{"id":"W4408195974","doi":"10.1038/s41597-025-04715-4","title":"A Comprehensive Dataset of Surface Water Quality Spanning 1940-2023 for Empirical and ML Adopted Research","year":2025,"lang":"en","type":"article","venue":"Scientific Data","topic":"Water Quality and Pollution Assessment","field":"Environmental Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Water quality; Quality (philosophy); Environmental science; Surface water; Data science; Computer science; Biology; Ecology; Environmental engineering; Physics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005678686,0.00009186316,0.0001735714,0.00006929688,0.000517057,0.0001742582,0.0009252465,0.000050531,0.0003872107],"category_scores_gemma":[0.0001532028,0.00006934122,0.00001989155,0.0004598481,0.001295996,0.0003411169,0.003031545,0.0001367354,0.0001526567],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005318153,"about_ca_system_score_gemma":0.00004145686,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008667393,"about_ca_topic_score_gemma":0.00023038,"domain_scores_codex":[0.9975587,0.0003940618,0.0003681617,0.0007744992,0.0005303436,0.0003742843],"domain_scores_gemma":[0.9980153,0.0002932935,0.00005172222,0.001504522,0.00004616458,0.00008896823],"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":[0.00006211922,0.0001339671,0.009434135,0.0001346398,0.00002128539,0.000001116098,0.0006840425,0.0001160709,0.03352586,0.0001883096,0.9547334,0.0009650591],"study_design_scores_gemma":[0.0003916841,0.00002530066,0.03231186,0.00002841531,0.00001248932,6.060114e-7,0.0005230276,0.002148731,0.01007305,0.001126977,0.9532478,0.0001100161],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9394497,0.00007021656,0.001851276,0.004191236,0.0007366185,0.0006579733,0.05234741,0.00002296735,0.0006726017],"genre_scores_gemma":[0.9353229,0.000009236838,0.009065219,0.0004667519,0.00003117523,0.00001711138,0.04696737,0.00001056841,0.008109631],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02345281,"threshold_uncertainty_score":0.4775151,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4412245770869852,"score_gpt":0.520185085393014,"score_spread":0.0789605083060288,"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."}}