In recent years, we can see a lot of influences on natural environments due to climate change. For example, it has been confirmed that water temperature increases in many public waters in Japan. In this study, we focus on an influence of climate change on riverine water temperature. Tadokoro et al. (2018) has already developed an evaluation way using a single correlation between air temperature and riverine water temperature. But, because riverine water temperature can be determined not only air temperature, but also river discharge, precipitation, solar radiation, and so on, we need to take other effects into accounts. Thus, we try to develop a new evaluation way by using the artificial intelligence (AI) technique. We adapt the neural network (NNW) method for this modeling as an AI technique. In the new evaluation, we use four variables as explanatory variables: air temperature, one hour accumulative precipitation, and amount of global solar radiation, which are measured by Japan Weather Agency (JMA) at AMeDAS stations, and riverine discharge, which is measured by MLIT. Also, riverine water temperature in each rivers was measured by the small data-logging type thermometer in our continuous measurement for more than one year. A model for estimating river water temperature from meteorological condition data was developed using an artificial neural network. The reproducibility of daily mean value by the optimized model was high, however the daily variation had a room for improvement.
|出版ステータス||出版済み - 2020|
|イベント||22nd Congress of the International Association for Hydro-Environment Engineering and Research-Asia Pacific Division: Creating Resilience to Water-Related Challenges, IAHR-APD 2020 - Sapporo, Virtual, 日本|
継続期間: 9 14 2020 → 9 17 2020
|会議||22nd Congress of the International Association for Hydro-Environment Engineering and Research-Asia Pacific Division: Creating Resilience to Water-Related Challenges, IAHR-APD 2020|
|Period||9/14/20 → 9/17/20|
All Science Journal Classification (ASJC) codes