{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:LJ6JEZDOU6F7TD4UKUSXX62LJA","short_pith_number":"pith:LJ6JEZDO","canonical_record":{"source":{"id":"2604.06398","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"physics.ao-ph","submitted_at":"2026-04-07T19:32:25Z","cross_cats_sorted":["cs.LG","physics.comp-ph"],"title_canon_sha256":"8264a9d39757dc2e959cf7e49bdfd13c3dc2141ad5193e2b5a86a9a21ce7d594","abstract_canon_sha256":"2e77b3c36d386097da71ad6f1729e3c7fe8e9ddd1fac6c476fdeea743d5e6b1b"},"schema_version":"1.0"},"canonical_sha256":"5a7c92646ea78bf98f9455257bfb4b483dfc47d306996a62f4e89b7864224030","source":{"kind":"arxiv","id":"2604.06398","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.06398","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"arxiv_version","alias_value":"2604.06398v2","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.06398","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"pith_short_12","alias_value":"LJ6JEZDOU6F7","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"pith_short_16","alias_value":"LJ6JEZDOU6F7TD4U","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"pith_short_8","alias_value":"LJ6JEZDO","created_at":"2026-05-20T00:04:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:LJ6JEZDOU6F7TD4UKUSXX62LJA","target":"record","payload":{"canonical_record":{"source":{"id":"2604.06398","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"physics.ao-ph","submitted_at":"2026-04-07T19:32:25Z","cross_cats_sorted":["cs.LG","physics.comp-ph"],"title_canon_sha256":"8264a9d39757dc2e959cf7e49bdfd13c3dc2141ad5193e2b5a86a9a21ce7d594","abstract_canon_sha256":"2e77b3c36d386097da71ad6f1729e3c7fe8e9ddd1fac6c476fdeea743d5e6b1b"},"schema_version":"1.0"},"canonical_sha256":"5a7c92646ea78bf98f9455257bfb4b483dfc47d306996a62f4e89b7864224030","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:30.883838Z","signature_b64":"2vVUE8qJo1Mu1qn3ZfBKa8rgKCBiLWzahhQu64fq12fwqPIf3kEfAwOmF4uH7lHvrO+ql+5tj9UOisdocbuhAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a7c92646ea78bf98f9455257bfb4b483dfc47d306996a62f4e89b7864224030","last_reissued_at":"2026-05-20T00:04:30.882839Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:30.882839Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.06398","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xo6+sTLB9GnOuVnbo2lNRtcelk7RRvAeMZySk3XrYe+HGV9hQ99D5N6SwsJ/4BiIkG8joxBmTGHtvkWO1ypABQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T07:47:19.984230Z"},"content_sha256":"346893185b01d6adbbad1932de5b10e9e920a2df18d9d3a60d999ef5a86427d1","schema_version":"1.0","event_id":"sha256:346893185b01d6adbbad1932de5b10e9e920a2df18d9d3a60d999ef5a86427d1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:LJ6JEZDOU6F7TD4UKUSXX62LJA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Calibration of a neural network ocean closure for improved mean state and variability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Calibrating a neural network parameterization of mesoscale eddies with ensemble inversion reduces errors in coarse ocean model mean state and variability by a factor of two.","cross_cats":["cs.LG","physics.comp-ph"],"primary_cat":"physics.ao-ph","authors_text":"Alistair Adcroft, Laure Zanna, Pavel Perezhogin","submitted_at":"2026-04-07T19:32:25Z","abstract_excerpt":"Global ocean models exhibit biases in the mean state and variability, particularly at coarse resolution, where mesoscale eddies are unresolved. To address these biases, parameterization coefficients are typically tuned ad hoc. Here, we formulate parameter tuning as a calibration problem using Ensemble Kalman Inversion (EKI). We optimize parameters of a neural network parameterization of mesoscale eddies in two idealized ocean models at coarse resolution. The calibrated parameterization reduces errors by factors of 1.7-3.3 in the time-averaged fluid interfaces and their variability compared to "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The calibrated parameterization reduces errors in the time-averaged fluid interfaces and their variability by approximately a factor of two compared to the unparameterized model or the offline-trained parameterization.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the neural network form can faithfully represent the net effect of mesoscale eddies once its parameters are calibrated, and that the chosen target statistics from idealized models remain informative when transferred to global configurations.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Calibrating a neural network mesoscale eddy parameterization via Ensemble Kalman Inversion halves errors in mean state and variability for coarse ocean models.