Quantitative assessment of the changes in regional water flow systems caused by the 2016 Kumamoto Earthquake using numerical modeling

Coseismic changes in water levels and discharges are well-known phenomena worldwide. Reliable hypotheses to explain these hydrological changes qualitatively have been proposed using various data and analyses. However, there are few quantitative assessments of these changes, which require hydrological simulators. A dense monitoring network revealed coseismic river and groundwater level changes induced by the 2016 M-w 7.0 Kumamoto crustal earthquake associated with many post-seismic fracture systems. In this study, we reproduced these changes using a physically based, fully distributed, integrated watershed modeling tool (GETFLOWS). In particular, the formation of new fracture systems is key to understanding the hydrological changes observed after the earthquake, which were incorporated into our model by changing parameters. Previous studies suggested that the observed water level drop immediately after the earthquake is explained by the drawdown of surface and aquifer waters to fill open spaces in the upper crust along the fracture systems generated under extensional stress field. After the initial water drop, water levels rose in the long-term because additional water sources were released from surrounding mountains due to the coseismic increase in the permeability of the soils and rocks around the compressional fracture systems. These hypotheses were validated in our model by adjusting the depths of fracture systems. Consequently, we constructed the three-dimensional distribution of these property changes over the study area. Furthermore, the model calculation yielded the volumes of missing and released mountain waters in the studied watershed (ca. 10(6) and 10(8) m(3), respectively). This study demonstrates the utility of an integrated watershed modeling tool for investigating coseismic hydrological changes in active hydrological systems quantitatively.