Abstract. Hydroclimatic extremes such as intense rainfall, floods, droughts, heatwaves, and wind or storms have devastating effects each year. One of the key challenges for society is understanding how these extremes are evolving and likely to unfold beyond their historical distributions under the influence of multiple drivers such as changes in climate, land cover, and other human factors. Methods for analysing hydroclimatic extremes have advanced considerably in recent decades. Here we provide a review of the drivers, metrics, and methods for the detection, attribution, management, and projection of nonstationary hydroclimatic extremes. We discuss issues and uncertainty associated with these approaches (e.g. arising from insufficient record length, spurious nonstationarities, or incomplete representation of nonstationary sources in modelling frameworks), examine empirical and simulation-based frameworks for analysis of nonstationary extremes, and identify gaps for future research.

Abstract. Long short-term memory (LSTM) models are recurrent neural networks from the field of deep learning (DL) which have shown promise for time series modelling, especially in conditions when data are abundant. Previous studies have demonstrated the applicability of LSTM-based models for rainfall–runoff modelling; however, LSTMs have not been tested on catchments in Great Britain (GB). Moreover, opportunities exist to use spatial and seasonal patterns in model performances to improve our understanding of hydrological processes and to examine the advantages and disadvantages of LSTM-based models for hydrological simulation. By training two LSTM architectures across a large sample of 669 catchments in GB, we demonstrate that the LSTM and the Entity Aware LSTM (EA LSTM) models simulate discharge with median Nash–Sutcliffe efficiency (NSE) scores of 0.88 and 0.86 respectively. We find that the LSTM-based models outperform a suite of benchmark conceptual models, suggesting an opportunity to use additional data to refine conceptual models. In summary, the LSTM-based models show the largest performance improvements in the north-east of Scotland and in south-east of England. The south-east of England remained difficult to model, however, in part due to the inability of the LSTMs configured in this study to learn groundwater processes, human abstractions and complex percolation properties from the hydro-meteorological variables typically employed for hydrological modelling.

Abstract. Long short-term memory models (LSTMs) are recurrent neural networks from the emerging field of Deep Learning (DL), which have shown recent promise when predicting time-series especially when data are abundant. Rainfall-runoff modelling presents a challenge, yet accurate hydrological models are vital for flood forecasting, hazard impact assessment, and to assess the potential effects of climate change on floods and water resources. In this study, we compare the performance of two DL-based models, a LSTM and an Entity Aware LSTM (EA LSTM). The DL models were trained using a newly published data set, CAMELS-GB, for a sample of 518 catchments across Great Britain. To identify spatial and seasonal patterns in model performance, we compare the DL models against benchmark outputs from four lumped conceptual models recently configured for rainfall-runoff modelling in Great Britain. Our findings show that the LSTM models simulate discharge with consistently high model performance scores, including in catchments typically considered difficult to model. The LSTM achieves a mean catchment NSE of 0.88 (0.86 for the EALSTM), which represents a performance improvement of 10 %–16 % compared with the benchmark conceptual models. Seasonal and spatial patterns indicate that the largest performance improvement relative to the benchmark is in the drier summer months and in drier catchments in the South East of England. By comparing LSTMs with conceptual models, we diagnose possible reasons for their different performance. We suggest that LSTMs offer useful predictive capability for rainfall-runoff modelling in Great Britain and elsewhere and note their value to support process understanding in locations where processes are less well understood.

Abstract. Hydroclimatic extremes such as intense rainfall, floods, droughts, heatwaves, and wind/storms have devastating effects each year. One of the key challenges for society is understanding how these extremes are evolving and likely to unfold beyond their historical distributions under the influence of multiple drivers such as changes in climate, land cover, and other human factors. Methods for analysing hydroclimatic extremes have advanced considerably in recent decades. Here we provide a review of the drivers, metrics and methods for the detection, attribution, prediction and projection of nonstationary hydroclimatic extremes. We discuss issues and uncertainty associated with these approaches (e.g arising from insufficient record length, spurious nonstationarities, or incomplete representation of nonstationary sources in modelling frameworks), examine empirical and simulation-based frameworks for analysis of nonstationary extremes, and identify gaps for future research.

Shifts in climate characteristics, such as precipitation totals, phase, and timing are widely considered to be the dominant drivers of hydrologic change, but land cover changes, consisting of changes in the biological or physical features present in a landscape (e.g., forested or urban area), also have the potential to drive changes in streamflow, potentially even offsetting the influence of climate (Slater, Anderson, et al., 2021). The effects of land cover changes on hydrological extremes, such as worsening flood risk (e.g., Bradshaw et al., 2007; van Dijk et al., 2009), or as potential mechanisms by which hydrological and climatic risks may be managed or offset (e.g., Dadson et al., 2017;Dixon et al., 2016) remain poorly understood. It is clear that land cover changes can alter hydrological response to precipitation events by influencing the degree and rate at which water is intercepted and evaporated, stored, or allowed to run off into a river channel (Filoso et al., 2017;Jacobson, 2011;Shuster et al., 2005), however, the extent to which they do so lacks clear definition.There are a number of ways in which land cover change might be expected to influence the magnitude of high, mean, and low daily streamflows. Widely discussed in the literature, urbanization is typically expected to increase high flows and flood risk (