{
  "age_assignment": "Only the 84 committed comparable calibrated age claims on chronology 29939 enter the 500-cal-BP point-age bins. Retain the two refused negative-bound source samples and their count mass separately. Do not reallocate by age bounds or infer a posterior.",
  "bin_width_cal_bp": 500,
  "candidate_bin_ids": [
    0,
    1,
    2,
    3,
    4,
    5,
    6,
    7,
    8,
    9,
    10,
    11,
    12,
    13,
    14,
    15,
    16,
    17,
    18,
    19,
    20,
    21,
    22
  ],
  "claim_scope": "New-site descriptive direct observed-pollen development evidence after scoring; not an untouched independent confirmation, LRA, measured-vegetation accuracy, basin-buffer validation or broad Swedish generalization.",
  "count_admission": "For each of all 86 source sample IDs, admit only Neotoma observations with source_element=pollen, source_element_type=pollen, source_unit=NISP, unit_family=count, detection_status=reported_value, nonnegative integer source_value and a nonblank native source_reported_name. Sum all admitted native pollen NISP. Sum counts whose native label exactly matches one of the 16 literal response labels as closed-axis NISP; outside-axis NISP is native minus closed. Do not reinterpret excluded native labels as absent vegetation.",
  "frozen_predictions_sha256": "844db7dd47a93eba5d0630719df992d211ced59464e917314e2b5ff131d36a17",
  "methods": [
    "training_mean",
    "nearest_site",
    "inverse_distance_squared"
  ],
  "paired_support": "Aggregate every method only over the intersection of bins successfully scored by all three frozen methods; also retain individual method scores and every attempted candidate-bin/method status. No winner selection from this target.",
  "prediction_freeze_sha256": "bc89f334d95b727d2a051efaa794a131f1d13eee0613f1c32f6e6ace67a6e997",
  "prediction_protocol_sha256": "9d04c225b571a98362fca38e1ef0976d19a9eb87e20b7e66e09529f4b09f0699",
  "prepared_input_manifest_sha256": "47d9b2cc54c416373fd6d6ad11722ad63e1c54b2bc8e4af9222a6dd488b62285",
  "prepared_metadata_sha256": "9054bfc34bd4444070f616e5810f180d827bca10a54c09c72133c0374e01dcb5",
  "primary_method": "inverse_distance_squared",
  "primary_metric": "Equal-bin mean absolute error of the 16-label closed observed pollen fractions, using score_pollen; one site, so equal-bin within-site mean is site-macro MAE.",
  "read_order": [
    "verify supplied and committed prediction freeze SHA-256, predictions SHA-256, prepared input-manifest and metadata SHA-256, training profile and receipt SHA-256",
    "verify current relational manifest against committed source hash and all recorded site/core/dataset/sample/age/chronology/control metadata part hashes",
    "verify every recorded observation part SHA-256 against the current relational manifest before interpreting any target taxon observation",
    "parse Neotoma observation parts and select only exact 86 target sample IDs for source count reconciliation and scoring"
  ],
  "refused_age_sample_count": 2,
  "response_taxa": [
    "Alnus",
    "Amaranthaceae",
    "Betula",
    "Cichorioideae",
    "Cyperaceae",
    "Filipendula",
    "Frangula alnus",
    "Nuphar",
    "Plantago lanceolata",
    "Plantago media",
    "Populus",
    "Quercus",
    "Rosaceae undiff.",
    "Salix",
    "Thalictrum",
    "Ulmus"
  ],
  "schema_version": "lilla-gloppsjon-direct-pollen-scoring-protocol.v1",
  "secondary_metric": "Equal-bin mean Hellinger distance on the same 16-label closed observed pollen fractions and same paired bins.",
