{"id":"W4415011885","doi":"10.22541/essoar.172798936.65411263/v3","title":"Adjustment of decadal ocean carbon sink predictions using deep learning","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Atmospheric and Environmental Gas Dynamics","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Artificial neural network; Carbon flux; Carbon sink; Earth system science; Sink (geography); Deep learning; Climate system; Carbon cycle","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006770655,0.0005351227,0.0002804125,0.0002918761,0.00021337,0.0005604391,0.0005698288,0.000488507,0.001063906],"category_scores_gemma":[0.003119154,0.0003336164,0.0001947868,0.0002695917,0.0002478146,0.000640263,0.00074569,0.0009185362,0.000296877],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005748459,"about_ca_system_score_gemma":0.0008518262,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0126703,"about_ca_topic_score_gemma":0.01681522,"domain_scores_codex":[0.9998722,0.0000205763,0.00000793555,0.00004810702,0.00003007444,0.00002110129],"domain_scores_gemma":[0.9995122,0.000159064,0.000073541,0.00007913905,0.0001463296,0.00002980822],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00008637135,0.0000657811,0.007666388,0.00002204113,0.00005310609,0.00004273474,0.00003552328,0.9250826,0.005538484,0.001367864,0.001387668,0.05865145],"study_design_scores_gemma":[0.000003358635,0.000005167878,0.0006212937,0.000001366337,0.000002089497,0.000002415755,0.000002118748,0.9976069,0.001116856,0.0004592084,0.0001763343,0.000002933569],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5805737,0.0003242738,0.4064596,0.0007435173,0.000301817,0.00004234447,0.0009716601,0.004886223,0.005696925],"genre_scores_gemma":[0.9530678,0.00006841683,0.04427048,0.0001001252,0.00002953567,0.0000241194,0.0006447434,0.0001486369,0.001646131],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0126703,"threshold_uncertainty_score":0.0251931,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008725802284131046,"score_gpt":0.2215751452318005,"score_spread":0.2128493429476694,"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."}}