Uncertainty Analysis of Monthly Precipitation in GCMs Using Multiple Bias Correction Methods under Different RCPs

Differences in extremes and uncertainties in future runoff simulations using SWAT and LSTM for SSP scenarios

This study compared the performance of Long Short-Term Memory networks (LSTM) and Soil Water Assessment Tool (SWAT) in simulating observed runoff and projecting future runoff using 11 CMIP6 GCMs. The projected runoff was estimated for two Shared Socioeconomic Pathways (SSPs), 2-4.5 and 5-8.5 for near (2021-2060) and far (2061-2100) futures, respectively. The biases in GCM simulated climate variables were corrected using quantile mapping considering observations at 6 weather stations as reference data over the historical period (1985-2014). Five evaluation metrics were used to quantify the GCM's and hydrological models' capability to reconstruct climate variables and runoff in the Yeongsan Basin of South Korea. Uncertainties in LSTM and SWAT simulated runoff for the historical and projected periods were quantified using Bayesian Model Averaging (BMA) and reliability ensemble averaging (REA), respectively. The results showed significant improvement in bias-corrected GCMs in replicating observations in terms of all evaluation metrics. The extreme runoff estimated using general extreme value (GEV) distribution revealed the better capability of LSTM than SWAT in reproducing observed runoff at all gauging locations. The SWAT projected an increase (17.7%) while LSTM projected a decrease (-13.6%) in the future runoff for both SSPs at most locations. The uncertainty in LSTM simulated runoff was lower than in SWAT runoff at all stations for the historical period. However, the uncertainty in SWAT projected runoff was lower than LSTM projected runoff for both SSPs. This study helps assessing the ability of deep-learning versus physically-based models in hydrological modeling and therefore opens new perspectives for hydrological modeling applications.

The Runoff in the Upper Taohe River Basin and Its Responses to Climate Change

Climate change has a significant impact on water resources. Forecasts and simulations of climate runoff processes are essential for assessing the impact of global climate change on runoff variations. This study focuses on the upper Taohe River Basin, which is an important watershed in the semi-arid regions of northwest China. To assess the runoff in the upper Taohe River Basin and the responses to climate change, the SWAT hydrological model was used to analyze future climate change scenarios and their effects on water resources. The results indicate that the minimum temperature would increase gradually in the 21st century and that the minimum temperature change would be more significant than the maximum temperature change, which indicates that minimum temperature changes would make an obvious contribution to future regional warming. Under RCP2.6, the average precipitation would decrease; at the same time, under RCP4.5 and RCP8.5, the average precipitation would increase. In the future, under different climate scenarios, the runoff will exhibit droughts and flood disasters. These research results provide scientific support for water resource utilization and management in the Taohe River Basin.

Uncertainties in evapotranspiration projections associated with estimation methods and CMIP6 GCMs for South Korea

This study compared the performance capabilities of three potential evapotranspiration (PET) methods, Thornthwaite (TW), Hargreaves and Samani (HS), and Penman-Monteith (PM), to simulate historical and future daily PET levels in South Korea using climate variables from Coupled Model Intercomparison Project 6 (CMIP6) Global Climate Models (GCMs). Five evaluation metrics were used to quantify the reproducibility of the climate variables and PETs at ten stations in South Korea for the historical period used here (1985-2014). The changes and uncertainty associated with the changes in PET in the near (2031-2060) and far (2071-2100) futures were calculated for two shared socioeconomic pathways (SSPs) of 2-4.5 and 5-8.5. As a result, PETs estimated using the three methods for the historical period showed high performance in terms of five evaluation metrics. Overall, PETs showed an increase for both the future periods and the SSPs. The PET estimated using the PM method showed the greatest increase, while that estimated using HS showed the most modest increase in the future. The PM method also showed the highest reliability and lowest uncertainty in the PET estimations, while the opposite was true for HS. This study contributes to our understanding of rational PET methods by which to calculate hydrological factors such as drought indexes for future periods via GCM simulations.

Comparison of Projection in Meteorological and Hydrological Droughts in the Cheongmicheon Watershed for RCP4.5 and SSP2-4.5

Due to the recent appearance of shares socioeconomic pathway (SSP) scenarios, there have been many studies that compare the results between Coupled Model Intercomparison Project (CMIP)5 and CMIP6 general circulation models (GCMs). This study attempted to project future drought characteristics in the Cheongmicheon watershed using SSP2-4.5 of Australian Community Climate and Earth System Simulator-coupled model (ACCESS-CM2) in addition to Representative Concentration Pathway (RCP) 4.5 of ACCESS 1-3 of the same institute. The historical precipitation and temperature data of ACCESS-CM2 were generated better than those of ACCESS 1-3. Two meteorological drought indices, namely, Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI) were used to project meteorological drought while a hydrological drought index, Standardized Streamflow Index (SDI), was used to project the hydrological drought characteristics. The metrological data of GCMs were bias-corrected using quantile mapping method and the streamflow was obtained using Soil and Water Assessment Tool (SWAT) and bias-corrected meteorological data. As a result, there were large differences of drought occurrences and severities between RCP4.5 and SSP2-4.5 for the values of SPI, SPEI, and SDI. The differences in the minimum values of drought index between near (2021–2060) and far futures (2061–2100) were very small in SSP2-4.5, while those in RCP4.5 were very large. In addition, the longest drought period from SDI was the largest because the variation in precipitation usually affects the streamflow with a lag. Therefore, it was concluded that it is important to consider both CMIP5 and CMIP6 GCMs in establishing the drought countermeasures for the future period.

Estimation of Water-Use Rates Based on Hydro-Meteorological Variables Using Deep Belief Network

This study proposed a deep learning-based model to estimate stream water-use rate (WUR) using precipitation (P) and potential evapotranspiration (PET). Correlations were explored to identify relationships among accumulated meteorological variables for various time durations (three-, four-, five-, and six-month cumulative) and WUR, which revealed that three-month cumulative meteorological variables and WUR were highly correlated. A deep belief network (DBN) based on iterating parameter tuning was developed to estimate WUR using P, PET, and antecedent stream water-use rate (DWUR). The training and validation periods were 2011–2016, and 2017–2019, respectively. The results showed that the PET-DWUR based model provided better performances in Nash–Sutcliff efficiency (NSE), root mean square error (RMSE), and determination coefficient (R2) than the P-PET-DWUR and P-DWUR models. The framework in this study can provide a forecast model for deficiencies of stream water use coupled with a weather forecast model.