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Calibrating a neural network parameterization of mesoscale eddies with ensemble inversion reduces errors in coarse ocean model mean state and variability by a factor of two.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"bf530cfd584307ef4ab08ce21563499f51195452ff6e5b6d0cdafa6b16109f8d"},"source":{"id":"2604.06398","kind":"arxiv","version":2},"verdict":{"id":"36ca1ed6-b1f9-401b-87e9-91db1073b54a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T17:57:16.839387Z","strongest_claim":"The calibrated parameterization reduces errors in the time-averaged fluid interfaces and their variability by approximately a factor of two compared to the unparameterized model or the offline-trained parameterization.","one_line_summary":"Calibrating a neural network mesoscale eddy parameterization via Ensemble Kalman Inversion halves errors in mean state and variability for coarse ocean models.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the neural network form can faithfully represent the net effect of mesoscale eddies once its parameters are calibrated, and that the chosen target statistics from idealized models remain informative when transferred to global configurations.","pith_extraction_headline":"Calibrating a neural network parameterization of mesoscale eddies with ensemble inversion reduces errors in coarse ocean model mean state and variability by a factor of two."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.06398/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"18035ead48975735df5805178b87aa17d6ff9f2702de72883d55707201627e71"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"36ca1ed6-b1f9-401b-87e9-91db1073b54a"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"StcQroMxhhuAOwMxpVb+zVxrCJw/+gVMzMqTr/zHT8vtqm/3wgUT78cHYM2XKefRin4AttrgKu94a9s/d492BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T07:47:19.985074Z"},"content_sha256":"bb1592ed8ab003e76d9a1ca887b2f476833e15e2bd01be2e91dbd8a32f3fea08","schema_version":"1.0","event_id":"sha256:bb1592ed8ab003e76d9a1ca887b2f476833e15e2bd01be2e91dbd8a32f3fea08"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LJ6JEZDOU6F7TD4UKUSXX62LJA/bundle.json","state_url":"https://pith.science/pith/LJ6JEZDOU6F7TD4UKUSXX62LJA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LJ6JEZDOU6F7TD4UKUSXX62LJA/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-31T07:47:19Z","links":{"resolver":"https://pith.science/pith/LJ6JEZDOU6F7TD4UKUSXX62LJA","bundle":"https://pith.science/pith/LJ6JEZDOU6F7TD4UKUSXX62LJA/bundle.json","state":"https://pith.science/pith/LJ6JEZDOU6F7TD4UKUSXX62LJA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LJ6JEZDOU6F7TD4UKUSXX62LJA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:LJ6JEZDOU6F7TD4UKUSXX62LJA","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2e77b3c36d386097da71ad6f1729e3c7fe8e9ddd1fac6c476fdeea743d5e6b1b","cross_cats_sorted":["cs.LG","physics.comp-ph"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"physics.ao-ph","submitted_at":"2026-04-07T19:32:25Z","title_canon_sha256":"8264a9d39757dc2e959cf7e49bdfd13c3dc2141ad5193e2b5a86a9a21ce7d594"},"schema_version":"1.0","source":{"id":"2604.06398","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.06398","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"arxiv_version","alias_value":"2604.06398v2","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.06398","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"pith_short_12","alias_value":"LJ6JEZDOU6F7","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"pith_short_16","alias_value":"LJ6JEZDOU6F7TD4U","created_at":"2026-05-20T00:04:30Z"},{"alias_kind":"pith_short_8","alias_value":"LJ6JEZDO","created_at":"2026-05-20T00:04:30Z"}],"graph_snapshots":[{"event_id":"sha256:bb1592ed8ab003e76d9a1ca887b2f476833e15e2bd01be2e91dbd8a32f3fea08","target":"graph","created_at":"2026-05-20T00:04:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"The calibrated parameterization reduces errors in the time-averaged fluid interfaces and their variability by approximately a factor of two compared to the unparameterized model or the offline-trained parameterization."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the neural network form can faithfully represent the net effect of mesoscale eddies once its parameters are calibrated, and that the chosen target statistics from idealized models remain informative when transferred to global configurations."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Calibrating a neural network mesoscale eddy parameterization via Ensemble Kalman Inversion halves errors in mean state and variability for coarse ocean models."