The change of groundwater levels after the 2016 Mw 7.0 Kumamoto crustal earthquake was evaluated using a simple conceptual hydrological model in an attempt to show the presence, intensity, and probable mechanism of water level rise observed in Kumamoto where a comprehensive observation-well network exists. A tank model was applied to verify 16 wells in the study field. In the model groundwater levels were first calibrated for the periods in ca. 2 years before the main shock using several hydrological parameters including precipitation, evapotranspiration, water recharge and discharge, and artificial recharge by irrigation. Water levels were then simulated by extrapolating this law of water fluctuating patterns for ca. 2.5 years after the main shock of the earthquake, without considering hydrogeological changes due to the earthquake. A difference in groundwater levels between observation and simulation results yields a degree of coseismic water level rises for each well. The coseismic abnormal water level increase was calculated to be ~11 m in 4–5 month after the main shock and was most significantly on the western slope of the Aso caldera rim mountains. The spatial distribution of the coseismic water increases clarified that the most dominate increasing anomalies prevail at mountain feet surrounding the plains, suggesting the occurrence of coseismic mountain water release resulting in the rise of water levels in downslope aquifers. Identified coseismic water level increases still continue up to 2.5 years after the earthquake, probably because changes in hydrogeological properties in mountain aquifers, i.e., permeability, are still sustained. Our forecasting water recovering trends require ca. 3.5–5 year after the earthquake for complete recovery to the original conditions. We demonstrated that our approaches are capable of describing coseismic water level changes and could potentially be applied to other fields.
ASJC Scopus subject areas