climate+ offers consistent climate prediction data in hourly resolution until the end of the 21st century for air temperature, precipitation, and a wide range of derived variables (such as the number of tropical nights, or heat days per year). All data can be downloaded as CSV files, and can be easily imported into various applications such as Microsoft® Excel or OpenOffice. The hourly time series of air temperature and precipitation can be downloaded for each RCP emission scenario, and for the desired time period (e.g., 2071 – 2080).

Note that these are synthetic data, i.e., they can be used for statistical applications but NOT to forecast the weather for a specific point in time (e.g., 21st January 2071 10:00 UTC). We provide three time series: “Low”, “Medium” and “High” because, as indicated above, it is scientifically impossible to forecast precipitation and air temperature for a specific point in time this far in the future. This method helps the users to understand the uncertainty range within the interannual variability. The classification of the “Low”, “Medium” and “High” values is based on yearly average air temperatures of the 30-year reference period. Please do not hesitate to contact us in the free monthly webinar if you have further questions regarding the dataset.

Limitations

As mentioned, it is not possible to forecast the exact temperature or precipitation for a specific point in time in the distant future. There are limitations to the extent of weather forecasts, typically 5 – 7 days, depending on the particular weather variable (e.g., air temperature). Therefore, hourly time series until the end of the century are a synthetic dataset, which can be used to calculate statistics or run external models (e.g., crop models). We recommend averaging the hourly values of a 10-year (or preferably a 30-year) period for the analyses of the future climate for your target use case.

Note that it is possible that some time steps of the time series in the “Low” values are higher than the “High” values. That is caused by the fact that the classification of these three time-series is based on yearly average temperatures from the past, and the selected year may differ from the others.

We utilise climate prediction data from the CCSM4 model (spatial resolution 100 km) and downscale the data to 30 km spatial resolution (ERA5 grid). Thus, climate+ data is based on a global climate prediction model, and does not exceed a spatial resolution of 30 km.

Consequently, regional and local climate prediction model data are not reflected in climate+. Please contact us at [email protected] if you are interested in regional climate prediction models, or if you are interested in other global climate prediction models. Alternatively, please consider the meteoblue Climate data & Visuals.