A closed-loop on-site wastewater treatment system (OWT) was studied comprising steps of septic tank to remove organics (Biological Oxygen Demand (BOD)), biofiltration clarifier for biological removal of nitrogen (N), phosphorus (P) and BOD, reactive Polonite® filter for chemical adsorption and precipitation removal of dissolved P, and tidal flow constructed wetland (TFCW) sand filter for polishing the effluent to low P and N effluent Swedish standards. The field experimental data that have been used to optimize TFCW design in the numerical modelling using HYDRUS-2D coupled with and without PHREEQC indicated that the adsorption efficiency of the reactive Polonite® adsorbent was nearly double to that obtained in TFCW sand filters for PO4-P (95 %) and Total-P (85 %) removal in summer at a high temperature range (15.4–18.8 °C) and pH range (9.9–10.8). The weaker PO4-P (53 %) and Total-P (25 %) removal efficiency in winter was due to a low temperature (1.5–8.1 °C) and low pH (7.2–7.9). This decrease in pH was attributed to salinity in the domestic wastewater and dilution of rainwater. Modelling results revealed that the transport mechanisms and rate of P adsorption kinetics in the TFCW sand filters enhanced with calcium and iron flow from chemical dissolution in the preceding Polonite® adsorbent was increased with the increase in temperature. However, the P adsorption was less sensitive at high ferrihydrite (Fe(OH)3) dose, suggesting limited effects of cations dissolution and abundance of metal oxides and hydroxide ions at the mineral surface for anions exchange with phosphate for surface complexation. The strategy of combining field data and modelling provided valuable insights for assessing adaptability and optimizing TFCW design under variable fluxes and scenario effects of insulated/uninsulated and dilution by rainwater in cold-climate regions.

Inadequate data and spatial dependence in the observations during geochemical studies are among the disturbing conditions when estimating environmental factors contributing to the local variability in the pollutants of interest. Usually, spatial dependence occurs due to the researcher 's imperfection on the natural scale of occurrence which affects the sampling strategy. As a consequence, observations on the study variable are significantly correlated in space. In this study, the machine learning approach was developed and used to study the environmental factors controlling the local variability in fluoride concentrations in drinking water sources of northern Tanzania within the East African Rift Valley. The approach constituted the use of geographical information systems (GIS) technology, exploratory spatial data analysis (ESDA) methods, and spatial regression modeling at a local level. The environmental variables used to study the local variation in fluoride concentration include topography, tectonic processes, water exchanges between hydrogeological layers during lateral movement, mineralization processes (EC), and water pH. The study was based on 20 local spatial regimes determined using GIS based on water sources density in the four hydrogeological environments. Specifically, the nonparametric (one-way Kruskal-Wallis sum ranks test and Multiple Comparisons Dunn Test), spatial statistics (Global Moran 's I statistic), ordinary least squares (OLS) regression, and spatial lag models were used to quantify the effects of topography, tectonic processes, water exchange between hydrogeological environments and water physiochemical parameters (pH and EC) on the spatial variability of fluoride concentrations in drinking water sources at a local scale. In order of significance, the local spatial variation in fluoride concentration is influenced by the EC, topography, tectonic processes, pH, and water exchange between hydrogeological layers during water movement. The results presented in this paper are crucial for safe water access planning in naturally contaminated aquifer systems.

