This article examines the short- and long-run effects of water price, system input, income, temperature on domestic water demand for Amman area over the period of 1980–2012. An empirical, dynamic autoregressive distributed lag (ARDL) model for water demand is developed on a yearly basis. This approach is capable of testing and analysing the dynamic relationship with time series data using a single equation regressions. Results show the ability of the model to predicting future trends (short- and long-run association). The main results indicate that water demand in limited water environment is partially captured in the long-run by the amount of water reaching the customer. The short- and long-run elasticities of water price (–0.061, –0.028) and high temperature (0.023, 0.054) indicate inelastic behaviour on water demand both in short- and long-run, while the lagged water price has a significant effect on demand. Income represented by gross domestic product (GDP) slightly affects water consumption in the long-run and insignificantly in the short-run (0.24, 0.24). Water consumption is strongly linked to consumption habits measured by lagged billed amount 0.35, and is strongly linked to amount of supplied water both in short- and long-run (0.47, 0.53). These results suggest that water needs should be satisfied first to allow controlling water demand through a good pricing system.
Moreover, the association identified between demand and water system input, and the lesser elasticities of water price and other explanatory variables confirm the condition of water deficit in Amman area and Jordan. The results could be rolled out to similar cities suffering scarce water resources with arid and semi-arid weather conditions.

Exploring the drivers of changes in ecosystem services is crucial to maintain ecosystem functionality, especially in the diverse Central Citarum watershed. This study utilises the integrated valuation of ecosystem service and trade-offs (InVEST) model and multiscale geographically weighted regression (MGWR) model to examine ecosystem services patterns from 2006 to 2018. The InVEST is a hydrological model to calculate water availability and evaluate benefits provided by nature through simulating alterations in the amount of water yields driven by land use/cover changes. Economic, topographic, climate, and vegetation factors are considered, with an emphasis on their essential components. The presence of a geographical link between dependent and explanatory variables was investigated using a multiscale geographic weighted regression model. The MGWR model is employed to analyse spatial impacts. The integration of both models simplified the process and enhanced its understanding. The findings reveal the following patterns: 1) decreasing land cover and increasing ecosystem services demand in the watershed, along with a decline in water yield, e.g. certain sub-districts encounter water scarcity, while others have abundant water resources; 2) the impact of natural factors on water yield shifts along vegetation > climate > topography (2006) changes to climate > vegetation > topography (2018).