Search for: All records

Accurate hydrologic modeling is vital to characterizing how the terrestrial water cycle responds to climate change. Pure deep learning (DL) models have been shown to outperform process-based ones while remaining difficult to interpret. More recently, differentiable physics-informed machine learning models with a physical backbone can systematically integrate physical equations and DL, predicting untrained variables and processes with high performance. However, it is unclear if such models are competitive for global-scale applications with a simple backbone. Therefore, we use – for the first time at this scale – differentiable hydrologic models (full name δHBV-globe1.0-hydroDL, shortened to δHBV here) to simulate the rainfall–runoff processes for 3753 basins around the world. Moreover, we compare the δHBV models to a purely data-driven long short-term memory (LSTM) model to examine their strengths and limitations. Both LSTM and the δHBV models provide competitive daily hydrologic simulation capabilities in global basins, with median Kling–Gupta efficiency values close to or higher than 0.7 (and 0.78 with LSTM for a subset of 1675 basins with long-term discharge records), significantly outperforming traditional models. Moreover, regionalized differentiable models demonstrated stronger spatial generalization ability (median KGE 0.64) than a traditional parameter regionalization approach (median KGE 0.46) and even LSTM for ungauged region tests across continents. Nevertheless, relative to LSTM, the differentiable model was hampered by structural deficiencies for cold or polar regions, highly arid regions, and basins with significant human impacts. This study also sets the benchmark for hydrologic estimates around the world and builds a foundation for improving global hydrologic simulations. 

Abstract Global water models are increasingly used to understand past, present and future water cycles, but disagreements between simulated variables make model-based inferences uncertain. Although there is empirical evidence of different large-scale relationships in hydrology, these relationships are rarely considered in model evaluation. Here we evaluate global water models using functional relationships that capture the spatial co-variability of forcing variables (precipitation, net radiation) and key response variables (actual evapotranspiration, groundwater recharge, total runoff). Results show strong disagreement in both shape and strength of model-based functional relationships, especially for groundwater recharge. Empirical and theory-derived functional relationships show varying agreements with models, indicating that our process understanding is particularly uncertain for energy balance processes, groundwater recharge processes and in dry and/or cold regions. Functional relationships offer great potential for model evaluation and an opportunity for fundamental advances in global hydrology and Earth system research in general. 

Abstract Droughts that exceed the magnitudes of historical variation ranges could occur increasingly frequently under future climate conditions. However, the time of the emergence of unprecedented drought conditions under climate change has rarely been examined. Here, using multimodel hydrological simulations, we investigate the changes in the frequency of hydrological drought (defined as abnormally low river discharge) under high and low greenhouse gas concentration scenarios and existing water resource management measures and estimate the time of the first emergence of unprecedented regional drought conditions centered on the low-flow season. The times are detected for several subcontinental-scale regions, and three regions, namely, Southwestern South America, Mediterranean Europe, and Northern Africa, exhibit particularly robust results under the high-emission scenario. These three regions are expected to confront unprecedented conditions within the next 30 years with a high likelihood regardless of the emission scenarios. In addition, the results obtained herein demonstrate the benefits of the lower-emission pathway in reducing the likelihood of emergence. The Paris Agreement goals are shown to be effective in reducing the likelihood to the unlikely level in most regions. However, appropriate and prior adaptation measures are considered indispensable when facing unprecedented drought conditions. The results of this study underscore the importance of improving drought preparedness within the considered time horizons. 

Abstract Droughts are anticipated to intensify in many parts of the world due to climate change. However, the issue of drought definition, namely the diversity of drought indices, makes it difficult to compare drought assessments. This issue is widely known, but its relative importance has never been quantitatively evaluated in comparison to other sources of uncertainty. Here, encompassing three drought categories (meteorological, agricultural, and hydrological droughts) with four temporal scales of interest, we evaluated changes in the drought frequency using multi-model and multi-scenario simulations to identify areas where the definition issue could result in pronounced uncertainties and to what extent. We investigated the disagreement in the signs of changes between drought definitions and decomposed the variance into four main factors: drought definitions, greenhouse gas concentration scenarios, global climate models, and global water models, as well as their interactions. The results show that models were the primary sources of variance over 82% of the global land area. On the other hand, the drought definition was the dominant source of variance in the remaining 17%, especially in parts of northern high-latitudes. Our results highlight specific regions where differences in drought definitions result in a large spread among projections, including areas showing opposite signs of significant changes. At a global scale, 7% of the variance resulted independently from the definition issue, and that value increased to 44% when 1st and 2nd order interactions were considered. The quantitative results suggest that by clarifying hydrological processes or sectors of interest, one could avoid these uncertainties in drought assessments to obtain a clearer picture of future drought change. 

Abstract. We develop a new large-scale hydrological and water resources model, theCommunity Water Model (CWatM), which can simulate hydrology both globallyand regionally at different resolutions from 30 arcmin to 30 arcsec atdaily time steps. CWatM is open source in the Python programming environmentand has a modular structure. It uses global, freely available data in thenetCDF4 file format for reading, storage, and production of data in acompact way. CWatM includes general surface and groundwater hydrologicalprocesses but also takes into account human activities, such as water useand reservoir regulation, by calculating water demands, water use, andreturn flows. Reservoirs and lakes are included in the model scheme. CWatMis used in the framework of the Inter-Sectoral Impact Model IntercomparisonProject (ISIMIP), which compares global model outputs. The flexible modelstructure allows for dynamic interaction with hydro-economic and water qualitymodels for the assessment and evaluation of water management options.Furthermore, the novelty of CWatM is its combination of state-of-the-arthydrological modeling, modular programming, an online user manual andautomatic source code documentation, global and regional assessments atdifferent spatial resolutions, and a potential community to add to, change,and expand the open-source project. CWatM also strives to build a communitylearning environment which is able to freely use an open-source hydrologicalmodel and flexible coupling possibilities to other sectoral models, such asenergy and agriculture.