{"title":"用于 GNSS-R 风速检索的深度残差全连接网络及其解释","authors":"","doi":"10.1016/j.rse.2024.114375","DOIUrl":null,"url":null,"abstract":"<div><p>Global navigation satellite system reflectometry (GNSS-R) has emerged as a new technique to provide L-band bistatic measurements for ocean wind speed retrieval, in which traditional geophysical model functions (GMFs) or shallow neural networks (NNs) are normally used. However, it is still challenging to identify and consider all relevant parameters in the GMF. Meanwhile, NN models face limitations due to the degradation problem, which restricts their depth and consequently their performance. Furthermore, the interpretation of NN models for GNSS-R wind retrieval is another issue. To this end, we propose a residual fully connected network (RFCN) fusing auxiliary information such as geometry, receiver gain, significant wave height, and current speed with track-wise corrected <span><math><msub><mrow><mi>σ</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span>. Referred to the European Centre for Medium-Range Weather Forecast (ECMWF) ERA5 wind product, the root mean square error (RMSE) and bias of RFCN winds are 1.031 m/s and -0.0003 m/s, respectively, with a 6% improvement in RMSE compared to debiased NOAA Cyclone Global Navigation Satellite System (CYGNSS) Version 1.2 (V1.2) wind speed retrieval. Moreover, in an intertropical convergence zone (ITCZ) area with large current speeds, the RMSE and bias are 1.006 m/s and -0.022 m/s: an improvement of 11.6% and 87.9% compared to debiased NOAA CYGNSS V1.2 winds. The bias ‘strips’ in these areas are nearly eliminated. Daily averaged error analyses also demonstrate that RFCN winds are more robust and consistent with ECMWF winds. For wind speeds larger than 20 m/s, referred to Soil Moisture Active Passive (SMAP) Level 3 final wind products, the RMSE and bias of fine-tuning RFCN (FT_RFCN) winds are reduced by 25.7% and 91.5% compared to NOAA winds. Finally, the RMSE and bias of retrievals in tropical cyclones, measured by Stepped Frequency Microwave Radiometer (SFMR) during 2021-2022, reveal an improvement of 3.5% and 21.2% compared to NOAA winds. Through SHapley Additive exPlanations (SHAP) models developed for RFCN and FT_RFCN, the contribution of each feature is quantitatively evaluated, while providing insights into their interactions within the ‘black-box’ NN models with clear physical meanings.</p></div>","PeriodicalId":417,"journal":{"name":"Remote Sensing of Environment","volume":null,"pages":null},"PeriodicalIF":11.1000,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0034425724004012/pdfft?md5=a2ed0e9eca433424ca0050057487e60b&pid=1-s2.0-S0034425724004012-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Deep residual fully connected network for GNSS-R wind speed retrieval and its interpretation\",\"authors\":\"\",\"doi\":\"10.1016/j.rse.2024.114375\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Global navigation satellite system reflectometry (GNSS-R) has emerged as a new technique to provide L-band bistatic measurements for ocean wind speed retrieval, in which traditional geophysical model functions (GMFs) or shallow neural networks (NNs) are normally used. However, it is still challenging to identify and consider all relevant parameters in the GMF. Meanwhile, NN models face limitations due to the degradation problem, which restricts their depth and consequently their performance. Furthermore, the interpretation of NN models for GNSS-R wind retrieval is another issue. To this end, we propose a residual fully connected network (RFCN) fusing auxiliary information such as geometry, receiver gain, significant wave height, and current speed with track-wise corrected <span><math><msub><mrow><mi>σ</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span>. Referred to the European Centre for Medium-Range Weather Forecast (ECMWF) ERA5 wind product, the root mean square error (RMSE) and bias of RFCN winds are 1.031 m/s and -0.0003 m/s, respectively, with a 6% improvement in RMSE compared to debiased NOAA Cyclone Global Navigation Satellite System (CYGNSS) Version 1.2 (V1.2) wind speed retrieval. Moreover, in an intertropical convergence zone (ITCZ) area with large current speeds, the RMSE and bias are 1.006 m/s and -0.022 m/s: an improvement of 11.6% and 87.9% compared to debiased NOAA CYGNSS V1.2 winds. The bias ‘strips’ in these areas are nearly eliminated. Daily averaged error analyses