Groundwater is consumed by over 2 billion people globally, though it can be impacted by microbial and chemical contamination in both rural and (peri-)urban areas. This issue is particularly pertinent in regions like East Africa, where rapid urbanisation has strained local infrastructure, including water and sanitation systems. We use selected tracers of human and animal waste to assess the quality of community drinking sources with regards to surface-derived groundwater inputs and to compare urban versus rural water quality, under the rapidly developing urban area of Gulu, Northern Uganda. Specifically, we examine bulk and fluorescent dissolved organic matter (DOM), microorganisms (total coliforms, E. coli) and inorganic tracers of anthropogenic waste (NO3−, SO42−, Cl/Br) from various sources: boreholes (12–76 m depth; n = 90), protected springs (n = 11) and municipal taps (n = 4). Our results show that NO3− and SO42− were elevated in groundwater sources in the Gulu city urban area and the Cl/Br ratio was elevated in springs, compared to concentrations in the more rural Aswa and Omoro County area (p < 0.05). Interestingly, human and animal waste indicators E. coli and Tryp:FA (the ratio of tryptophan-like to fulvic-like fluorescence) displayed no significant difference between rural and urban settings (p > 0.05), though total coliforms were significantly higher in rural boreholes (p < 0.05). The presence of a pollution source, pollution carrier and a breakdown of a sanitary barrier at the borehole, as spot-checked by a visual sanitary risk assessment, was significantly associated with groundwater E. coli abundances. Evidence suggests monitoring and mitigation should be improved for all water types in Gulu District to meet WHO and Uganda Standard guidelines for potable water. This study offers valuable insights for water management planning and risk assessment of community water sources particularly in the context of East Africa and similar settings.

This study examined the adsorption capacity and treatment efficiency of sand filters in on-site treatment systems for cold climate regions. The effects of different operating conditions, porosity and kinetics parameters were investigated in column experiments and COMSOL Multiphysics® modelling, to comprehensively reveal the mechanisms and optimize treatment efficiency of nitrogen (N) and phosphorus (P) removal in a field tidal flow constructed wetland (TFCW), treating effluent from a package treatment plant with P filter material. The results from column experiments with sand showed that Total-P adsorption rate was dependent on feed water quality (Septic tank >0.77 ± 0.06 g kg−1; Biotreatment >0.41 ± 0.07 g kg−1; Reactive material Polonite® <0.18 ± 0.07 g kg−1). In the field TFCW trial, Total-P adsorption in the top layer (>1.42 ± 0.55 g kg−1) and middle layer (>1.06 ± 0.51 g kg−1) was twice that in laboratory columns, due to strong interaction with the air-water interface and use of fluctuated domestic wastewater solutions. The breakthrough curve (BTCs) of the coarse sand matched the physical behaviour of tracer electrical conductivity (EC) in effluent from the sand column experiments. The modelling results demonstrated that high filter porosity and low hydraulic load were significant factors for optimal removal of NH4–N, Total-N, PO4–P, Total- P in the top layer (>99.95 ± 0.03 %, 44.37 ± 28.75%, 70.89 ± 28.30%, 76.18 ± 20.3%), middle layer (>98.94 ± 1.77%, 18.23 ± 23.04%, 76.62 ± 28.73%, 65.40 ± 31.85%) and deep layer (>99.99 ± 0.02%, 65.50 ± 20.64%, 75.53 ± 23.16%, 41.54 ± 28.81%) in the TFCW system, respectively. The results show that on-site wastewater treatment in cold climate TFCW can be applied as a technology to polish effluent from a three-step pretreatment system. However, hydraulic optimization is an important factor for the design of the TFCW to receive a successful long-term operating system.

The study evaluates the performance of bauxite for arsenic (As) elucidation from water. The raw, calcined, and alkaline ferric-bauxite composite was applied in batch experiments to evaluate the influence of dosage, initial As concentrations, contact time, and pH. The X-ray diffraction studies revealed significant content of gibbsite (Al(OH)3 in the bauxite. Visual MINTEQ simulation indicated As removal increases with an increase in dosage, the pH range between acidic and near neutral favors maximum removal. The 100 g/L calcined bauxite at pH 7.4 removed 99.9% As to below 0.001 mg/L after 20 minutes from an initial concentration of 1 mg/L. The raw 100 g/L bauxite at pH 6 removed 99.86% As to below 0.003 mg/L after 1 hour from an initial As concentration of 1 mg/L. The alkaline ferric-bauxite composite used for treatment of 2 mg/L As raised pH from pH 4 to 12, and removal efficiency declined to 31% after 4 hours. Aluminium (Al) was sensitive to pH, and about 435 mg/L was released in water at pH 12. Despite the decrease in specific surface area during calcination at 500 °C, the As removal was more improved for the calcined bauxite. The removal capacity was high, up to 6 mg/g, when less dosage of 0.5 g/L bauxite was used. The kinetic reaction process using 5 g/L reaction obeys pseudo-secondorder with R2 of 0.99 while its removal isotherm obeys Langmuir with R2 of 0.98 and is confirmed favorable.

