Digital Elevation Models (DEMs) have been successfully used in a large range of environmental issues. Several methods such as digital contour interpolation and remote sensing have allowed the generation of DEMs, some of which are now freely available for almost the entire globe. The Soil and Water Assessment Tool (SWAT) is a widely used semi-distributed model operating at the watershed level and has previously been shown to be very sensitive to the quality of the input topographic information. The objective of this study was to evaluate the impact of DEMs generated from different data sources, respectively DLG5m (local Digital Line Graph, 5 m interval), ASTER30m (1 arc-s ASTER Global DEM Version 1, approximately 30 m resolution), and SRTM90m (3 arc-s SRTM Version 4, approximately 90 m resolution), on SWAT predictions for runoff, sediment, total phosphor (TP) and total nitrogen (TN). Eleven resolutions, from 5 m to 140 m, were considered in this study. Results indicate that the predictions of TPs and TNs decreased substantially with coarser resampled resolution. Slightly decreased trends could be found in the predicted sediments when DEMs were resampled to coarser resolutions. Predicted runoffs were not sensitive to resampled resolutions. The predicted outputs based on DLG5m were more sensitive to resampled resolutions than those based on ASTER30m and SRTM90m. At original resolutions, the predicted outputs based on ASTER30m and SRTM90m were similar, but the predicted TNs and TPs based on ASTER30m and SRTM90m were much lower than the one based on DLG5m. For the predicted TNs and TPs, which were substantially sensitive to DEM resolutions, the output accuracies of SWAT derived from ASTER30m and SRTM90m could be improved by down-scaled resampling, but they could not improve on finer DEM (DLG5m) at the same resolution. This study helps GIS environmental model users to understand the sensitivities of SWAT to DEM resolution, and choose feasible DEM data for environmental models

Abstract. Streamflow forecasts are traditionally effective in mitigating water scarcity and flood defense. This study developed an artificial intelligence (AI)-based management methodology that integrated multi-step streamflow forecasts and multi-objective reservoir operation optimization for water resource allocation. Following the methodology, we aimed to assess forecast quality and forecast-informed reservoir operation performance together due to the influence of inflow forecast uncertainty. Varying combinations of climate and hydrological variables were input into three AI-based models, namely a long short-term memory (LSTM), a gated recurrent unit (GRU), and a least-squares support vector machine (LSSVM), to forecast short-term streamflow. Based on three deterministic forecasts, the stochastic inflow scenarios were further developed using Bayesian model averaging (BMA) for quantifying uncertainty. The forecasting scheme was further coupled with a multi-reservoir optimization model, and the multi-objective programming was solved using the parameterized multi-objective robust decision-making (MORDM) approach. The AI-based management framework was applied and demonstrated over a multi-reservoir system (25 reservoirs) in the Zhoushan Islands, China. Three main conclusions were drawn from this study: (1) GRU and LSTM performed equally well on streamflow forecasts, and GRU might be the preferred method over LSTM, given that it had simpler structures and less modeling time; (2) higher forecast performance could lead to improved reservoir operation, while uncertain forecasts were more valuable than deterministic forecasts, regarding two performance metrics, i.e., water supply reliability and operating costs; (3) the relationship between the forecast horizon and reservoir operation was complex and depended on the operating configurations (forecast quality and uncertainty) and performance measures. This study reinforces the potential of an AI-based stochastic streamflow forecasting scheme to seek robust strategies under uncertainty.

Investigation of the role of multiple general circulation model (GCM) ensembles in obtaining comprehensive knowledge of hydrological responses across the Yellow River Basin (YRB), China, is still of substantial importance. This study evaluates the performance of the Coupled Model Intercomparison Project Phase 6 (CMIP6) models in simulating the hydrological regime in the YRB and compares the results with those from CMIP 5 (CMIP5). The comparison is performed between 21 GCMs from CMIP6 under three Shared Socioeconomic Pathway scenarios and 18 GCMs from CMIP5 under three Representative Concentration Pathway scenarios. Raw CMIP outputs are first corrected and downscaled by the Bias Correction and Spatial Disaggregation methods, and the bias-corrected GCM outputs are then employed to drive the Soil and Water Assessment Tool hydrological model and project streamflow. After correction and downscaling, areal averages for future changes (relative to 1971–2000) of temperature and precipitation are found larger in CMIP6 than in CMIP5. The emblematic annual mean temperature of CMIP6 increases by 1.64–2.20 and 2.31–5.29 °C for the future period of 2026–2055 and 2066–2095, while the counterpart of CMIP5 is 1.92–2.39 and 1.68–4.76 °C, respectively. In terms of precipitation, for CMIP6, it increases by 3.45–4.70 and 6.77–15.40%, and for CMIP5 by 2.58–2.96 and 3.83–9.95%. It is further concluded that: (1) future streamflow will probably decrease less under CMIP6 than that under CMIP5 in most cases, and climate changes of this kind will affect regional water supply and security in the YRB; (2) uncertainty in the projected streamflow is dominated by GCMs uncertainty with the contribution rate of >75%; (3) the streamflow is more sensitive to precipitation changes in comparison with temperature changes in the near future. In contrast, streamflow reduction is more attributed to an increase in temperature with a contribution rate of almost >60% than in precipitation in the far future.

No abstract

No abstract