{"id":"W4417062489","doi":"10.2139/ssrn.5871350","title":"Adaptive Machine Learning for Chlorophyll-a Retrieval in Optically Complex Lakes: Enhancing Performance and Exposing Transferability Limits","year":2025,"lang":"","type":"preprint","venue":"SSRN Electronic Journal","topic":"Marine and coastal ecosystems","field":"Earth and Planetary Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University","funders":"","keywords":"Random forest; Transferability; Support vector machine; Artificial neural network; Boosting (machine learning); Scalability; Colored dissolved organic matter; Interoperability","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"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":[0.001627385,0.0004985732,0.0004475279,0.0003404646,0.0003212904,0.0006371925,0.0006887089,0.001189137,0.00124707],"category_scores_gemma":[0.00695979,0.0001993289,0.0002976583,0.0004685071,0.0005210693,0.001576253,0.001094563,0.0009330898,0.000434858],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003590228,"about_ca_system_score_gemma":0.0005962959,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003861393,"about_ca_topic_score_gemma":0.002687451,"domain_scores_codex":[0.9996878,0.0001290962,0.0000199155,0.00007803511,0.00004722235,0.00003798521],"domain_scores_gemma":[0.996569,0.002729732,0.0001327005,0.0002824941,0.000226657,0.00005930468],"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":[0.0009732601,0.0003797029,0.008669947,0.000244482,0.0001574319,0.00008648446,0.0002443135,0.4631574,0.06919611,0.004329408,0.001453616,0.4511079],"study_design_scores_gemma":[0.00001377038,0.00005817796,0.0008551728,0.000003792887,0.000008995833,0.00001304706,0.00001414886,0.9906075,0.006546928,0.001765869,0.000105997,0.000006540846],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.468752,0.0006350972,0.5259874,0.0006507157,0.00007273446,0.00007840656,0.0001071575,0.00106375,0.002652792],"genre_scores_gemma":[0.921859,0.0001555655,0.07630704,0.0001157542,0.00005231476,0.0000429562,0.0001136429,0.00005950793,0.001294245],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003861393,"threshold_uncertainty_score":0.008606553,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01800940117346582,"score_gpt":0.225985856666864,"score_spread":0.2079764554933982,"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."}}