Fluoride is essential for the human body and a global groundwater contaminant (the recommended WHO limit is 1.5mg/L). The mobilization and genesis of fluoride depend on fluoride-bearing rocks (e.g., fluorite, fluor-apatite, biotite, etc.) that are a part of the natural geogenic process, which later contaminate the groundwater. More specifically, the dissolution process (via infiltration), lateral water flow, ion exchange, climatic factors, and chemical weathering of “rocks and minerals” are highly responsible for the release of elevated concentrations of fluoride in groundwater. The intake of fluoride-contaminated groundwater and anthropogenically produced daily usable products (e.g., dental products, foods, etc.) causes physiological and metabolic disturbances in animals and humans. However, this fluoride can be removed effectively from water by technology-enhanced processes (e.g., reverse osmosis, nano-filtration, coagulation, adsorption, electrochemical, membrane distillation, ion exchange, and precipitation). This, in turn, means that climate-dependent contamination, mobilization mechanism, and bioaccumulation will be essential for selecting efficient, cost-effective green technologies. Adequate information should be provided to overcome people’s wrong perceptions concerning fluoride-related issues, especially in lower socioeconomic groups. Policy interventions are required to improve the quality of life in the developing world, where there is a lack of awareness about health issues. Extensive research in this field can identify fluoride “hot spots” (through regular monitoring) and removal technique(s) utilizing public-private sector collaboration.

Geogenic contamination of groundwater due to elevated fluoride (F-) concentrations is a significant issue worldwide (including in Tanzania). The present study focussed to assess the adsorption capacity of thermally treated (calcined) bauxite to remove the F- from contaminated water. Characterization of bauxite by X-ray fluorescence spectroscopy (XRF) revealed Al2O3, Fe2O3, and SiO2 as the major oxides in both raw and calcined bauxite. The major mineral phase in the raw bauxite was gibbsite, which disappeared after calcination. The optimum calcination temperature, dosage and contact time for F- removal by calcined bauxite were 400 degrees C, 40 g/L and 8 min, respectively. The experimental data revealed Freundlich isotherm as the best model to fit the F -adsorption process with kF and 1/n being 0.1537 mg/g and 0.8607, respectively. The pseudo-second-order ki-netic and intra-particle diffusion models explained well the F- adsorption process with the rate constants of 115.43 g/mg min and 0.0025 mg/g min0.5, respectively. The values of Delta G, Delta H and Delta S indicate the F- adsorption on bauxite surface indicated that the adsorption process was spontaneous, endothermic and structural changes occurred during the adsorption process. The F- adsorption under optimum conditions lowered the pH and F -concentration to WHO and Tanzania Bureau of Standards (TBS) standards.

Arsenic (As) removal studies were carried out through batch experiments to investigate the performance of the locally available calcined magnesite mineral rocks from Tanzania. Natural water from a stream source in Tanzania and the prepared synthetic water at the laboratory were used for the studies. Parameters such as initial As concentration, calcined magnesite dosage, contact time and pH were evaluated for As removal using an overhead rea×2 shaker. Arsenic concentration was reduced from 5.3 to 1.1 mg/L As(V) at 180 min when 0.5 g/L calcined magnesite was applied to a synthetic water sample, whereas the concentration of 117 μg/L As(V) and 5.2 μg/L As(III) was reduced to below 0.1 μg/L in natural water. An increase in calcined magnesite dosage resulted in increased As removal up to below 0.01 mg/L. The calcined magnesite raised the pH of the water sample from 6.8 to 10 when the applied dosage increased between 0.002 g/L and 0.05 g/L. The pH was constant at around 10 even when the amount of 0.05 g/L was added 2000 times. Despite the high pH, the amount of magnesium released in water was low. The calcination of magnesite at 500 ◦C increased surface area by 4 times as compared to the natural magnesite and X-ray diffraction showed presence of MgCO3 phase as the dominant phase at this temperature. The reaction kinetics of As removal on 0.5 g/L calcined magnesite fitted with the pseudo-second-order (R2 = 0.96). Reaction isotherm was strongly fitted with Freundlich isotherm (R2 = 0.98). Linear regression and artificial intelligence neural network showed the As removal was influenced by both contact time and pH. Arsenic can be removed from As water using calcined magnesite and will be suitable for water treatment around gold mining areas. 

