DECIPHeR model estimates of daily flow for 1366 gauged catchments in Great Britain (1962-2015) using observed driving data

This dataset provides 100 model realisations of daily river flow in cubic metres per second (m3/s) for 1,366 catchments, for the period 1962 to 2015. The dataset is model output from the DECIPHeR hydrological model driven by observed climate data (CEH-GEAR rainfall and CHESS-PE potential evapotranspiration). The modelled catchments correspond to locations of National River Flow Archive (NRFA) gauging stations and provide good spatial coverage across the UK. The dataset was produced as part of MaRIUS (Managing the Risks, Impacts and Uncertainties of drought and water Scarcity) to provide national scale probabilistic flow simulations and predictions for UK drought risk analysis. MaRIUS was a UK NERC-funded research project (2014-2017) that developed a risk-based approach to drought and water scarcity.,The modelled flow data are output from the DECIPHeR hydrological model (Coxon et al, 2019; https://doi.org/10.5194/gmd-12-2285-2019). The meteorological inputs to the model were rainfall from CEH-GEAR (Tanguy et al. 2016; https://doi.org10.5285/33604ea0-c238-4488-813d-0ad9ab7c51ca) and CHESS-PE (Robinson et al, 2016; https://doi.org/10.5285/8baf805d-39ce-4dac-b224-c926ada353b7). DECIPHeR (V1) was run from 1961-2015 using 100,000 monte carlo model parameter sets. Model performance was evaluated over a 54 year timeseries (1962-2015) using a multi-objective approach comprising four evaluation metrics. The evaluation metrics were chosen to quantify the model’s ability to capture a range of hydrologic behaviour including maintaining overall water balance, capturing flow variability, reproducing low and high flows and the timing of flows. The top ‘multi-objective’ scoring 100 model simulations for each of the 1,366 gauges are supplied as daily river flow timeseries. Metadata comprising the model parameters and performance for the 100 ensemble members is also supplied to allow the user to interrogate the performance of the model and assist in the interpretation of the results. See supporting information document for more details.,