Search

Although global- and catchment-scale hydrological models are often shown to accurately simulate long-term runoff time-series, far less is known about their suitability for capturing hydrological extremes, such as droughts. Here we evaluated simulations of hydrological droughts from nine catchment scale hydrological models (CHMs) and eight global scale hydrological models (GHMs) for eight large catchments: Upper Amazon, Lena, Upper Mississippi, Upper Niger, Rhine, Tagus, Upper Yangtze and Upper Yellow. The simulations were conducted within the framework of phase 2a of the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP2a). We evaluated the ability of the CHMs, GHMs and their respective ensemble means (Ens-CHM and Ens-GHM) to simulate observed hydrological droughts of at least one month duration, over 31 years (1971–2001). Hydrological drought events were identified from runoff-deficits and the Standardised Runoff Index (SRI). In all catchments, the CHMs performed relatively better than the GHMs, for simulating monthly runoff-deficits. The number of drought events identified under different drought categories (i.e. SRI values of -1 to -1.49, -1.5 to -1.99, and ≤-2) varied significantly between models. All the models, as well as the two ensemble means, have limited abilities to accurately simulate drought events in all eight catchments, in terms of their occurrence and magnitude. Overall, there are opportunities to improve both CHMs and GHMs for better characterisation of hydrological droughts.

This study presents a novel application of machine learning to deliver optimised, multi-model combinations (MMCs) of Global Hydrological Model (GHM) simulations. We exemplify the approach using runoff simulations from five GHMs across 40 large global catchments. The benchmarked, median performance gain of the MMC solutions is 45% compared to the best performing GHM and exceeds 100% when compared to the EM. The performance gain offered by MMC suggests that future multimodel applications consider reporting MMCs, alongside the EM and intermodal range, to provide endusers of GHM ensembles with a better contextualised estimate of runoff. Importantly, the study highlights the difficulty of interpreting complex, non-linear MMC solutions in physical terms. This indicates that a pragmatic approach to future MMC studies based on machine learning methods is required, in which the allowable solution complexity is carefully constrained.

Global-scale hydrological models are routinely used to assess water scarcity, flood hazards and droughts worldwide. Recent efforts to incorporate anthropogenic activities in these models have enabled more realistic comparisons with observations. Here we evaluate simulations from an ensemble of six models participating in the second phase of the Inter-Sectoral Impact Model Inter-comparison Project (ISIMIP2a). We simulate observed monthly runoff in 40 catchments, spatially distributed across 8 global hydrobelts. The performance of each model and the ensemble mean is examined with respect to their ability to replicate observed mean and extreme runoff under human-influenced conditions. Application ofa novel integrated evaluation metric to quantify the models’ ability to simulate timeseries of monthly runoff suggests that the models generally perform better in the wetter equatorial and northern hydrobelts than in drier southern hydrobelts. When model outputs are temporally aggregated to assess mean annual and extreme runoff, the models perform better. Nevertheless, we find a general trend in the majority of models towards the overestimation of mean annual runoff and allindicators of upper and lower extreme runoff. There are particular challenges associated with reproducing both the timing and magnitude of seasonal cycles; the models struggle to capture the timing of the seasonal cycle, particularly in northern hydrobelts, while in southern hydrobelts the models struggle to reproduce the magnitude of the seasonal cycle. It is noteworthy that over all hydrological indicators, the ensemble mean fails to perform better than any individual model – afinding that challenges the commonly held perception that model ensemble estimates deliver superior performance over individual models. The study highlights the need for continued model development and improvement. It also suggests that caution should be taken when summarising the simulations from a model ensemble based upon its mean output.

Climate change impacts on water availability and hydrological extremes are major concerns as regards the Sustainable Development Goals. Impacts on hydrology are normally investigated as part of a modelling chain, in which climate projections from multiple climate models are used as inputs to multiple impact models, under different greenhouse gas emissions scenarios, which are resulting in different amounts of global temperature rise. While the goal is generally to investigate the relevance of changes in climate for the water cycle, water resources or hydrological extremes, it is often the case that variations in other components of the model chain obscure the effect of climate scenario variation. This is particularly important when assessing the impacts of relatively lower magnitudes of global warming, such as those associated with the aspirational goals of the Paris Agreement. In our study, we use ANOVA (ANalyses Of VAriance) to allocate and quantify the main sources of uncertainty in the hydrological impact modelling chain. In turn we determine the statistical significance of different sources of uncertainty. We achieve this by using a set of 5 climate models and up to 13 hydrological models, for 9 large scale river basins across the globe, under 4 emissions scenarios. The impact variable we consider in our analysis is daily river discharge. We analyze overall water availability and flow regime, including seasonality, high flows and low flows. Scaling effects are investigated by separately looking at discharge generated by global and regional hydrological models respectively. Finally, we compare our results with other recently published studies. We find that small differences in global temperature rise associated with some emissions scenarios have mostly significant impacts on river discharge – however, climate model related uncertainty is so large that it obscures the sensitivity of the hydrological system.

We investigated dam behaviours during high-flow events and their robustness against perturbations in meteorological conditions using the H08 global hydrological model. Differences in these behaviours were examined by comparing simulation runs, with and without dams and using multiple meteorological datasets, at a case-study site, Fort Peck Dam on the Missouri River, USA. The results demonstrated that dam-regulated river flow reduced temporal variability over large time periods and also dampened inter-forcing discrepancies in river discharge (smoothing effects). However, during wet years, differences in peak flow were accentuated downstream of the dam, resulting in divergence in simulated peak flow across the meteorological forcing (pulsing effect). The pulsing effect was detected at other major dams in global simulations. Depending upon the meteorological forcing, the dams act as a selective filter against high-flow events. Synergy between a generic dam scheme and differences in meteorological forcing data might introduce additional uncertainties in global hydrological simulations.

We performed a twofold intercomparison of river discharge regulated by dams under multiple meteorological forcings among multiple global hydrological models for a historical period by simulation. Paper II provides an intercomparison of river discharge simulated by five hydrological models under four meteorological forcings. This is the first global multimodel intercomparison study on dam-regulated river flow. Although the simulations were conducted globally, the Missouri-Mississippi and Green-Colorado Rivers were chosen as case-study sites in this study. The hydrological models incorporate generic schemes of dam operation, not specific to a certain dam. We examined river discharge on a longitudinal section of river channels to investigate the effects of dams on simulated discharge, especially at the seasonal time scale. We found that the magnitude of dam regulation differed considerably among the hydrological models. The difference was attributable not only to dam operation schemes but also to the magnitude of simulated river discharge flowing into dams. That is, although a similar algorithm of dam operation schemes was incorporated in different hydrological models, the magnitude of dam regulation substantially differed among the models. Intermodel discrepancies tended to decrease toward the lower reaches of these river basins, which means model dependence is less significant toward lower reaches. These case-study results imply that, intermodel comparisons of river discharge should be made at different locations along the river’s course to critically examine the performance of hydrological models because the performance can vary with the locations.