Data record

ERA5–Drought consists of the SPI and the SPEI calculated over a range of accumulation periods (1, 3, 6, 12, 24, 36, 48 months) for the ERA5 reanalysis and the corresponding ensemble, consisting of 10 ensemble members. Indices for the entire ERA5 dataset from 1940–today are provided. In addition, quality criteria that rate the reliability of the derived indices, are provided. 

Data streams

Here, two products from the ERA5 dataset are used to calculate drought indices:

• the deterministic model output (simply called 'reanalysis', based on the 'moda' stream from ERA5), and 
• the probabilistic model output that consists of 10 ensemble members (simply called 'ERA5 ensemble', based on the 'edmo' stream from ERA5).

The corresponding daily averages from the 'moda' and the 'edmo' stream were converted to monthly sums by multiplying the fluxes with the number of days in each month. 
Both SPI and SPEI are calculated for each of those products, rendering also a deterministic version and a probabilistic version of each drought index. The latter ensemble allows to assess uncertainties associated with ERA5 data. 

Please note that the ensemble spread provides an indication of part of the uncertainty in the underlying reanalysis but should primarily be considered a measure of confidence in representing the synoptic situation. The derivation of SPI and SPEI introduces additional uncertainties (e.g., through the fit of a statistical distribution, the selection of a 30-year reference period, etc.) that is not addressed here. Additional uncertainties due to biases in the input variables of SPI and SPEI may be present.

Definition of indices and naming convention

ERA5–Drought consists of the SPI and the SPEI calculated over a range of accumulation periods (1, 3, 6, 12, 24, 36, 48 months), where the accumulation period is indicated in the name of the drought index. For example, an SPI calculated with an accumulation period of 1 month is denoted as SPI-1, and an SPI calculated with an accumulation period of 12 months is denoted as SPI-12, and so on. Analogous to other data sets, the valid date of both drought indices is set to the end of each accumulation period. E.g., the SPI-12 for January 2024 is valid at the end of January 2024 and takes into account the previous 12 months, i.e. from 1 February 2023 to 31 January 2024. 

Data availability and updates

Both ERA5 and ERA5–Drought data are available from January 1940. However, drought indices with accumulation periods larger than 1 month start with a corresponding delay. E.g., the SPI-12 only starts at the end of December 1941, as the previous 12 months are required for the accumulation.

Dataset are updated regularly in what is referred to as the 'Consolidated dataset' on the Climate Data Store (CDS). As the process of consolidating and releasing the official dataset takes about two months, the CDS also hosts a so-called 'Intermediate dataset' of ERA5. This experimental dataset is published behind real time and could be affected by changes before the official release date. An intermediate release is also issued for ERA5–Drought, enabling a near-real time monitoring of droughts with just one month of delay. However, the ERA5–Drought dataset is updated with each update of the consolidated ERA5 dataset as well. 

Data format

ERA–Drought is made available on a regular unprojected grid with spherical coordinates expressed in decimal degrees (EPSG:4326). Latitudes span the range from -90 to +90 degrees and are referenced to the equator. Longitudes are in the range from 0 to 360 degrees, referenced to the Greenwich Prime Meridian, consistent with ERA5 and derived products that use ERA5 meteorological fields for other application such as fire prediction or heat stress. The spatial resolution is 0.25 x 0.25 degrees (about 28 km grid cell size). Ensemble data are concatenated in the 'realization' dimension.

Data are provided in netCDF format following the CF-standard convention v1.8.

File naming

Data are provided in netCDF files for each drought index (SPI, SPEI), accumulation window (1, 3, 6, 12, 24, 36, 48), distribution (gamma for SPI, log-logistic for SPEI), dataset (ERA5 for the consolidated dataset or ERA5T for the unconsolidated dataset), stream (moda for the deterministic data stream, edmo for the probabilistic data stream, both following the ERA5 stream naming convention), and year and month separately. The files are named as follows:

(index)(window)_(distribution)_global_(dataset)_(stream)_ref1991to2020_(yyyy)(mm).nc.

The file naming for the quality criteria (pzero, significance) is analogous and follows the structure:

(index)(window)_(index) (quality_criterion)_(distribution)_global_(dataset)_(stream)_ref1991to2020_(mm).nc.

Quality criteria are calculated for the reference period (i.e., 1991–2020) but only one time step and file for each month is provided (thus omitting the year in the filename).

Versioning

The current version is version 1.0.

New versions of the dataset are published as settings, such as the reference period, are updated or bug fixes are applied.
Bug Fixes will be released using a minor revision (i.e. v1.1), while changes to the reference period will be released as major revisions (i.e. v2.0). Bug Fixes will be published to the Known Issues area on the Documentation tab.

Overview of data records and updates