The need for high resolution
Global climate models (GCMs) or regional climate models (RCMs) provide climate data of a spatial scale that is much coarser than required for impact studies. GCMs have resolutions of approximately 1–3° (hundreds of kilometers) while RCMs can be as fine as tens of kilometers. However, climate impact studies require resolutions of 1km down to 100m or even the equivalent of point observations.
Also, study sites are usually situated far away from weather stations and thus measured data needs to be interpolated between these stations to form a contiguous gridded surface from which data can then be sampled at the study site's position.
The goal of this project is to provide current and scenario climate grids of resolutions adequate for ecological studies and applications. This is achieved by downscaling GCM and RCM output using various methods or by interpolating weather station data. The data is currently only accessible over the intranet but shall be published later on the web.
The change factor or delta change method
This method requires a high resolution observed climate baseline and coarse-scale GCM/RCM climate change projections. It is fast and easy, computationally inexpensive and thus allows to downscale many GCM/RCM model chains.
The moving window method
This method (Zimmermann & Roberts, 2001) is described here and it is applicable if only one climate layer is available. A relationship of the climate variable to local surface variables (elevation, latitude and longitude) is established in a moving window by deriving lapse rates and intercepts using single or multiple linear regression. These derived grids are interpolated to high resolution and the climate variable is recalculated using the linear relationship to the high resolution surface variables.
Daymet (Thornton et al., 1997 and https://daymet.ornl.gov/) is an interpolation software designed to produce gridded estimates of daily weather parameters over large regions. As input serve daily minimum, maximum, mean temperature and precipitation observations from climate stations and a digital elevation model.
Various downscaled gridded climate parameters are available for Switzerland, Europe or other continents. They cover current climate or different future climate scenarios from many GCMs or RCMs in different spatial and time resolutions. Unfortunately, the data sets for Switzerland are only available for WSL internal use.