< Back to previous page

Project

Design of Rooftop Rainwater Harvesting and Development of Low-Cost Water Treatment Technologies for Dairy Cattle and Farmers

The quality of drinking water for animals is rarely a concern for many dairy farmers but yet a vital component of dairy cattle's daily required intake. Rainwater harvesting can supplement other water sources and the operation and maintenance costs remain low after the initial investment to meet the increasing freshwater demand. Additionally, the drinking water quality standards of dairy cattle differ from those of humans, and this is not usually emphasised especially in low-income countries where both animals and humans share the same source of water. Water containing high percentages of microbes can lead to outbreaks of coliform-related diseases. Heavy metal consumption through drinking water can result in bioaccumulation and have negative effects on dairy cattle. Therefore, this PhD research aims at the design of rooftop rainwater harvesting and development of low-cost water treatment technologies for dairy cattle and farmers. The specific objectives included investigating the reliability and economic assessment of rainwater harvesting systems for dairy production; determining the spatial and temporal variability of drinking water quality for dairy cattle and this informed the design and evaluation of a large-scale water treatment technology for dairy cattle and design and evaluation a low-cost point-of-use water treatment technology for dairy farmers. The most novel part of this research is the multi-level filter for the treatment of water for dairy cattle that showed abilities to reduce bacteriological contaminants to the required standard.

The study was conducted in the Rwenzori region in Uganda, specifically in the Kabarole, Kamwenge and Kyenjojo districts. Princeton Global Forcing (PGF) data gridded at (0.25° × 0.25°) were used after comparison with observed data. The water balance model was used to determine the optimal tank size and hence assess its time-based reliability. The economic viability of the optimised tank sizes was determined using the net present value, benefit-cost ratio and discount payback period. To determine the spatial and temporal variation of selected pollutants in drinking water for dairy cattle, water samples were collected from 40 dairy farms for one year. The different contaminants were analysed for conformity with livestock drinking water quality standards. A two-step multi-layer filter (MLF) system was designed and evaluated for the treatment of drinking water for dairy cattle. The filter consisted of graded sand, supporting gravel and permeable filter layers consisting of fine sand, rice husks, charcoal, and iron arranged in a brick layering format. Multi-layer filters were operated under field conditions along a polluted river and a small stream at a constant flow rate of 750 L/day and then optimised by varying the material composition. Finally, the ceramic water filters coated with zinc oxide nanoparticles biosynthesised with Moringa oleifera were evaluated for the treatment of drinking water for humans.

The results showed that the PGF underestimates the precipitation at gauging stations in the study area. PGF was better at estimating precipitation in the December, January and February season. Bias correction was applied before the data was used for further analysis. The optimum water tank size varied based on the collection area and the climatic conditions. According to the results, the maximum reliability obtained was 99% at a collection area of 500 m2 for a demand of 750 L/day. Reliability was investigated for different tank volumes and an insignificant increase in reliability was observed for tank sizes greater than 50 m3. A roof collection area of at least 250 m2 and a tank size of 120 m3 and above are recommended to be able to achieve more than 50% reliability. Installation of RWH systems is viable with a payback period of fewer than 4 years. The payback period is short due to benefits realised from milk sales that accounted for 95% of the benefits. Therefore, there is a need to consider the use of alternative water sources to supplement RWH when rainwater is not sufficient.

Water quality assessment of various alternative water sources showed that the presence of heavy metal varied from low to very high. For most sources,  metal concentrations were within acceptable ranges except iron, nickel, chromium, aluminium, lead and copper where the percentage of samples above the recommended limit was 82%, 26%, 22%, 9%, 8% and 1%, respectively. The microbial profile from the water sampled revealed isolation of E. coli exceeded the limits by 54% times out of the samples analysed on the cattle farms. There was a significant spatial difference in the concentration of all pollutants across districts except E. coli, lead and nickel. All the other parameters whose limit was exceeded (except for Fe, turbidity, TDS and E. coli) were higher in the dry season compared to the wet season. This could indicate most of the pollutants are naturally occurring in the study area. The water quality index showed that the best water source was rainwater, and the worst was water from the utility. Water from open wells and valley dams was indicated as poor and unsuitable, respectively. Therefore, water for both animals and humans needs to be treated before consumption.

The designed multi-level filters for the treatment of water for dairy cattle show that most contaminants are removed at the first stage. The second stage shows an improved reduction in all pollutants especially faecal coliforms and total coliforms. The results demonstrated that the best filter reduced turbidity, sulphates and nitrates by 90%, 38%, and 27%, while electrical conductivity and total dissolved solids increased by 8% and 9% respectively. The mean reduction efficiency for the overall filter is 92%, 54%, 47%, 51%, 28%  and 93% for aluminium, copper, chromium, iron, fluoride and nickel, respectively for heavy metals. For bacterial indicators, a combination of the filters achieved log removals between 2.19 and 0.5 log units. The effectiveness of the filters was influenced by the quality of the influent water. The small stream had higher contaminant concentration and presented higher removal efficiencies than the river. A significant relationship was found between influent contaminant concentration and removal efficiency. The filter was able to remove bacteriological contaminants up to the required stand for drinking water for dairy cattle. The overall cost of the filter installation is estimated at 500 Euro. The application of MLF could be considered a promising solution for on-site drinking water treatment for dairy cattle in low-income countries. It is recommended to consider monitoring these water sources and putting in place national regulations for livestock drinking water quality.

Furthermore, the results show that ZnO NPs were biosynthesised using Moringa oleifera. The CWFs coated with biosynthesised zinc oxide nanoparticles (CWF-Zn) for the treatment of drinking water for humans are a promising solution. The CWF-Zn showed up to a maximum of 3.2 log removal of total coliform compared to CWF coated with silver nitrate (CWF-Ag) which showed a maximum of 2.5 log removal. The average log removal efficiencies of the filters (1.07  and 0.95 log removal for CWF-Zn and CWF-plain, respectively) did not meet the required WHO standards. The log removals were low compared to other studies specifically for CWF-Ag. This could be due to the limitation of the coliforms in the influent since this was a field application study. The filters showed a significant difference in E. coli removal efficiency with water from different sources indicating the concentration of the influent influences the performance of the filters. Further optimisation of the CWF-Zn and study over a long period is required. CWF-Zn could be considered a more sustainable and environmentally friendly solution over CWF-Ag.

Date:3 Sep 2019 →  27 Jan 2023
Keywords:Rainwater harvesting, water quality
Disciplines:Water engineering not elsewhere classified
Project type:PhD project