Projected changes of groundwater levels in northeastern Romania according to climate scenarios for 2020–2100

Abstract

Study regionA plain region covering 8000 km2 in extreme north-eastern Romania in which two groundwater levels are identified, near the surface (1.5 m) and in-depth (4.5 m) respectively. Study focusThe statistical relationship between climate conditions and underground water resources was assessed by means of correlation analysis and autoregressive neural networks (ANNs). Based on this, the study explores the quantitative changes in groundwater levels towards the end of this century according to most relevant climate scenarios regarding the evolution of air temperature and precipitation amount. New hydrological insights for the regionThe results of the ANNs model show that air temperature together with precipitation amount explain between 50% and 70% of the groundwater levels variability for 1983–2017 interval. For local conditions instead, analysed at the level of main regional weather stations, the explained variability by the climate conditions decreases between 40% and 60%. This overall good statistical performance enables us to assimilate into the model the climate scenarios data for the region. The groundwater projections indicate that the decline in groundwater levels is expected to be general in both climate scenarios analysed. Thus, the impact of climate change on groundwater supply is expected to be very high over the study region, adding to the onset effect of diminishing water quality, raising a great risk on water availability for natural and socio-economic environment.