{"url":"/dataset/earthnet2021","name":"EarthNet2021","full_name":"EarthNet2021: Earth Surface Forecasting","description_markdown":"Satellite images are snapshots of the Earth surface. We propose to forecast them. We frame Earth surface forecasting as the task of predicting satellite imagery conditioned on future weather. EarthNet2021 is a large dataset suitable for training deep neural networks on the task. It contains Sentinel~2 satellite imagery at $20$~m resolution, matching topography and mesoscale ($1.28$~km) meteorological variables packaged into $32000$ samples. Additionally we frame EarthNet2021 as a challenge allowing for model intercomparison. Resulting forecasts will greatly improve ($>\\times50$) over the spatial resolution found in numerical models. This allows localized impacts from extreme weather to be predicted, thus supporting downstream applications such as crop yield prediction, forest health assessments or biodiversity monitoring. Find data, code, and how to participate at www.earthnet.tech.","description_withheld":null,"homepage":"https://www.earthnet.tech/docs/ds-download/","introduced_date":"2021-04-16","introduced_date_note":null,"introduced_by":{"paper":"/paper/earthnet2021-a-large-scale-dataset-and","title":"EarthNet2021: A large-scale dataset and challenge for Earth surface forecasting as a guided video prediction task","first_author":"Christian Requena-Mesa","url":null},"license":{"name":"CC-BY-NC-SA 4.0","url":"https://www.earthnet.tech/license/"},"modalities":[],"tasks":[{"name":"Time Series Forecasting","url":"/task/time-series-forecasting","datasets_with_task":"/datasets/task/time-series-forecasting"},{"name":"Multivariate Time Series Forecasting","url":"/task/multivariate-time-series-forecasting","datasets_with_task":"/datasets/task/multivariate-time-series-forecasting"},{"name":"Video Prediction","url":"/task/video-prediction","datasets_with_task":"/datasets/task/video-prediction"},{"name":"Earth Surface Forecasting","url":"/task/earth-surface-forecasting","datasets_with_task":"/datasets/task/earth-surface-forecasting"},{"name":"Time Series Prediction","url":"/task/time-series-prediction","datasets_with_task":"/datasets/task/time-series-prediction"},{"name":"Video Forensics","url":"/task/video-forensics","datasets_with_task":"/datasets/task/video-forensics"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["EarthNet2021","EarthNet2021 IID Track","EarthNet2021 OOD Track","EarthNet2021 Extreme Track","EarthNet2021 Seasonal Track"],"data_loaders":[{"repo":"https://github.com/earthnet2021/earthnet-toolkit","url":"https://github.com/earthnet2021/earthnet-toolkit","frameworks":[]}],"num_papers_in_archive":10,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/earth-surface-forecasting-on-earthnet2021-iid","task":"Earth Surface Forecasting","dataset_variant":"EarthNet2021 IID Track","rows":7,"metrics":["EarthNetScore"],"first_row_in_archive_order":{"model":"Earthformer","paper":"/paper/earthformer-exploring-space-time-transformers","metrics":{"EarthNetScore":"0.3425"},"code_links":[{"title":"amazon-science/earth-forecasting-transformer","url":"https://github.com/amazon-science/earth-forecasting-transformer"},{"title":"PaddlePaddle/PaddleScience","url":"https://github.com/PaddlePaddle/PaddleScience"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/earth-surface-forecasting-on-earthnet2021-ood","task":"Earth Surface Forecasting","dataset_variant":"EarthNet2021 OOD Track","rows":7,"metrics":["EarthNetScore"],"first_row_in_archive_order":{"model":"Earthformer","paper":"/paper/earthformer-exploring-space-time-transformers","metrics":{"EarthNetScore":"0.3252"},"code_links":[{"title":"amazon-science/earth-forecasting-transformer","url":"https://github.com/amazon-science/earth-forecasting-transformer"},{"title":"PaddlePaddle/PaddleScience","url":"https://github.com/PaddlePaddle/PaddleScience"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/earth-surface-forecasting-on-earthnet2021","task":"Earth Surface Forecasting","dataset_variant":"EarthNet2021 Extreme Track","rows":6,"metrics":["EarthNetScore"],"first_row_in_archive_order":{"model":"SGConvLSTM","paper":"/paper/deep-learning-for-satellite-image-forecasting","metrics":{"EarthNetScore":"0.2740"},"code_links":[{"title":"rudolfwilliam/satellite_image_forecasting","url":"https://github.com/rudolfwilliam/satellite_image_forecasting"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/earth-surface-forecasting-on-earthnet2021-1","task":"Earth Surface Forecasting","dataset_variant":"EarthNet2021 Seasonal Track","rows":6,"metrics":["EarthNetScore"],"first_row_in_archive_order":{"model":"Persistence Baseline","paper":"/paper/earthnet2021-a-large-scale-dataset-and","metrics":{"EarthNetScore":"0.2676"},"code_links":[{"title":"earthnet2021/earthnet-model-intercomparison-suite","url":"https://github.com/earthnet2021/earthnet-model-intercomparison-suite"},{"title":"earthnet2021/earthnet-toolkit","url":"https://github.com/earthnet2021/earthnet-toolkit"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/deep-learning-for-satellite-image-forecasting","title":"Deep learning for satellite image forecasting of vegetation greenness","date":"2022-08-17","rows_on_this_dataset":8,"code_links":1,"syntology":null},{"paper":"/paper/earthformer-exploring-space-time-transformers","title":"Earthformer: Exploring Space-Time Transformers for Earth System Forecasting","date":"2022-07-12","rows_on_this_dataset":2,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":11,"samples_ran":0,"samples_unverified":11,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/understanding-the-role-of-weather-data-for","title":"Understanding the Role of Weather Data for Earth Surface Forecasting using a ConvLSTM-based Model","date":"2022-06-20","rows_on_this_dataset":4,"code_links":1,"syntology":null},{"paper":"/paper/earthnet2021-a-large-scale-dataset-and","title":"EarthNet2021: A large-scale dataset and challenge for Earth surface forecasting as a guided video prediction task","date":"2021-04-16","rows_on_this_dataset":12,"code_links":2,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":1,"samples_harvested":11,"samples_ran":0,"samples_unverified":11,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":1,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}