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Calibrating a neural network parameterization of mesoscale eddies with ensemble inversion reduces errors in coarse ocean model mean state and variability by a factor of two."}],"snapshot_sha256":"bf530cfd584307ef4ab08ce21563499f51195452ff6e5b6d0cdafa6b16109f8d"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"18035ead48975735df5805178b87aa17d6ff9f2702de72883d55707201627e71"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.06398/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Global ocean models exhibit biases in the mean state and variability, particularly at coarse resolution, where mesoscale eddies are unresolved. To address these biases, parameterization coefficients are typically tuned ad hoc. Here, we formulate parameter tuning as a calibration problem using Ensemble Kalman Inversion (EKI). We optimize parameters of a neural network parameterization of mesoscale eddies in two idealized ocean models at coarse resolution. The calibrated parameterization reduces errors by factors of 1.7-3.3 in the time-averaged fluid interfaces and their variability compared to ","authors_text":"Alistair Adcroft, Laure Zanna, Pavel Perezhogin","cross_cats":["cs.LG","physics.comp-ph"],"headline":"Calibrating a neural network parameterization of mesoscale eddies with ensemble inversion reduces errors in coarse ocean model mean state and variability by a factor of two.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"physics.ao-ph","submitted_at":"2026-04-07T19:32:25Z","title":"Calibration of a neural network ocean closure for improved mean state and variability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.06398","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T17:57:16.839387Z","id":"36ca1ed6-b1f9-401b-87e9-91db1073b54a","model_set":{"reader":"grok-4.3"},"one_line_summary":"Calibrating a neural network mesoscale eddy parameterization via Ensemble Kalman Inversion halves errors in mean state and variability for coarse ocean models.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Calibrating a neural network parameterization of mesoscale eddies with ensemble inversion reduces errors in coarse ocean model mean state and variability by a factor of two.","strongest_claim":"The calibrated parameterization reduces errors in the time-averaged fluid interfaces and their variability by approximately a factor of two compared to the unparameterized model or the offline-trained parameterization.","weakest_assumption":"That the neural network form can faithfully represent the net effect of mesoscale eddies once its parameters are calibrated, and that the chosen target statistics from idealized models remain informative when transferred to global configurations."}},"verdict_id":"36ca1ed6-b1f9-401b-87e9-91db1073b54a"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:346893185b01d6adbbad1932de5b10e9e920a2df18d9d3a60d999ef5a86427d1","target":"record","created_at":"2026-05-20T00:04:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2e77b3c36d386097da71ad6f1729e3c7fe8e9ddd1fac6c476fdeea743d5e6b1b","cross_cats_sorted":["cs.LG","physics.comp-ph"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"physics.ao-ph","submitted_at":"2026-04-07T19:32:25Z","title_canon_sha256":"8264a9d39757dc2e959cf7e49bdfd13c3dc2141ad5193e2b5a86a9a21ce7d594"},"schema_version":"1.0","source":{"id":"2604.06398","kind":"arxiv","version":2}},"canonical_sha256":"5a7c92646ea78bf98f9455257bfb4b483dfc47d306996a62f4e89b7864224030","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5a7c92646ea78bf98f9455257bfb4b483dfc47d306996a62f4e89b7864224030","first_computed_at":"2026-05-20T00:04:30.882839Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:30.882839Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2vVUE8qJo1Mu1qn3ZfBKa8rgKCBiLWzahhQu64fq12fwqPIf3kEfAwOmF4uH7lHvrO+ql+5tj9UOisdocbuhAg==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:30.883838Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.06398","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:346893185b01d6adbbad1932de5b10e9e920a2df18d9d3a60d999ef5a86427d1","sha256:bb1592ed8ab003e76d9a1ca887b2f476833e15e2bd01be2e91dbd8a32f3fea08"],"state_sha256":"c3c89e2389000147f1b5e3c1b8aeac9b4f62482f34606b8c02ec6b9775a4f023"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1Xlzvvu9x60cJ9YSahp1vFVKVdQ/QdhucW1eH2SLDElTkpwl0r5IjNE709921MKyZgpE9EaJJobInDc55qwqBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T07:47:19.988645Z","bundle_sha256":"be99fe2d38d73620eefd2fc1d407fbc10fc0a89c468f856dd361137360543d3d"}}