Search

Previous studies have shown that regional deep learning (DL) models can improve runoff prediction by leveraging large hydrological datasets. However, training a DL regional model using all data without screening may degrade local performance. This study focuses on constructing enhanced local models through the utilization of cross‐basin data. To this end, we propose an approach that employs a novel training strategy to optimize DL model training for specific basins. The approach measures the impact of any one basin's gradient on the loss of the basin of interest, providing insights into the relationships between different basins. The approach was validated using 531 basins from the CAMELS dataset. Results suggest that local performance degradation is a common occurrence in regional models, and imbalanced data are likely to result in a specific pattern dominating the entire regional model. In comparison to a regional model simply trained with all basins, the median Nash‐Sutcliffe efficiency (NSE) for our models is 0.031 higher. In particular, the increase in NSE can exceed 0.2 for some dry basins. Our findings indicate that this novel DL strategy can significantly improve model performance in specific basins using large hydrological datasets, while mitigating local performance loss. Plain Language Summary: In the realm of deep learning, incorporating more data into the training process typically results in a more potent model. Conventionally, large datasets have been employed to train regional models with the intention of predicting rainfall‐runoff processes across all basins. However, such regional models often encounter performance degradation when applied to local basins. This degradation can be attributed to the extraction of overly general features, leading to a loss of specificity. In this paper, our objective is to harness information from a large dataset to establish a more robust local model. We have introduced a method that autonomously learns the similarities in rainfall‐runoff behavior among basins. Subsequently, we utilize this learned similarity to selectively choose data that proves advantageous for training the local model. Our results demonstrate that this strategy can significantly enhance runoff prediction, particularly in arid basins. Key Points: Training a DL regional model using all data without screening may degrade local performanceProper selection of training data is crucial for enhancing DL model training for individual basins, especially arid basinsData from different basins might act as mutual noise during the training process [ABSTRACT FROM AUTHOR]

In northern Ethiopian highlands, climate variability and land use/land cover (LULC) change has been observed during the past decades. However, it remains unclear that to which extent and how these key drivers contributed to changes in hydrological extremes. As a case study, the respective impacts of LULC change and climate variability on hydrological extremes with respect to magnitude, duration, and frequency metrics were assessed in the Upper Tekeze basin, Ethiopia. A scenario‐based approach using LULC and hydro‐climatological data was developed and five simulation experiments were conducted using Soil and Water Assessment Tool model. Results showed that, during 1986–2015, low‐flow duration has decreased by 5.0 days/yr, and the magnitude and occurrence of high flow has increased by 5.23 m3/s/yr and 5.11 days/yr, respectively. Meanwhile, about 47.42% of the landscape has changed its category while no significant change was detected in annual rainfall during the past 31 yr. Particularly, rocky area increased by 82% which was mostly contributed by losses of bare, sand, and low biomass areas. Significant positive and negative contributions to the changes in high and low‐flow extremes, respectively, were detected from LULC change, while no significant contribution was detected from climate change. The increased rock area tends to be one of the major contributors to the increased high flow while climate change seemed to play a minor role. The potential mechanism is the expansion in rock area increased runoff and high flow, which may lead to significant land deterioration and loss of water storage capacity in the mountainous basin. Key Points: Significant increase in high flow events and decrease in low flow events were detected in the Upper Tekeze River basin (UTB)Changes in the flow extremes was largely attributed to land use/land cover change and not significantly influenced by climate changePotential mechanism that increased land degradation led to more streamflow in the UTB [ABSTRACT FROM AUTHOR]

