date: 2026-07-07T12:16:17Z
pdf:PDFVersion: 1.7
pdf:docinfo:title: Filling Satellite Microwave Observation Gaps via Generative Synthesis
xmp:CreatorTool: LaTeX with hyperref
access_permission:can_print_degraded: true
subject: Polar-orbiting microwave radiometers provide indispensable all-weather measurements of the atmospheric state, yet revisit intervals of many hours leave critical gaps during rapidly evolving weather events. To address this limitation, we developed MIDAS (Microwave Inference via Diffusion Across Satellites), a probabilistic framework that estimates microwave brightness temperature (BT) fields across the geostationary full-disk domain from infrared observations at 10 min intervals. This study focuses on the five Microwave Humidity Sounder-2 (MWHS-2) humidity-sounding channels near 183 GHz, which provide vertically resolved water vapor information. MIDAS achieves relative errors below 0.5% for the majority of cases, with a channel-averaged mean absolute error of 1.15 K, outperforming a deterministic U-Net baseline (1.43 K). Beyond per-sample evaluation, MIDAS reproduces large-scale climatological patterns across the full-disk domain over a three-month summer period, consistent with Radiative Transfer for TOVS?Scattering (RTTOV-SCATT) simulations. In deep convective scenes where reconstruction is most difficult, the ensemble spread naturally tracks reconstruction difficulty, providing a built-in indicator of prediction confidence. Notably, MIDAS incorporates real-time polar-orbiting observations as physical constraints via a merge-sampling mechanism, reducing ensemble RMSE by over 20% and improving probabilistic calibration by more than 30%. Proof-of-concept assimilation experiments for two high-impact weather cases show that MIDAS-generated fields yield forecast improvements comparable to those from real satellite observations, reducing tropical cyclone track errors from approximately 110 km to 40 km and improving heavy precipitation forecasts at extreme rainfall thresholds where direct infrared assimilation shows no benefit. Overall, our framework demonstrates the potential of generative models to supplement sparse observational coverage and provide physically plausible microwave humidity fields for downstream applications.
dc:format: application/pdf; version=1.7
pdf:docinfo:creator_tool: LaTeX with hyperref
access_permission:fill_in_form: true
pdf:encrypted: false
dc:title: Filling Satellite Microwave Observation Gaps via Generative Synthesis
modified: 2026-07-07T12:16:17Z
cp:subject: Polar-orbiting microwave radiometers provide indispensable all-weather measurements of the atmospheric state, yet revisit intervals of many hours leave critical gaps during rapidly evolving weather events. To address this limitation, we developed MIDAS (Microwave Inference via Diffusion Across Satellites), a probabilistic framework that estimates microwave brightness temperature (BT) fields across the geostationary full-disk domain from infrared observations at 10 min intervals. This study focuses on the five Microwave Humidity Sounder-2 (MWHS-2) humidity-sounding channels near 183 GHz, which provide vertically resolved water vapor information. MIDAS achieves relative errors below 0.5% for the majority of cases, with a channel-averaged mean absolute error of 1.15 K, outperforming a deterministic U-Net baseline (1.43 K). Beyond per-sample evaluation, MIDAS reproduces large-scale climatological patterns across the full-disk domain over a three-month summer period, consistent with Radiative Transfer for TOVS?Scattering (RTTOV-SCATT) simulations. In deep convective scenes where reconstruction is most difficult, the ensemble spread naturally tracks reconstruction difficulty, providing a built-in indicator of prediction confidence. Notably, MIDAS incorporates real-time polar-orbiting observations as physical constraints via a merge-sampling mechanism, reducing ensemble RMSE by over 20% and improving probabilistic calibration by more than 30%. Proof-of-concept assimilation experiments for two high-impact weather cases show that MIDAS-generated fields yield forecast improvements comparable to those from real satellite observations, reducing tropical cyclone track errors from approximately 110 km to 40 km and improving heavy precipitation forecasts at extreme rainfall thresholds where direct infrared assimilation shows no benefit. Overall, our framework demonstrates the potential of generative models to supplement sparse observational coverage and provide physically plausible microwave humidity fields for downstream applications.
