climatemaps

An interactive web application for visualizing global historical climate data and future climate projections. It show temperature, precipitation, and other climate variables with support for multiple climate scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5) and time periods (2020-2100).

Demo

openclimatemap.org

Data

Historic

• CHELSA Climatologies at high resolution for the earth’s land surface areas: https://www.chelsa-climate.org/datasets
• Historic 1970-2000 WorldClim data: https://www.worldclim.org/data/worldclim21.html

Projections (predictions)

• WorldClim: Downscaled CMIP5 and CMIP6 model outputs for 2021–2100: https://www.worldclim.org/data/cmip6/cmip6_clim5m.html

Other (not used)

• https://interactive-atlas.ipcc.ch/
• CRU TS (Climatic Research Unit Time-Series): https://catalogue.ceda.ac.uk/uuid/ec331e93d21347e6bf5fa4b9c68dbd2c/
• ERA5 monthly averaged data on single levels from 1940 to present: https://cds.climate.copernicus.eu/datasets/reanalysis-era5-single-levels-monthly-means?tab=overview
• ERA5 post-processed daily statistics on single levels from 1940 to present: https://cds.climate.copernicus.eu/datasets/derived-era5-single-levels-daily-statistics?tab=overview
• Copernicus Climate Data Store - CMIP6 climate projections: https://cds.climate.copernicus.eu/datasets/projections-cmip6?tab=overview
• NASA NEX-GDDP-CMIP6 (~25 km resolution): https://registry.opendata.aws/nex-gddp-cmip6/

Development

Dependencies

GDAL

GDAL is needed for the gdal2tiles.py script that creates map-tiles from a single image (matplotlib plot).

Install GDAL with conda to prevent the need to install a hugh number of system level dependencies:

conda install -c conda-forge gdal==3.11.0

Tippecanoe

WARNING: tippecanoe 1.19.1 is the latest version that produces valid GeoJSON due to issue mapbox/tippecanoe#652

TODO: migrate to active fork https://github.com/felt/tippecanoe

sudo apt install libsqlite3-dev
git clone https://github.com/mapbox/tippecanoe.git
cd tippecanoe
git checkout tags/1.19.1
make -j
make install

tileserver-gl

npm install -g tileserver-gl

Run

Create contour data

Run:

python scrips/create_contour.py [--month=1] [-dataset-type=historic]

to create contour and raster mbtiles.

Create tileserver config

python scripts/create_tileserver_config.py

to generate the tileserver config.

Run the backend (FastAPI server)

uvicorn api.main:app --reload

Run the tileserver (tileserver-gl)

tileserver-gl --config tileserver_config_dev.json --port 8080

Or with auto-restart on config changes (requires nodemon: npm install -g nodemon):

nodemon --watch tileserver_config_dev.json --exec "tileserver-gl --config tileserver_config_dev.json --port 8080"

Run the client (Angular)

In ./client run:

ng serve

Or run against the openclimatemap.org API and tileserver:

ng serve --configuration=production-backend

Alternative: Docker Compose (recommended for production-like setup)

Start both the API server and tileserver using Docker Compose:

docker compose up -d

This will start:

• API server on localhost:8000
• Tileserver on localhost:8080

To view logs:

docker compose logs -f

To stop the services:

docker compose down

Note: Make sure you have already created the tiles and tileserver config before starting the services.

Tests

Run:

pytest

Build and deploy (to openclimatemap.org)

Everything

Build and deploy client and backend:

bash scripts/deploy.sh

TODO: include create tiles, create config and upload tiles

Client

Deploy the client angular app:

bash scripts/deploy_client.sh

API and TileServer

bash scripts/deploy_backend.sh

Logs

View logs for the API (FastAPI):

docker compose logs -f api

View logs for the TileServer:

docker compose logs -f tileserver

Upload tiles (optional)

If you have locally created tiles, upload them to the server with:

rsync -avP data/tiles openclimatemap.org:/home/bart/climatemaps/data/