and liver histopathology in Clarias gariepinus and Oreochromis mossambicus from three impacted rivers flowing into a hyper-eutrophic freshwater system: A pilot study. Environ Toxicol Pharmacol 71:103222. https://doi.org/10.1016/j.etap.2019.103222
- Morsy KM, Mishra AK, Galal MM (2020) Water Quality Assessment of the Nile Delta Lagoons. Air Soil and Water Res 13:117862212096307. https://doi.org/10.1177/1178622120963072
- Ndlela LL, Oberholster PJ, van Wyk JH, Cheng PH (2018) Bacteria as biological control agents of freshwater cyanobacteria: is it feasible beyond the laboratory? Appl Microbiol Biotechnol 102:9911–9923. https://doi.org/10.1007/s00253-018-9391-9
- Njagi DM, Routh J, Odhiambo M et al (2022) A century of human-induced environmental changes and the combined roles of nutrients and land use in Lake Victoria catchment on eutrophication. Sci Total Environ 835:155425. https://doi.org/10.1016/j.scitotenv.2022.155425
- Oberholster PJ, Cheng P-H, Genthe B, Steyn M (2019) The environmental feasibility of low-cost algae-based sewage treatment as a climate change adaption measure in rural areas of SADC countries. J Appl Phycol 31:355–363. https://doi.org/10.1007/s10811-018-1554-7
- Oberholster PJ, Steyn M, Botha A-M (2021) A Comparative Study of Improvement of Phycoremediation Using a Consortium of Microalgae in Municipal Wastewater Treatment Pond Systems as an Alternative Solution to Africa’s Sanitation Challenges. Processes 9:1677. https://doi.org/10.3390/pr9091677
- Odume O, Griffin N, Mensah P (2018) Literature Review and Terms of Reference for Case Study for Linking the Setting of Water Quality License Conditions with Resource Quality Objectives and/or Site-Specific Conditions in the Vaal Barrage Area and Associated Rivers within the Lower Sections of the Upper Vaal River Catchment. Private Bag X03, Gezina, 0031
- Olokotum M, Troussellier M, Escalas A et al (2021) High nutrient loading and climatic parameters influence the dominance and dissimilarity of toxigenic cyanobacteria in northern bays of Lake Victoria. J Great Lakes Res 47:985–996. https://doi.org/10.1016/j.jglr.2021.04.021
- Pieterse AJ, Toerien DF (1978) The Phosphorus-Chlorophyll relationship in Roodeplaat Dam. Water SA 4:105–112
- Pretorius W (1983) Should the phosphate concentration in sewage effluents be restricted? IMIESA 8:2329
- Quayle L, Dickens C, Graham M et al (2010) Investigation of the positive and negative consequences associated with the introduction of zero-phosphate detergents into South Africa. WRC report no. TT 446/10. Private Bag X03, Gezina, 0031, Pretoria, South Africa
- Republic of South Africa (RSA) (2019) Government Notice (GN) No. 42775: Determination of Resource Quality Objectives in the Mokolo, Matlabas, Crocodile (West), and Marico Catchments in the Limpopo North West Water Management Area (WMA 01): Resource Quality Objectives and Numerical Limits. Vol. 652. http://www.gpwonline.co.za, South Africa
- Rossouw J, Harding W, Fatoki O (2008) A Guide to Catchment-Scale Eutrophication Assessments for Rivers, Reservoirs and Lacustrine Wetlands (WRC Project No K5/1343). Private Bag X03, Gezina, 0031, Pretoria, South Africa
- Searcy J (1969) Flow-Duration Curves - Manual of Hydrology: Part 2. Low-Flow Techniques, 3rd edn. Washington
- Slaughter AR (2011) Modelling the relationship between flow and water quality in South African rivers. PhD Thesis. Rhodes University, Grahamstown, South Africa
- Slaughter AR, Hughes DA (2013) A simple model to separately simulate point and diffuse nutrient signatures in stream flows. Hydrol Res 44:538–553. https://doi.org/10.2166/nh.2012.213
- Teague A, Bedient PB, Guven B (2011) Targeted application of seasonal load duration curves using multivariate analysis in two watersheds flowing into lake Houston. J Am Water Resour Assoc 47:620–634. https://doi.org/10.1111/j.1752-1688.2011.00529.x
- Tesoriero AJ, Duff JH, Wolock DM et al (2009) Identifying Pathways and Processes Affecting Nitrate and Orthophosphate Inputs to Streams in Agricultural Watersheds. J Environ Qual 38:1892–1900. https://doi.org/10.2134/jeq2008.0484
- United Nations Environmental Programme (UNEP) (2022) The Sustainable Development Goals Extended Report. https://unstats.un.org/sdgs/report/2022/extended-report/Extended-Report_Goal-6.pdf. Accessed 19 Nov 2022
- United State-Environmental Protection Agency (US-EPA) (2007) An Approach for Using Load Duration Curves in the Development of TMDLs. Washington, DC
- Van Ginkel CE (2007) Investigating the Applicability of Ecological Informatics Modelling Techniques for Predicting Harmful Algal Blooms in Hypertrophic Reservoirs of South Africa. North-West University, Potchefstroom, South Africa
- van Ginkel CE, Silberbauer MJ (2007) Temporal trends in total phosphorus, temperature, oxygen, chlorophyll a and phytoplankton populations in Hartbeespoort Dam and Roodeplaat Dam, South Africa, between 1980 and 2000. Afr J Aquat Sci 32:63–70. https://doi.org/10.2989/AJAS.2007.32.1.9.146
- Water Research Commission (WRC) (2010) Eutrophication Research Impact Assessment (WRC Project No. K8/897). Pretoria, South Africa
- Wei Z, Yu Y, Yi Y (2022) Spatial distribution of nutrient loads and thresholds in large shallow lakes: The case of Chaohu Lake, China. J Hydrol (Amst) 613:128466. https://doi.org/10.1016/j.jhydrol.2022.128466
- Winton RS, Kleinschroth F, Calamita E et al (2020) Potential of aquatic weeds to improve water quality in natural waterways of the Zambezi catchment. Sci Rep 10:15467. https://doi.org/10.1038/s41598-020-72499-1
- Yao X, Zhang Y, Zhang L, Zhou Y (2018) A bibliometric review of nitrogen research in eutrophic lakes and reservoirs. J Environ Sci 66:274–285. https://doi.org/10.1016/j.jes.2016.10.022
- Younger P (2007) Groundwater in the environment: An introduction. John Wiley and Sons, London, United Kingdom
Tables 1 to 8 are available in the Supplementary Files section
No competing interests reported.