  "selected_age_sample_count": 84,
  "source_access_history": "The earlier governed training source-validation screen parsed all Neotoma observation partitions, including site 3234, but did not select that site into training or predictions. The freeze was made without interpreting target count outcomes. This scoring protocol is declared before the target outcomes are opened for this consumer.",
  "source_additional_metadata_sha256": {
    "surfaces/chronology_controls/part-00001.json": "bb84e43a97c3ac548505ad34e489c551ce44ab6c072d28aaa88bc287dbeeac84",
    "surfaces/collection_units/part-00001.json": "dea88b9cadd08256b94f537a5d6a48c8b6b6c5820e5f340ae4786c3e28066780"
  },
  "source_relational_manifest_sha256": "6f6777a40b67bd0704eb1a982f19e6d451cad0d6e4a41dfebf5e3f28be33105b",
  "source_sample_count": 86,
  "source_surface_sha256": {
    "manifest.json": "6f6777a40b67bd0704eb1a982f19e6d451cad0d6e4a41dfebf5e3f28be33105b",
    "surfaces/age_claims/part-00001.json": "6b8983b43138ec68895d96070508218d0b7c4f9908bf87860f96b79d2725c558",
    "surfaces/chronologies/part-00001.json": "0703828826bde8a4860d5ea03100cf85145702a042bb3d29dafcb8a3bef4f569",
    "surfaces/datasets/part-00001.json": "be7fccfa6e0284ac7984e64c844618a6d8682c03282296cef5cabaab594e27e9",
    "surfaces/observations/part-00001.json": "6afa864a9e06384678e63e9f85b13ca79882a87f5c14751cb27af46df60539be",
    "surfaces/observations/part-00002.json": "571b55edc7ce7b6472cacc0db41c41c0e5ccb7d11f0890f88e7ba5f99ef3c719",
    "surfaces/observations/part-00003.json": "a96543819902f5cf5b2b2de355558fd1dd85946c2c08254ec84fad6f9e43fcb5",
    "surfaces/observations/part-00004.json": "b78776421961eb1bf64d886340e2e0f5a42dfe6d0a4ca937d028eb66b1ca0219",
    "surfaces/observations/part-00005.json": "c4705ced69897dcb8f154435d0ba6bdfb0baa942642de94b18f3c856cbfc2e3a",
    "surfaces/observations/part-00006.json": "a33597470799a6b85add81bb699c8a32ec6b55881809345f44351f5530d4d2b7",
    "surfaces/observations/part-00007.json": "48a285e8be25f0ffc9d41feb0cd08ac8c2b6ba54783cf7f3ceeb57e4595ef1bf",
    "surfaces/observations/part-00008.json": "517a9b6fd807cd420dbaf7f2d09a5c93d6a4d4374bc823d25fba7307bc0852dd",
    "surfaces/samples/part-00001.json": "c959e77f7c645b814bffe92a92ff2d8fabcc077b4fc4c894e6315e448d14100c",
    "surfaces/sites/part-00001.json": "b6ca4e49d0aec59aee576fc540c77b6c2e21601ce28338227e98717ee0bb93e9"
  },
  "status_rule": "Create one coverage row per 23 candidate bins and one score row per bin and fixed method. Refuse a bin if any age-selected sample assigned to it has no admitted direct pollen count row; separately refuse zero closed-axis count or frozen training-support refusal. Preserve score-pollen validation failures with a reason and null metrics. A missing exact label within a positive observed closed-axis count vector is a zero count on this declared subcomposition only. Preserve every source sample row and native/closed/outside mass; do not silently drop failed bins.",
  "target_chronology_id": "neotoma:chronology:4071:29939",
  "target_core_id": "neotoma:collection-unit:4071",
  "target_dataset_doi": "10.21233/vpnw-pv89",
  "target_dataset_id": "neotoma:dataset:4200",
  "target_site_id": "neotoma:site:3234",
  "training_profile_sha256": "4ec0de377bba7ad501ca5bf141374b4a3bde97ea266e64b0135a639d309a8647",
  "training_receipt_sha256": "4f161a235fd77953fc8a0411d2649ad2d7268144328e20f6acdc0df2c783f381"
}