Storm Eunice (February 2022): Pre-industrial, current, and future climate scenarios using IFS EPS CY47R3 at 8, 4, and 2 days lead time

This data accompanies the submitted article Shirin Ermis et al. (under review): Event attribution of a midlatitude cyclone using ensemble weather forecasts.

Description

Please note that the data in this dataset is post-processed and not raw data to make storage more efficient. The study correcponding to this dataset compares the operational forecast (curr) as well as a future (fut or incr) and pre-industrial (pi) scenario of storm Eunice which hit the UK on February 18, 2022. Forecasts are run using the ECMWF ensemble prediction system (IFS EPS, CY47R3) with 51 ensemble members. We use forecasts at 8, 4, and 2 days lead time to the storm hitting the UK. Acordindly, the forecasts were initialised at 00:00UTC on Febryary 10th, 14th and 18th, 2022. For the counterfactual scenarios of the storm, we adjusted ocean temperatures in 3d, including sea surface temperatures (SST) as well as CO2 concentrations. The concentrations of CO2 were 285ppm, 421ppm, and 625ppm for pi, curr, and fut respectively. 

The dataset aslo contains data from the operational analysis and forecast initialised at 00:00UTC on February 18th, 2022.

Raw data can be downloaded by ECMWF MARS users under the United Kingdom memberstate data and experiment codes b2nn, b2ns, b2nq (all for pi), and b2no, b2nr, b2nt (for incr). The curr simulations can be found in the operational ensemble prediction system archive for the respective initialisation dates.

For any further information, please refer to the article and references therein.

Usage

Each of the data files is needed to run the code which is publicly available here. The notebooks folder contains the python notebooks PAPER1, PAPER2, PAPER3, and PAPER4 which produce the plots as shown in the paper.

To be able to run the code fully with all aspects of the figures, reanalysis data from ERA5 (Hersbach et al. 2020) is also necesary. This data is freely available online for research purposes. We use the variables mean sea level pressure (msl) and wind gusts at 10m (fg10) for February 2022.

To run our code, please create a conda environment with the environment.yml file in the docs folder. 

In the PAPERx notebooks, please import the necessary packages at the start and then skip to the sectio "Save or load data". In most cases this will load all the necessary data for the plotting cell below, although in some cases loading and processing ERA5 data might be neceessary.

Acknowledgements

S.E. was supported by the Natural Environment Research Council (NERC) under Grant NE/S0074747/1 and a Graduate Scholarship from St Cross College Oxford. N.J.L. was supported by the Natural Environment Research Council under Grant NE/L002612/1 and the European Union's Horizon 2020 project FORCeS under Grant GA 821205.  A.W. was supported by the NERC CANARI project (NE/W004984/1) and by European Union’s Horizon Europe research and innovation programme under grant agreement No. 101081460. F.C.L. was supported by the Met Office Hadley Centre Climate Programme funded by DSIT. S.N.S. was supported by the NERC Centre for Greening Finance and Investment (CGFI) under Grant NE/V017756/1. The results contain modified Copernicus Climate Change Service information [2022]. Neither the European Commission nor ECMWF is responsible for any use that may be made of the Copernicus information or data it contains. Acknowledgement is made for the use of ECMWF’s computing and archive facilities in this research under the special project spgbleac.