also demonstrate that RFCN winds are more robust and consistent with ECMWF winds. For wind speeds larger than 20 m/s, referred to Soil Moisture Active Passive (SMAP) Level 3 final wind products, the RMSE and bias of fine-tuning RFCN (FT_RFCN) winds are reduced by 25.7% and 91.5% compared to NOAA winds. Finally, the RMSE and bias of retrievals in tropical cyclones, measured by Stepped Frequency Microwave Radiometer (SFMR) during 2021-2022, reveal an improvement of 3.5% and 21.2% compared to NOAA winds. Through SHapley Additive exPlanations (SHAP) models developed for RFCN and FT_RFCN, the contribution of each feature is quantitatively evaluated, while providing insights into their interactions within the ‘black-box’ NN models with clear physical meanings.</p></div>\",\"PeriodicalId\":417,\"journal\":{\"name\":\"Remote Sensing of Environment\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":11.1000,\"publicationDate\":\"2024-08-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S0034425724004012/pdfft?md5=a2ed0e9eca433424ca0050057487e60b&pid=1-s2.0-S0034425724004012-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Remote Sensing of Environment\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0034425724004012\",\"RegionNum\":1,\"RegionCategory\":\"地球科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENVIRONMENTAL SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Remote Sensing of Environment","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0034425724004012","RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
Deep residual fully connected network for GNSS-R wind speed retrieval and its interpretation
Global navigation satellite system reflectometry (GNSS-R) has emerged as a new technique to provide L-band bistatic measurements for ocean wind speed retrieval, in which traditional geophysical model functions (GMFs) or shallow neural networks (NNs) are normally used. However, it is still challenging to identify and consider all relevant parameters in the GMF. Meanwhile, NN models face limitations due to the degradation problem, which restricts their depth and consequently their performance. Furthermore, the interpretation of NN models for GNSS-R wind retrieval is another issue. To this end, we propose a residual fully connected network (RFCN) fusing auxiliary information such as geometry, receiver gain, significant wave height, and current speed with track-wise corrected . Referred to the European Centre for Medium-Range Weather Forecast (ECMWF) ERA5 wind product, the root mean square error (RMSE) and bias of RFCN winds are 1.031 m/s and -0.0003 m/s, respectively, with a 6% improvement in RMSE compared to debiased NOAA Cyclone Global Navigation Satellite System (CYGNSS) Version 1.2 (V1.2) wind speed retrieval. Moreover, in an intertropical convergence zone (ITCZ) area with large current speeds, the RMSE and bias are 1.006 m/s and -0.022 m/s: an improvement of 11.6% and 87.9% compared to debiased NOAA CYGNSS V1.2 winds. The bias ‘strips’ in these areas are nearly eliminated. Daily averaged error analyses also demonstrate that RFCN winds are more robust and consistent with ECMWF winds. For wind speeds larger than 20 m/s, referred to Soil Moisture Active Passive (SMAP) Level 3 final wind products, the RMSE and bias of fine-tuning RFCN (FT_RFCN) winds are reduced by 25.7% and 91.5% compared to NOAA winds. Finally, the RMSE and bias of retrievals in tropical cyclones, measured by Stepped Frequency Microwave Radiometer (SFMR) during 2021-2022, reveal an improvement of 3.5% and 21.2% compared to NOAA winds. Through SHapley Additive exPlanations (SHAP) models developed for RFCN and FT_RFCN, the contribution of each feature is quantitatively evaluated, while providing insights into their interactions within the ‘black-box’ NN models with clear physical meanings.
期刊介绍:
Remote Sensing of Environment (RSE) serves the Earth observation community by disseminating results on the theory, science, applications, and technology that contribute to advancing the field of remote sensing. With a thoroughly interdisciplinary approach, RSE encompasses terrestrial, oceanic, and atmospheric sensing.
The journal emphasizes biophysical and quantitative approaches to remote sensing at local to global scales, covering a diverse range of applications and techniques.
RSE serves as a vital platform for the exchange of knowledge and advancements in the dynamic field of remote sensing.