{"success":true,"data":{"author":"Google Research (NeuralGCM team)","bucketAccessType":"public","bucketListable":true,"bucketName":"neuralgcm","bucketPrefix":"aimip","category":"Climate & Weather","cloudProvider":"gcp","createdAt":"2026-11-14T14:15:25.956Z","description":"Output of the NeuralGCM hybrid machine-learning general circulation model (Kochkov et al., Nature 2024) prepared for the AI Model Intercomparison Project (AIMIP), an AMIP-style protocol in which AI atmosphere models are driven by observed sea surface temperature and sea ice and compared on climate statistics (AIMIP Phase 1, arXiv 2605.06944). Files follow CMIP6 conventions (DRS path GOOGLE/<model>/<experiment>/<member>/<table>/<variable>/gn/<version>) as netCDF, one file per variable covering October 1978 to December 2024. Two model configurations are included: NeuralGCM on a 2.8 degree grid and NeuralGCM-HRD, a TL63 model with a high-resolution decoder writing a 1x1 degree 360x180 grid, trained on 1979-2014 ERA5. Each has the historical aimip experiment plus aimip-p2k and aimip-p4k uniform SST warming experiments, with ensemble members r1i1p1f1 to r5i1p1f1 for NeuralGCM, and monthly (Amon) and daily (day) tables of specific humidity, precipitation, surface pressure, air temperature, 2 m temperature, 2 m dewpoint, skin temperature, u and v wind, 10 m wind and geopotential height. Measured 2026-11-14: NeuralGCM holds 545 files (30.7 GB) and NeuralGCM-HRD 540 files (238.7 GB). Open with xarray.open_dataset or xarray.open_mfdataset. The file metadata states CMIP6 model data produced by GOOGLE is licensed CC BY 4.0.","displayTitle":"NeuralGCM AIMIP Atmosphere-Only Climate Simulations (1978-2024, +2K, +4K)","fuseMountBrowsable":true,"id":"aa1d74c6-3b91-540c-ac1f-9eb13fe42ec3","license":"CC-BY-4.0","licenseEvidenceUrl":"https://storage.googleapis.com/neuralgcm/aimip/GOOGLE/NeuralGCM/aimip/r1i1p1f1/Amon/hus/gn/v20251203/hus_Amon_NeuralGCM_aimip_r1i1p1f1_gn_197810-202412.nc","objectCountEstimate":1,"probeStatus":"ok","referenceUrl":"https://arxiv.org/abs/2605.06944","region":"us","schemaCard":{"accessHints":[{"language":"python-gcsfs","snippet":"import gcsfs\nfs = gcsfs.GCSFileSystem(token='anon')\nfs.ls('neuralgcm/aimip/', detail=False)[:10]"},{"language":"python-storage","snippet":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nlist(client.bucket('neuralgcm').list_blobs(prefix='aimip/', max_results=10))"}],"caveats":["Files are binary — use a domain-specific parser rather than text I/O."],"format":"binary","generatedAt":"2026-11-14T14:15:31.719Z","layout":{"example":"aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn/v20251209/hus_Amon_NeuralGCM-HRD_aimip-p2k_r1i1p1f1_gn_197810-202412.nc","pattern":"{id}/{seg1}/{seg2}/{seg3}/{seg4}/{seg5}/{seg6}/{seg7}/{seg8}/{filename}"},"mountPath":"/mnt/gcs-archive/data","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn/v20251209/hus_Amon_NeuralGCM-HRD_aimip-p2k_r1i1p1f1_gn_197810-202412.nc","truncatedAt":512},{"bytes":"folder","encoding":"utf8","key":"aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn/v20251209_$folder$","truncatedAt":512}],"sampleObjects":["aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn/v20251209/hus_Amon_NeuralGCM-HRD_aimip-p2k_r1i1p1f1_gn_197810-202412.nc","aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn/v20251209_$folder$","aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn_$folder$","aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus_$folder$","aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/pr/gn/v20251209/pr_Amon_NeuralGCM-HRD_aimip-p2k_r1i1p1f1_gn_197810-202412.nc"],"sourceObjectKey":"aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn/v20251209/hus_Amon_NeuralGCM-HRD_aimip-p2k_r1i1p1f1_gn_197810-202412.nc","starterCell":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nbucket = client.bucket('neuralgcm')\nblob = bucket.blob('aimip/GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn/v20251209/hus_Amon_NeuralGCM-HRD_aimip-p2k_r1i1p1f1_gn_197810-202412.nc')\nwith blob.open(\"rb\") as fh:\n    print(fh.read(64))","starterSamplePath":"GOOGLE/NeuralGCM-HRD/aimip-p2k/r1i1p1f1/Amon/hus/gn/v20251209/hus_Amon_NeuralGCM-HRD_aimip-p2k_r1i1p1f1_gn_197810-202412.nc","totalObjects":"1 top-level prefixes"},"summary":"Output of the NeuralGCM hybrid machine-learning general circulation model (Kochkov et al., Nature 2024) prepared for the AI Model Intercomparison Project (AIMIP), an AMIP-style…","title":"NeuralGCM AIMIP Atmosphere-Only Climate Simulations (1978-2024, +2K, +4K)","updatedAt":"2026-11-14T17:34:03.505Z","visibility":"public","websiteUrl":"https://neuralgcm.readthedocs.io/en/latest/"}}