{"id":"W4406881213","doi":"10.5194/essd-17-259-2025","title":"BCUB – a large-sample ungauged basin attribute dataset for British Columbia, Canada","year":2025,"lang":"en","type":"article","venue":"Earth system science data","topic":"Hydrocarbon exploration and reservoir analysis","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Sample (material); Structural basin; Geology; Environmental science; Paleontology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005280836,0.001468121,0.0008931427,0.00415247,0.00223415,0.001886021,0.00285433,0.001258165,0.01175682],"category_scores_gemma":[0.00429128,0.0005079313,0.0008483968,0.01043174,0.0007131392,0.0008172713,0.001577371,0.00149312,0.01047949],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.01014479,"about_ca_system_score_gemma":0.01865627,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.9429365,"about_ca_topic_score_gemma":0.9681355,"domain_scores_codex":[0.9992423,0.00005716497,0.00006550847,0.0001954677,0.0002467143,0.0001928705],"domain_scores_gemma":[0.9973856,0.0002542205,0.0001023817,0.0003817863,0.001500669,0.0003753225],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001162829,0.00004943757,0.008695981,0.0003973881,0.00009829274,0.0001178948,0.000120342,0.001952492,0.000385437,0.0008472658,0.9773651,0.009854076],"study_design_scores_gemma":[0.0002412778,0.00002387368,0.08556817,0.0006814966,0.0001158669,0.0001963483,0.0008709045,0.01179176,0.001348593,0.00164603,0.8973603,0.0001554787],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00305345,0.0001716495,0.000185711,0.0001102185,0.00002025779,0.0000263501,0.9947925,0.0006121442,0.001027727],"genre_scores_gemma":[0.004268742,0.00009288942,0.0005143702,0.00003472414,0.000003009631,0.00005700559,0.9943566,0.00005416377,0.0006185556],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.05706352,"threshold_uncertainty_score":0.1147991,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01777164094568939,"score_gpt":0.2415731403703857,"score_spread":0.2238014994246963,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}