Framework for Feasibility Assesment and Performance Analysis of Riverbank Filtration Systems for Water Treatment

Saroj K. Sharma, Daniel Chaweza, Nelson Bosuben, Ekkehard Holzbecher and Gary Amy
UNESCO-IHE Institute for Water Education, P. O. Box 3015, 2601 DA, Delft, The Netherlands. Email: s.sharma@unesco-ihe.org, Tel: +31 15 2151772
ABSTRACT

Riverbank Filtration (RBF) is a reliable and proven natural water treatment technology, in which surface water contaminants are removed or degraded as water moves through the soil/aquifer to a recovery well(s). Because of its ability to remove even the most persistent contaminants and microbes, RBF can support or even replace other treatment processes in a water treatment scheme. RBF, however, is site specific and requires extensive site investigations and pilot studies to assess its feasibility based local conditions. The goal of this study was to develop a framework to assess the feasibility of RBF for water treatment under given local (site specific) conditions.

Initially, literature reviews were conducted to compile the removal rates of four main water quality parameters namely (i) bulk organic matter, (ii) trace organic compounds, (iii) nitrogen species and (iv) microbes. Furthermore, changes in other water quality parameters such as pH, O2, Fe, and Mn during RBF were also examined. Based on literature studies, removal rates of these main parameters with respect to travel time and/or distance were established.

A four step methodology has been developed for feasibility assessment of RBF under given conditions. Firstly, a hydraulic model MODFLOW is used to determine the travel time and pumping yield corresponding to given location of extraction wells. Furthermore, the number of the (vertical) wells required for a given production capacity and their spacing (from source and between wells) is determined. Secondly, the NASRI Bank Filtration Simulator is used to compute the share of bank filtrate and local groundwater for each well location. Both of these hydraulic models can be used to assess different scenarios by changing variables such as number of wells, well spacing, and well distance from riverbank, hydraulic conductivities, and initial hydraulic head. Thirdly, water quality from the RBF system is predicted using the Water Quality Prediction Algorithms developed based on the literature review. Finally, the capital and O&M costs of the RBF system are compared with that of existing conventional water treatment systems.

This assessment method was applied for feasibility assessment of BF in five cities in Africa; three cities in Malawi; Blantyre, Lilongwe, and Mzuzu; and two cities in Kenya; Eldoret and Nakuru. In most of these water supply systems surface water (river) is the main source of water and conventional surface water treatment (sedimentation, chemical coagulation/flocculation, clarification and chlorination) is applied. Feasibility studies revealed that in all of the five cities, the hydrogeological conditions are favourable for BF and with the proper design of the production wells, existing water demand can be met with the BF systems. Analysis in Malawi showed that by switching to BF, savings of over 80% on existing annual treatment costs (chemical and energy only) are likely for Blantyre and Lilongwe cities, but for Mzuzu the annual costs of BF and existing conventional water treatment are comparable. In Eldoret and Nakuru cities in Kenya, annual operational costs savings of about 16% and 32%, respectively, could be achieved by switching from conventional surface water treatment to a BF system.