Over the past decades, water scarcity has become prevalent in many regions of the world. Previous water scarcity assessments largely considered "blue water" (surface and groundwater) and overlooked "green water" (soil moisture), despite its key role in supporting crop production. Moreover, the long‐term evolution of water scarcity remains unclear. Here, we propose an agricultural water scarcity index (WSIAW) that explicitly combines green and blue water. The WSIAW measures the degree to which both green and blue water availabilities can satisfy the agricultural water requirement. The WSIAW can directly indicate water scarcity in both rainfed and irrigated areas, which cannot be captured by the blue water scarcity index alone. Using the WSIAW, we assessed the evolution of water scarcity across regions of the world during 1971–2010 at a spatial resolution of 0.5 arc degree. We found that 22% of the 228 river basins with a WSIAW of >1 presented a much broader geographical coverage compared with estimates of water scarcity based on blue water only (8% of total basins). The WSIAW also exhibited a clear trend of increasing severity over the study period, particularly in major agricultural regions. Expanding cropland and increasing/competing water withdrawals by other sectors were the main reasons for the intensification of water scarcity, which was pronounced in some Asian and African river basins. The proposed WSIAW can be applied to different regions with various scales and is important for addressing global water scarcity issues. Key Points: We propose a new agricultural water scarcity index (WSIAW) by explicitly incorporating green and blue water for crop productionWSIAW indicated severe water scarcity over a much broader geographical coverage compared with water scarcity based on blue water onlyWSIAW exhibited an increasing trend in many regions, mainly due to expanding cropland and increasing water withdrawals by other sectors [ABSTRACT FROM AUTHOR]

Spatiotemporal Variation of Drought and Associated Multi-Scale Response to Climate Change over the Yarlung Zangbo River Basin of Qinghai-Tibet Plateau, China. The authors wish to make the following corrections to this paper [[1]]: 1 Li H., Liu L., Shan B.Y., Xu Z.C., Niu Q.K., Cheng L., Liu X.C., Xu Z.X. Spatiotemporal Variation of Drought and Associated Multi-Scale Response to Climate Change over the Yarlung Zangbo River Basin of Qinghai-Tibet Plateau, China. [Extracted from the article]

Drought is one of the most widespread and threatening natural disasters in the world, which has terrible impacts on agricultural irrigation and production, ecological environment, and socioeconomic development. As a critical ecologically fragile area located in southwest China, the Yarlung Zangbo River (YZR) basin is sensitive and vulnerable to climate change and human activities. Hence, this study focused on the YZR basin and attempted to investigate the spatiotemporal variations of drought and associated multi-scale response to climate change based on the scPDSI (self-calibrating Palmer drought severity index) and CRU (climate research unit) data. Results showed that: (1) The YZR basin has experienced an overall wetting process from 1956 to 2015, while a distinct transition period in the mid 1990s (from wet to dry) was detected by multiple statistical methods. (2) Considering the spatial variation of the scPDSI, areas showing the significantly wetting process with increasing scPDSI values were mostly located in the arid upstream and midstream regions, which accounted for over 48% area of the YZR basin, while areas exhibiting the drying tendency with decreasing scPDSI values were mainly concentrated in the humid southern part of the YZR basin, dominating the transition period from wet to dry, to which more attention should be paid. (3) By using the EEMD (ensemble empirical mode decomposition) method, the scPDSI over the YZR basin showed quasi-3-year and quasi-9-year cycles at the inter-annual scale, while quasi-15-year and quasi-56-year cycles were detected at the inter-decadal scale. The reconstructed inter-annual scale showed a better capability to represent the abrupt change characteristic of drought, which was also more influential to the original time series with a variance contribution of 55.3%, while the inter-decadal scale could be used to portray the long-term drought variation process with a relative lower variance contribution of 29.1%. (4) The multi-scale response of drought to climate change indicated that changes of precipitation (PRE) and diurnal temperature range (DTR) were the major driving factors in the drought variation at different time scales. Compared with potential evapotranspiration (PET), DTR was a much more important climate factor associated with drought variations by altering the energy balance, which is more obvious over the YZR basin distributed with extensive snow cover and glaciers. These findings could provide important implications for ecological environment protection and sustainable socioeconomic development in the YZR basin and other high mountain regions. [ABSTRACT FROM AUTHOR]

