CMIP5 기반 하천유량 예측을 위한 딥러닝 LSTM 모형의 최적 학습기간 산정

초록

In this study, we suggested the optimal training period for predicting the streamflow using the LSTM (Long Short-Term Memory) model based on thedeep learning and CMIP5 (The fifth phase of the Couple Model Intercomparison Project) future climate scenarios. To validate the model performanceof LSTM, the Jinan-gun (Seongsan-ri) site was selected in this study. We comfirmed that the LSTM-based streamflow was highly comparable to themeasurements during the calibration (2000 to 2002/2014 to 2015) and validation (2003 to 2005/2016 to 2017) periods. Additionally, we compared theLSTM-based streamflow to the SWAT-based output during the calibration (2000∼2015) and validation (2016∼2019) periods. The results supportedthat the LSTM model also performed well in simulating streamflow during the long-term period, although small uncertainties exist. Then theSWAT-based daily streamflow was forecasted using the CMIP5 climate scenario forcing data in 2011∼2100. We tested and determined the optimaltraining period for the LSTM model by comparing the LSTM-/SWAT-based streamflow with various scenarios. Note that the SWAT-based streamflowvalues were assumed as the observation because of no measurements in future (2011∼2100). Our results showed that the LSTM-based streamflow wassimilar to the SWAT-based streamflow when the training data over the 30 years were used. These findings indicated that training periods more than30 years were required to obtain LSTM-based reliable streamflow forecasts using climate change scenarios.

키워드