Climate data

ECMWF - ERA5 Reanalysis

The ERA5 reanalysis dataset is a comprehensive, high-resolution global climate dataset produced by the European Centre for Medium-Range Weather Forecasts (ECMWF). Covering the period from January 1950 to the present, ERA5 assimilates a wide range of observational data using advanced modeling techniques, providing hourly estimates of atmospheric, oceanic, and land surface conditions.

This dataset is crucial for climate analysis, offering detailed insights into weather patterns, climate variability, and long-term trends. It is widely used for research in meteorology, hydrology, and environmental science. In RS Climate labe you will find tools enable seamless downloading and processing of ERA5 data.

RS Climate Lab offers tooling to submit download jobs to Climate Data Store (CDS). Downloading data from CDS is not a trivial task, it can take hours (sometimes days) for you to have access to the data after submitting a job, so you need to orchestrate download jobs and keep monitoring them.

You can use the rscl CLI tool from RS Climate Lab to download data. The first step is to define a configuration file such as the example below:

parallel_jobs: 10
variables:
    - total_precipitation
    - 2m_temperature
    - minimum_2m_air_temperature
    - maximum_2m_air_temperature
    - potential_evaporation
    - surface_net_solar_radiation
years:
    start: 2020
    end: 2022

Note

Note that RS Climate Lab (rscl) is focused in Rio Grande do Sul (RS) in Brazil, therefore it will automatically download only data for the state. We are planning to support other areas in near future.

After that you can run the following to download the data for the specified years:

rscl era5 download --config-file config.yml output-dir

This command will start and monitor the download jobs for the specified ECMWF ERA5 climate data from the CDS servers. It will also do that in parallel and keep track of errors so if some file fails, you can re-run the same command and it will download only the missing files.

Note

It is important to note that rscl will download the year data splitting the year in two parts (Jan to Jun and then Jul to Dec).