pdf:docinfo:subject: Polar-orbiting microwave radiometers provide indispensable all-weather measurements of the atmospheric state, yet revisit intervals of many hours leave critical gaps during rapidly evolving weather events. To address this limitation, we developed MIDAS (Microwave Inference via Diffusion Across Satellites), a probabilistic framework that estimates microwave brightness temperature (BT) fields across the geostationary full-disk domain from infrared observations at 10 min intervals. This study focuses on the five Microwave Humidity Sounder-2 (MWHS-2) humidity-sounding channels near 183 GHz, which provide vertically resolved water vapor information. MIDAS achieves relative errors below 0.5% for the majority of cases, with a channel-averaged mean absolute error of 1.15 K, outperforming a deterministic U-Net baseline (1.43 K). Beyond per-sample evaluation, MIDAS reproduces large-scale climatological patterns across the full-disk domain over a three-month summer period, consistent with Radiative Transfer for TOVS?Scattering (RTTOV-SCATT) simulations. In deep convective scenes where reconstruction is most difficult, the ensemble spread naturally tracks reconstruction difficulty, providing a built-in indicator of prediction confidence. Notably, MIDAS incorporates real-time polar-orbiting observations as physical constraints via a merge-sampling mechanism, reducing ensemble RMSE by over 20% and improving probabilistic calibration by more than 30%. Proof-of-concept assimilation experiments for two high-impact weather cases show that MIDAS-generated fields yield forecast improvements comparable to those from real satellite observations, reducing tropical cyclone track errors from approximately 110 km to 40 km and improving heavy precipitation forecasts at extreme rainfall thresholds where direct infrared assimilation shows no benefit. Overall, our framework demonstrates the potential of generative models to supplement sparse observational coverage and provide physically plausible microwave humidity fields for downstream applications.
pdf:docinfo:creator: Han Du, Baoxiang Pan, Fan Ping, Jin Xu, Congyi Nai, Sencan Sun, Jie Chao, Jingnan Wang, Shangshang Yang, Xi Chen, Jingyuan Li, Jiahua Mao, Lei Yin, Yupeng Li and Ziniu Xiao
PTEX.Fullbanner: This is pdfTeX, Version 3.141592653-2.6-1.40.25 (TeX Live 2023) kpathsea version 6.3.5
meta:author: Han Du, Baoxiang Pan, Fan Ping, Jin Xu, Congyi Nai, Sencan Sun, Jie Chao, Jingnan Wang, Shangshang Yang, Xi Chen, Jingyuan Li, Jiahua Mao, Lei Yin, Yupeng Li and Ziniu Xiao
trapped: False
meta:creation-date: 2026-07-07T12:08:27Z
created: 2026-07-07T12:08:27Z
access_permission:extract_for_accessibility: true
Creation-Date: 2026-07-07T12:08:27Z
Author: Han Du, Baoxiang Pan, Fan Ping, Jin Xu, Congyi Nai, Sencan Sun, Jie Chao, Jingnan Wang, Shangshang Yang, Xi Chen, Jingyuan Li, Jiahua Mao, Lei Yin, Yupeng Li and Ziniu Xiao
producer: pdfTeX-1.40.25; modified using OpenPDF 1.4.2
pdf:docinfo:producer: pdfTeX-1.40.25; modified using OpenPDF 1.4.2
pdf:unmappedUnicodeCharsPerPage: 0
Keywords: microwave radiometer; generative model; cross-platform synthesis; observation gap-filling; assimilation
access_permission:modify_annotations: true
dc:creator: Han Du, Baoxiang Pan, Fan Ping, Jin Xu, Congyi Nai, Sencan Sun, Jie Chao, Jingnan Wang, Shangshang Yang, Xi Chen, Jingyuan Li, Jiahua Mao, Lei Yin, Yupeng Li and Ziniu Xiao
dcterms:created: 2026-07-07T12:08:27Z
Last-Modified: 2026-07-07T12:16:17Z
dcterms:modified: 2026-07-07T12:16:17Z
title: Filling Satellite Microwave Observation Gaps via Generative Synthesis
Last-Save-Date: 2026-07-07T12:16:17Z
pdf:docinfo:keywords: microwave radiometer; generative model; cross-platform synthesis; observation gap-filling; assimilation
pdf:docinfo:modified: 2026-07-07T12:16:17Z
meta:save-date: 2026-07-07T12:16:17Z
pdf:docinfo:custom:PTEX.Fullbanner: This is pdfTeX, Version 3.141592653-2.6-1.40.25 (TeX Live 2023) kpathsea version 6.3.5
Content-Type: application/pdf
X-Parsed-By: org.apache.tika.parser.DefaultParser
creator: Han Du, Baoxiang Pan, Fan Ping, Jin Xu, Congyi Nai, Sencan Sun, Jie Chao, Jingnan Wang, Shangshang Yang, Xi Chen, Jingyuan Li, Jiahua Mao, Lei Yin, Yupeng Li and Ziniu Xiao
dc:subject: microwave radiometer; generative model; cross-platform synthesis; observation gap-filling; assimilation
access_permission:assemble_document: true
xmpTPg:NPages: 37
pdf:charsPerPage: 3155
access_permission:extract_content: true
access_permission:can_print: true
pdf:docinfo:trapped: False
meta:keyword: microwave radiometer; generative model; cross-platform synthesis; observation gap-filling; assimilation
access_permission:can_modify: true
pdf:docinfo:created: 2026-07-07T12:08:27Z