Ouranos Ensemble of Bias-adjusted Simulations - Global models CMIP6 - AHCCD v3 (ESPO-G6-AHCCD v1.0.0)

This dataset is an ensemble of CMIP6 climate model bias-adjusted simulations for total daily precipitation, minimum, maximum and mean daily temperature over the period 1950–2100. Daily time series are extracted at 692 locations for temperature, and 439 for precipitation, corresponding to weather stations from the "Adjusted and homogenized Canadian climate data" (AHCCD) published by Environment and Climate Change Canada (ECCC). Climate simulations grids are bilinearly interpolated at each station with a 1.5° buffer zone, then bias-adjusted using the detrended quantile mapping (DQM) method and AHCCD as the reference dataset. This ensemble has two distinguishing features. It is bias-adjusted at the weather station scale, namely all AHCCD stations with at least 25 years within a period of 35 years with valid months according to WMO criteria (a month is considered valid if it has has less than 5 consecutive missing days, and less than 11 missing days overall). Second, it includes a large number of model realizations from ScenarioMIP simulations (627 for precipitation and 561 for temperature) across four scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5). CMIP6 simulations were acquired via the zarr Pangeo Catalog (https://pangeo-data.github.io/pangeo-cmip6-cloud/overview.html) and biasdajusted with the xscen package (https://xscen.readthedocs.io/en/latest/). This dataset has been created as part of project funded by Infrastructure Canada assessing the likelihood of future climate hazards to built infrastructures.