This page presents some readings, datasets, and tools for the areas outlined in our paper, “Tackling Climate Change with Machine Learning.” These resources include general references about climate change, as well as resources organized by section of the paper.
If you find that we are missing some important resources, please submit them through this form!
For an overall introduction to the topic of climate change, we provide a (non-exhaustive) list of reports, academic papers, and conference proceedings below:
1. Electricity Systems
3. Buildings and Cities
The “metabolism” of a city includes the electricity used, waste generated, and GHG emitted.
5. Farms and Forests
6. CO2 Removal
7. Climate Prediction
The largest climate prediction datasets are ensembles of many climate simulations. These include simulations with varied physics, architectures, or initial conditions, and they are used to explore the range of possible climate futures. The largest ensembles include:
N.B. Climate model data is typically presented in netcdf4 format. These may be smoothly converted to csv files or pandas dataframes, but be aware that the data lies on irregular 3D spherical grids.
The Earth and climate science community is also working to create benchmark datasets: https://is-geo.org/benchmarks/
Climate science is a journal field. Noteworthy research appears in journals such as the Bulletin of the American Meteorological Society, Geophysical Research Letters and the Proceedings of the National Academy of Sciences.
8. Societal Impacts
Satellite imagery are used for ecological and social observation, and there are lists of publicly available datasets online. Here is one. For a satellite imagery dataset about food security specifically, consider this competition.
This competition describes an attempt to use mobile money effort to improve financial inclusion and resilience.
Improved disease surveillance and response is an important part of adaptation – here is one competition with this goal in mind.
9. Solar Geoengineering
10. Tools for Individuals
11. Tools for Society
Datasets are fairly limited in terms of data quantity (i.e. no single dataset would be enough to implement an ML system), but merging different sources of data together can yield interesting insights.