Abstract:
Modeling the relationships between surface water quality with rainfall and land-use patterns is
of prime importance in exploring effective method s for mitigating the water pollution in areas
like Colombo city. These models are important for land-use planning and resource management
to plan for best environmental management practices and it is possible to compile detailed
inventories and to monitor water quality in surface water bodies. This study focuses on
modeling the surface water quality change s of urban canals in Colombo from 2003 to 2009 in
relation to rainfall of different climatic seasons, land use types, and population density using a
G IS based approach together with statistical modeling. The data on land use, monthly rainfall,
population density and water quality from 200 3 to 200 9 was collected from the Department of
Survey, Department of Meteorology, Department of Census & Statistics and Sri Lanka Land
Reclamation & Development Company respectively. Results revealed that there is a strong
significant difference within the mea n values of Conductivity, Chemical Oxygen Demand
(COD), Biological Oxygen Demand (BOD) and Salinity percentage according to the climatic
seasons. The water quality attributes; Conductivity, Phosphate , BO D and Salinity percentage
were significantly correlated to land-use types (Built-up/Homesteads, Barren, Marsh/Paddy,
Park/Playground, Plantation, Scrub and Water bodies). A significant regression model could
be drawn to predict the surface water quality in Colombo city using one water quality indicator
(BOD) which ma y suggest that point-sources contribute more pollutants than non-point
sources. Recommendation s were mad e to consider more important water quality parameters
such as biological parameters within an extended study area for future studies because they
will be providing complex models of water quality with other factors.