Summary Freshwater ecoregion is currently widely used by biologists, conservators and resource managers. Most of ecoregion delineations are developed at the basin scale and are not fully adapted in a practical manner because operational water resources management is primarily conducted by political administrative departments. In this study, an ecoregion delineation framework was proposed to classify three-level ecoregions in Jinan City with geographic information systems and cluster analysis. The first level ecoregion was composed of three watersheds (a part of the Yellow River, Xiaoqing River and Tuhaimajia River) plus the urban area, which was primarily determined on the basis of the city administrative divisions and river watersheds. The classification of the second level ecoregion is primarily based on the spatial heterogeneity of land use. The third level ecoregion was delineated for each second level ecoregion by using the cluster analysis on water quality. At the same time, administrative boundaries were used to rectify the boundaries of each ecoregion in this study to facilitate the administration of each ecoregion. Furthermore, ecological health assessment (IBI) based on fish communities were employed to validate the freshwater ecoregion. The results demonstrated that 73.3% of ecoregions were in line with the distribution of fish IBI, indicating that the freshwater ecoregions are acceptable for future water resources management. [ABSTRACT FROM AUTHOR]

[Display omitted] • Precipitation is the most important factor for water yield during the wet season. • Vegetation is more decisive at predicting water yield during the dry season. • Vegetation greening has the potential to increase drought resilience. Water yield (WY) in the Upper Brahmaputra River (UBR) basin is important for sustaining the ecological environment in the headstream region and supplying valuable freshwater resources downstream. While recent studies indicated the presence of warming and greening trends in the region, the effects of these changes on WY are not yet understood. Here, on the basis of the Budyko framework and a random forest model, we assess the relative importance of climate and vegetation on WY changes in the UBR basin. We find that precipitation contributes to over 90% of the increase in annual WY, and is also the most important factor for WY during the wet season. During the dry season, however, vegetation, rather than precipitation, predicts the variability in WY. We suggest that vegetation may have changed the seasonality of water by storing water in the wet periods and releasing it in dry periods, which explains why WY increases significantly during the dry season. Our findings imply that vegetation greening in the UBR basin has the potential to increase drought resilience downstream, and protect food security from drought, highlighting the value of ecological restoration for water resource management. [ABSTRACT FROM AUTHOR]

• Climatic and reservoir impacts were quantified using a hydrological-reservoir model. • Reservoir strongly alter the streamflow in the period of 2008–2016 in the LMRB. • Upstream reservoir mainly affected the streamflow of the midstream stations. • Reservoir operation has reduced the flood risk increased by climatic change. The Lancang-Mekong River Basin (LMRB) is one of the most important transboundary river basins in Asia. While climate change perturbs the streamflow and affects flood events, reservoir operation may mitigate or aggravate this impact. Therefore, quantitative assessment of the climate change impact and reservoir effect on the LMRB is a vital prerequisite for future hydropower development and environmental protection. This study aimed to estimate the variation of the streamflow and flood characteristics affected by climate change and reservoir operation within the LMRB. A reservoir module was incorporated into the Variable Infiltration Capacity (VIC) model to simulate the streamflow susceptible to the reservoirs. It was found that the reservoirs had a substantial influence on the streamflow during 2008–2016, when many reservoirs were constructed in the LMRB. The reservoirs across the Lancang River (the upper Mekong River located in China) reduced the annual average streamflow by 5% at Chiang Sean station (northern Thailand) in 2008–2016, whereas their influence became undetectable downstream of Vientiane station (northern Laos). The streamflow changes downstream of Mukdahan station at southern Laos (including the stations in Cambodia and southern Vietnam) were mainly attributed to the local reservoirs and climate change. Compared with the baseline period of 1985–2007, the upstream reservoir operation dramatically affected streamflow at the midstream stations with higher dry season streamflow (+15% to +37%), but lower wet season streamflow was less affected (−2% to −24%) in 2008–2016. Climate change increased the magnitude and frequency of the flood by up to 14% and 45%, respectively, whereas the reservoir operation reduced them by 16% and 36%, respectively. Our findings provide insights into the interaction between climate change and reservoir operation and their integrated effects on the streamflow, informing and supporting water management and hydropower development in the LMRB. [ABSTRACT FROM AUTHOR]