Climate scenarios

You can access the CH2018 data via the NCCS website.

info

The CH2025 data are available in November 2025.

This is a draft that is being worked on until the official launch of the CH2025 climate scenarios on November 4th.

warning

Links in this documentation will only work after the launch of the CH2025 climate scenarios on November 4th.

The localised Climate CH2025 datasets consist of 30-year daily time series for different Global Warming Levels (GWL) (forthcoming) and the reference period 1991–2020 for several climate variables at individual Swiss stations (DAILY-LOCAL) and on a regular 1 km grid covering the area of Switzerland (DAILY-GRIDDED). This data is primarily useful for research purposes or professional consulting.

A detailed description of the localised Climate CH2025 datasets is available on the MeteoSwiss Website: Climate CH2025 datasets (forthcoming)

Data download

The Open Data from MeteoSwiss may be used without restriction; the source must be cited when reproducing or redistributing ("Source: MeteoSwiss & ETH Zurich (2025): Climate CH2025 - Daily Datasets. Federal Office of Meteorology and Climatology MeteoSwiss, Zurich, https://doi.org/10.18751/climate/scenarios/ch2025/data/1.0/").

✅ By using 'Open Data' from MeteoSchweiz, you confirm that you have taken note of the Terms of use.

Download options

Manual download via STAC Browser

Users that prefer to use a web interface to browse and download individual files can use the STAC Browser for DAILY-LOCAL and DAILY-GRIDDED.

Download using R

The script below shows how one could use the R language with the rstac library to query the STAC API and download files.

More information about the STAC specification and and R tutorial can be found on stacspec.org.

library(rstac)

# Define the STAC API to use
stac_source <- rstac::stac(
  "https://data.geo.admin.ch/api/stac/v1/"
)
stac_source

# Show info about STAC source
rstac::get_request(stac_source)

# Query source for available collections
available_collections <- stac_source |>
  rstac::collections() |>
  rstac::get_request()
available_collections

# define filter criteria for STAC search
collection_id <- "ch.meteoschweiz.ogd-climate-scenarios-ch2025"
collection <- stac_source |>
  rstac::collections(collection_id = collection_id) |>
  rstac::get_request()
collection

# Query all assets of the selected collection
assets <- rstac::stac_search(
  q = stac_source,
  collection = collection_id,
  limit = 999) |>
  rstac::get_request()

# Helper function to filter assets
# @param assets: return value of a call to rstac::stac_search c
# @param pattern: regexp pattern to search in asset name
# @param extension: optional file ending (default NULL)
assets_subset <- function(assets, pattern="", extension=NULL) {
  assertions::assert_class(assets, "doc_items")
  assertions::assert_character(pattern)
  if (!is.null(extension)) {
    assertions::assert_character(extension)
    pattern <- paste0(pattern, ".*\\.", extension)
  }
  subset <- rstac::assets_select(assets,
                                select_fn = function(asset){
                                  return(length(grep(pattern, asset$title)) > 0)
                                }
  )
  n_assets <- length(rstac::assets_url(subset))
  cat(n_assets, "assets found for pattern", paste0("'", pattern, "'."), fill = TRUE)
  return(subset)
}

# Filter assets using the helper function: find all csv assets for parameter hurs
hurs_assets <- assets_subset(assets, pattern = "_hurs_", extension = "csv")

# get download urls for these assets
rstac::assets_url(hurs_assets)

# download the selected assets
# rstac::assets_download(hurs_assets, output_dir = tempdir())

Download using Python

The script below shows how one could use the Python language with the pystac and pystac_client packages to query the STAC API and download files.

More information about the STAC specification and python tutorials can be found on stacspec.org.

import json

from pystac import Asset
from pystac_client import Client, CollectionClient

import re
import urllib.request

def main(): 
    catalog = Client.open('https://sys-data.int.bgdi.ch/api/stac/v1/')
    print(catalog.title)

    # get a pystac client for the DAILY-LOCAL collection
    collection_id = "ch.meteoschweiz.ogd-climate-scenarios-ch2025"
    collection: CollectionClient = catalog.get_collection(collection_id)
    print(collection.title)

    # print all items
    for item in collection.get_items():
        print(f"{item.id}: {len(item.assets)}")

    # create a dict with all items of the collection
    assets_dict = {}
    for item in collection.get_items():
        assets_dict = assets_dict | item.assets
    print(f"Number of assets: {len(assets_dict)}")

    # find matching keys
    hits = [k for k in assets_dict.keys() if k == "ogd-climate-scenarios-ch2025_zwk_pr_ref91-20.csv" ]

    # use regular expression to find matching keys
    pattern = "_zwk_" # find all assets for station zwk, the same can be done for parameters, GWLs etc. 
    pattern = re.compile("^.*" + pattern + ".*$")
    hits = [k for k in assets_dict.keys() if pattern.match(k) ]
    print(f"Hits for pattern {pattern}: {hits}")

    # download all hits to current directory
    for k in hits:
        print(assets_dict[k].href)
        #urllib.request.urlretrieve(url=assets_dict[k].href, filename=k)

Data structure and format

Here is a short overview of the datasets:

Attributes	DAILY-LOCAL	DAILY-GRIDDED
Number of Parameters	7	4
Formats	CSV, NetCDF (in ZIPs)	NetCDF
Data Volume per file	CSV: ~1.5 MB NetCDF: ~200KB	~1-2 GB

Detailed information on the available simulations and variables, limitations and a list of available Swiss stations can be found in the user documentation of the localised Climate CH2025 datasets:

• User documentation pdf on the MeteoSwiss website: (Link forthcoming)
• Climate CH2025 datasets on the MeteoSwiss Website (forthcoming)

Metadata

DAILY-LOCAL Parameter

ogd-climate-scenarios-ch2025_meta_parameters.csv provides a list of all parameter identifiers with explanation, time interval, decimal places, data type and unit of measurement.

DAILY-LOCAL Stations

All stations have a three-letter identifier (e.g. BER for "Bern/Zollikofen" or LUG for "Lugano").

ogd-climate-scenarios-ch2025_meta_stations.csv provides a list of all station identifiers with name, Wigos ID, altitude, coordinates and region.

DAILY-GRIDDED Parameter

ogd-climate-scenarios-ch2025-grid_meta_parameters.csv provides a list of all parameter identifiers with explanation, time interval, decimal places, data type and unit of measurement.

Contact and staying up to date

If you have questions contact: klimaszenarien@meteoschweiz.ch

To receive updates on the datasets and complementary products sign up for the "Climate newsletter":

• in German,
• in French or
• in Italian.