An Iterative Dynamic Model for Simulating the Pollutant Distributions in the Tank Cascade Ecosystems of Sri Lanka

Sri Lanka is a tropical island located in the Indian Ocean that consists of two distinct climatic zones: the dry zone and the wet zone. Rain water scarcity of the dry zone for agricultural use has resulted in the development of man-made tank systems referred to as tank cascade systems (TCS). Over the years, these systems have integrated with the natural environment of the dry zone. Recently, agrochemicals and other pollutants have come in contact with these traditional water-preserving TCSs, leading to health and environmental concerns. These systems are studied in qualitative ways in the literature. However, there is a lack of mathematical models to capture, predict, and better understand the dynamics of TCS pollution. Therefore, in this work, for the first time, a time-dependent mathematical model for a TCS is developed and applied to study pollutant dynamics. The mathematical model is developed by identifying key ecological components in a generic TCS and applying time dependent water and pollutant balance equations over these components. The equations are then numerically solved and simulations are conducted using the Xcos simulation environment of the open-source software Scilab. The water volume and pollutant concentration variations in the Kadirgama TCS are studied. The simulation results are then compared against available measured data from the literature. The simulation results indicate that the water storage in different TCS components, ranging from 30,000 to 60,000 m 3 in the Nawakkulama system and 100,000 to 300,000 m 3 in the Kadiragama system facilitate the ecological functioning of the component as well as the functioning of the entire TCS. Pollutant concentration peaks ranging from 0.4 to 7 kg m −3 are observed during times of low water availability in some components of the system. The results also indicated a system dynamic time scale in the order of several months, which demonstrated that the TCS are long residence time systems built to preserve water to cope with the water scarcity in the dry zone. The developed model in the present work provides a tool to study the pollutant dynamics in various ecological components of a TCS under measured or user given input data. It can therefore be used to identify the effects of modern agricultural and environmental decisions on the functioning of these components, therefore providing a tool to aid in the decision-making processes related to the TCS systems.