{"doi":"10.13016/m2hows-tdmj","title":"Deep Learning Approaches for Cloud Property Retrieval: Leveraging Geospatial Foundation Models and Multitask Frameworks","abstract":"With the rapid growth of Earth-observation datasets, geospatial foundation models (FMs) provide a scalable approach to learn transferable features across diverse satellite sensor data. However, their cross-sensor adaptation ability needs more exploration. To study this issue, we present a benchmarking study of SatVision-TOA, an FM pre-trained on over 20 years of MODIS data, when adapted to the GOES NOAA ABI sensor for four downstream cloud properties: cloud mask, cloud phase (segmentation), and cloud optical depth (COD) and cloud particle size (CPS) (regression). We propose a multi-task learning fine-tuning pipeline with a U-Net-based decoder and a lightweight preprocessor to address band-mismatch handling (14 MODIS bands for pre-training vs. 16 ABI bands for fine-tuning). To evaluate our pipeline, we benchmark fine-tuned models against from-scratch baselines, evaluate full fine-tuning (FFT) versus parameter-efficient fine-tuning (PEFT) methods (LoRA, VPT), and compare 14-band versus 16-band inputs. Our experiments show that multi-task learning improves efficiency and predictive quality in both fine-tuned and from-scratch settings. For the other four comparisons (FT vs. from-scratch, FFT vs. PEFT, 14-bands vs. 16-bands and loss functions), the results are mixed and there is no setup that always performs the best for all segmentation and regression tasks.","journal":"Open MIND","year":2025,"id":587924,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":0,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.952,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":5935,"name":"Kevin Zhang","orcid":null,"position":1,"is_corresponding":false},{"id":1504230,"name":"Caleb Parten","orcid":null,"position":2,"is_corresponding":false},{"id":1504231,"name":"Autumn Sterling","orcid":null,"position":3,"is_corresponding":false},{"id":1504232,"name":"Haoxiang Zhang","orcid":null,"position":4,"is_corresponding":false},{"id":1022771,"name":"Xingyan Li","orcid":"0009-0004-5603-4530","position":5,"is_corresponding":false},{"id":1504116,"name":"Jordan A. Caraballo‐Vega","orcid":"0000-0001-9125-5591","position":6,"is_corresponding":false},{"id":1401818,"name":"Jie Gong","orcid":"0000-0001-5897-7023","position":7,"is_corresponding":false},{"id":1504233,"name":"Mark Carroll","orcid":null,"position":8,"is_corresponding":false},{"id":1338816,"name":"Jianwu Wang","orcid":"0000-0002-4421-8058","position":9,"is_corresponding":false},{"id":230285,"name":"Danielle Murphy","orcid":"0000-0003-1448-9444","position":0,"is_corresponding":true}],"reference_count":0,"raw_metadata":null,"created_at":"2026-07-19T02:59:43.096742Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}