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.

中文翻译：

---

GNSS-R风速反演的深度残差全连接网络及其解释


全球导航卫星系统反射测量（GNSS-R）已成为一种为海洋风速反演提供L波段双基地​​测量的新技术，通常使用传统的地球物理模型函数（GMF）或浅层神经网络（NN）。然而，识别和考虑 GMF 中的所有相关参数仍然具有挑战性。同时，神经网络模型由于退化问题而面临局限性，这限制了它们的深度，从而限制了它们的性能。此外，GNSS-R 风反演的神经网络模型的解释是另一个问题。为此，我们提出了一种残差全连接网络（RFCN），融合了几何形状、接收器增益、有效波高和当前速度等辅助信息，并进行了逐轨校正。参考欧洲中期天气预报中心（ECMWF）ERA5风产品，RFCN风的均方根误差（RMSE）和偏差分别为1.031 m/s和-0.0003 m/s，改善了6% RMSE 与去偏 NOAA 旋风全球导航卫星系统 (CYGNSS) 1.2 版 (V1.2) 风速反演相比。此外，在大流速的热带辐合带（ITCZ）区域，RMSE和偏差分别为1.006 m/s和-0.022 m/s：与去偏差的NOAA CYGNSS V1.2风相比分别提高了11.6%和87.9%。这些区域的偏差“条带”几乎被消除。每日平均误差分析还表明 RFCN 风更加强劲并且与 ECMWF 风一致。对于大于20 m/s的风速，参考土壤湿度主动被动（SMAP）3级最终风产品，微调RFCN（FT_RFCN）风的RMSE和偏差比NOAA风降低了25.7%和91.5% 。 最后，2021-2022 年期间通过步进频率微波辐射计 (SFMR) 测量的热带气旋的 RMSE 和反演偏差显示，与 NOAA 风相比，热带气旋的 RMSE 和反演偏差分别提高了 3.5% 和 21.2%。通过为 RFCN 和 FT_RFCN 开发的 SHapley Additive exPlanations (SHAP) 模型，可以定量评估每个特征的贡献，同时提供对它们在具有明确物理意义的“黑盒”神经网络模型中相互作用的见解。