Forecasting Municipal Water Demands: Evaluating the Impacts of Population Growth, Climate Change, and Conservation Policies on Water End-Use

Abstract

Urban population growth and climate change have heightened concerns about the sustainability of municipal water supplies and the effectiveness of current management strategies. This study introduces a hybrid model, the Edmonton Water Demand Simulator (EWDS), that integrates system dynamics, artificial neural networks, and regression techniques to forecast long-term water demand and support planning for water utilities and governments. Using historical data for the Edmonton region, the EWDS accurately simulates past municipal and end-use water demands and provides predictions under different scenarios through 2100. The study evaluates water demand variability under scenarios involving climate change, population growth, and conservation policies, and analyzes their relative and combined effects. Under high population growth and climate change without additional conservation efforts, water demand is projected to double by 2066. However, lower growth and conservation measures could delay this by 30 years. Annual water demands range from 3.3–3.7 × 10⁵ ML, 3.4–4.1 × 10⁵ ML, and 2.9–3.4 × 10⁵ ML for population growth, climate change, and conservation policies, respectively. Population growth poses the greatest challenge, while conservation policies effectively curb demands.  The EWDS offers a transferable tool for projecting water demand in other communities and can help to support sustainable resource management.

Keywords: Municipal water demand, climate change, Artificial Neural Networks, system dynamics