Search

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.

The study assesses how climate change could affect Europe in eleven impact areas. Under a high warming scenario, several climate impacts show a clear geographical north-south divide. Most of the welfare losses, assessed for six impact areas, would be greatly reduced under a 2ᵒC scenario., JRC.C.6-Economics of Climate Change, Energy and Transport

This study uses five impact models that describe observed relationships between labour productivity and temperature, with climate model simulations from five climate models under a high emissions scenario (RCP8.5), i.e. 25 climate-impact model combinations, to assess the impact of climate change on outdoor and indoor labour productivity respectively, at the national-scale, across Europe. This is the first assessment to use multiple impact models with multiple climate models and to consider the potential effects of adaptation on lowering the impacts relative to no adaptation taking place. Impacts are estimated for the end of century (2071-2100) and near-term (2021-2050), relative to present-day (1981-2010). Impacts are also estimated under a mitigation scenario, where global-mean warming is 2°C relative to pre-industrial. Impacts are assessed with and without adaptation respectively. Planned adaptation is represented as an adjustment in work activities following recommendations by the US Occupational Safety & Health Administration to consider the adjustment of work shifts during hot periods – all labour takes place at night instead of day-time, under the adaptation assumption. Without climate change mitigation and adaptation, daily average outdoor labour productivity could decline by around 10-15% from present-day levels in several southern European countries by the end of the century (Bulgaria, Greece, Italy, Macedonia, Portugal, Spain and Turkey). Countries in northern Europe could also see declines in daily average outdoor labour productivity but the declines are considerably smaller than for the southern countries, at around 2-4% (Denmark, Estonia, Finland, Norway and Sweden). The magnitude of impact on indoor labour productivity is generally 2-4 percentage points lower than for impacts on outdoor labour productivity, for the three most sensitive impact models, while for the two least sensitive impact models, the differences are smaller. There is uncertainty in the magnitude of projected climate change impacts on labour productivity due to: 1) differences in the projections of climate between different climate models; and 2) the use of different impact models. Both sources of uncertainty are significant. The range in projected impacts due to using multiple climate models is comparable to the range in impacts from using multiple impact models with only one climate model. Adaptation and mitigation have the potential to significantly lessen the impacts of climate change on declines in labour productivity across Europe. For some countries the impacts can be up to around 10 percentage points lower with adaptation than without, for some climate-impact model combinations, at the end of the century under high emissions (e.g. Bulgaria, Croatia, Greece, Italy, Spain and Turkey). However, the declines in daily average outdoor labour productivity could still be around 5% relative to present-day in these countries (and up to 10% for Greece, with one climate-impact model combination). Whilst the potential benefits of adaptation are clear from this assessment, it is important to be aware of the caveats associated with the adaptation modelling approach employed. These include an assumption of the entire work force engaged in moderate to heavy labour shifting to night-time working, acknowledgment that night-time working can be associated with negative health effects, and potentially higher costs of night-time working due to energy requirements for lighting and higher wages for working unsocial hours. Such a change in working practices is optimistic, but not implausible, since currently around 20% of workers in Europe are employed on shift work involving night work. Limiting global warming to below 2°C (and assuming no adaptation) could avoid a substantial proportion of impacts in the European countries that see the largest impacts without mitigation (Bulgaria, Greece, Italy, Macedonia, Portugal, Spain and Turkey). With some climate-impact model combinations the declines in labour productivity can be up to 10 percentage points lower in these countries with mitigation when compared to without mitigation, JRC.C.6-Economics of Climate Change, Energy and Transport

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 monthly runoff in 40 catchments, spatially distributed across eight 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 of a 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 all indicators of upper and lower extreme runoff. 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—a finding 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., Environmental Research Letters, 13 (6), ISSN:1748-9326, ISSN:1748-9318

Human activities have a profound influence on river discharge, hydrological extremes, and water-related hazards. In this study, we compare the results of five state-of-the-art global hydrological models (GHMs) with observations to examine the role of human impact parameterizations (HIP) in the simulation of the mean, high, and low flows. The analysis is performed for 471 gauging stations across the globe and for the period 1971-2010. We find that the inclusion of HIP improves the performance of GHMs, both in managed and near-natural catchments. For near-natural catchments, the improvement in performance results from improvements in incoming discharges from upstream managed catchments. This finding is robust across GHMs, although the level of improvement and reasons for improvement vary greatly by GHM. The inclusion of HIP leads to a significant decrease in the bias of long-term mean monthly discharge in 36-73% of the studied catchments, and an improvement in modelled hydrological variability in 31-74% of the studied catchments. Including HIP in the GHMs also leads to an improvement in the simulation of hydrological extremes, compared to when HIP is excluded. Whilst the inclusion of HIP leads to decreases in simulated high-flows, it can lead to either increases or decreases in low-flows. This is due to the relative importance of the timing of return flows and reservoir operations and their associated uncertainties. Even with the inclusion of HIP, we find that model performance still not optimal. This highlights the need for further research linking the human management and hydrological domains, especially in those areas with a dominant human impact. The large variation in performance between GHMs, regions, and performance indicators, calls for a careful selection of GHMs, model components, and evaluation metrics in future model applications.