Geogenic contamination of groundwater is frequently associated with gold mining activities and related to drinking water quality problems worldwide. In Tanzania, elevated levels of trace elements (TEs) have been reported in drinking water sources within the Lake Victoria Basin, posing a serious health risk to communities. The present study aims to assess the groundwater quality with a focus on the concentration levels of geogenic contaminants in groundwater around the Lake Victoria goldfields in Geita and Mara districts. The water samples were collected from community drinking water sources and were analysed for physiochemical parameters (pH, EC, Eh), major ions, and trace elements. The analysed major ions included Na+, K+, Ca2+, Mg2+, SO42-, HCO3- and Cl- whereas the trace elements were As, Al, Li, Ba, B, Ti, V, U, Zr, Sr, Si, Mn Mo, Fe, Ni, Zn, Cr, Pb, Cd, and V. The present study revealed that the concentration levels of the major ions were mostly within the World Health Organization (WHO) drinking water standards in the following order of their relative abundance; for cations, Ca2+-Na+ >Mg2+ >K+ and for anions was HCO3- > SO42- > NO3-, Cl- > PO43-. Statistical and geochemical modelling software such as 31 Studio', IBM SPSS, geochemical workbench, visual MINTEQ were used to understand the groundwater chemistry and evaluate its suitability for drinking purpose. The concentration of As in groundwater sources varies between below detection limit (bdl) and 300 mu g/L, with highest levels in streams followed by shallow wells and boreholes. In approximately 48% of the analysed samples, As concentration exceeded the WHO drinking water guideline and Tanzania Bureau of Standards (TBS) guideline for drinking water value of 10 mu g/L. The concentration of the analyzed TEs and mean values of physicochemical parameters were below the guideline limits based on WHO and TBS standards. The Canadian Council of Ministries of the Environment Water Quality Index (CCME WQI) shows that the overall water quality is acceptable with minimum threats of deviation from natural conditions. We recommend further geochemical exploration and the periodic risk assessment of groundwater in mining areas where high levels of As were recorded.

The sorption capacities of sand filters used for onsite wastewater treatment and their associated risks of phosphorus (P) leaching on contact with rainwater were investigated in column experiments and with modelling tool for over 300 days. Columns packed with sand were exposed to real domestic wastewater of different characteristics and hydraulic loading modes. The wastewater fed into the columns was effluent collected from three different treatment units in the field: a septic tank (ST), biofiltration tank (BF) and Polonite® filter bag (PO). The risk of P leaching to groundwater and surface water was also assessed, by exposing the same sand columns to natural rainwater. Overall results indicated that sand soils can exhibit different adsorption and desorption capacities for electrical conductivity (EC), Total-P, phosphate-P and total suspended solids, depending on the characteristics of influent wastewater, loading rate and total operation time. The removal efficiencies of the sand columns increased in the order ST (98.16 %) > PO (93.36%) > BF (81.57%) for PO4-P and slightly decreased ST (97.11 %) > PO (92.06%) > BF (76.76%) for Total-P columns. All sand columns loaded with actual wastewater solutions from septic tanks and biofiltration tank have demonstrated high risks of phosphorus leaching (> 99.99%) to the groundwater. The modelling was successful captured behavior of EC tracer and adsorption of PO₄-P with acceptable prediction uncertainty in the PO < 8% columns. The modelling results indicated that the decrease of loading rate from 83.3 mL d^-1 to 20.83 mL d^-1 led to an average increase of removal efficiency and prolong operational lifetime and mass of adsorbed Total-P in the sand soil. This study concludes that sand is a valuable filter medium at low loading rate for phosphorus removal in full-scale operations of onsite treatment systems, however very vulnerable for leaching P when in contact with rainwater.