As extreme forms of tropical cyclones (TCs), typhoons pose significant threats to both human society and the natural environment. To better understand and predict their behavior, scientists have relied on numerical simulations. Current typhoon modeling primarily falls into two categories: (1) complex simulations based on fluid dynamics and thermodynamics, and (2) empirical parameterized models. Most comparative studies on these models have focused on wind speed below 50 m/s, with fewer studies addressing high wind speed (above 50 m/s). In this study, we design and compare four different simulation approaches to model two super typhoons: Typhoon Surigae (2102) and Typhoon Nepartak (1601). These approaches include: (1) The Weather Research and Forecasting (WRF) model simulation driven by NCEP Final Operational Global Analysis data (FNL), (2) WRF simulation driven by the fifth generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis data (ERA5), (3) the empirical parameterized Holland model, and (4) the empirical parameterized Jelesnianski model. The simulated wind fields were compared with the measured wind data from The Soil Moisture Active Passive (SMAP) platform, and the resulting wind fields were then used as inputs for the Simulating WAves Nearshore (SWAN) model to simulate typhoon-induced waves. Our findings are as follows: (1) for high wind speeds, the performance of the empirical models surpasses that of the WRF simulations; (2) using more accurate driving wind data improves the WRF model’s performance in simulating typhoon wind speeds, and WRF simulations excel in representing wind fields in the outer regions of the typhoon; (3) careful adjustment of the maximum wind speed radius parameter is essential for improving the accuracy of the empirical models.

Synthetic aperture radar (SAR) has been widely used for observing sea-surface wind fields (SSWFs), with many scholars having evaluated the performance of SAR in SSWF retrieval. Due to the large systems and high costs of traditional SAR, a tendency towards the development of smaller and more cost-effective SAR systems has emerged. However, to date, there has been no evaluation of the SSWF retrieval performance of miniaturized SAR systems. This study utilized 1053 HiSea-1 and Chaohu-1 miniaturized SAR images covering the Southeast China Sea to retrieve SSWFs. After a quality control procedure, the retrieved winds were subsequently compared with ERA5, buoy, and ASCAT data. The retrieved wind speeds demonstrated root mean square errors (RMSEs) of 2.42 m/s, 1.64 m/s, and 3.29 m/s, respectively, while the mean bias errors (MBEs) were found to be −0.44 m/s, 1.08 m/s, and −1.65 m/s, respectively. Furthermore, the retrieved wind directions exhibited RMSEs of 11.5°, 36.8°, and 41.7°, with corresponding MBEs of −1.3°, 2.4°, and −8.8°, respectively. The results indicate that HiSea-1 and Chaohu-1 SAR satellites have the potential and practicality for SSWF retrieval, validating the technical indicators and performance requirements implemented during the satellites’ design phase.

Synthetic-Aperture Radar (SAR) has emerged as an important tool for monitoring tropical cyclones (TCs) due to its high spatial resolution and cloud-penetrating capability. Recent advancements in SAR technology have led to smaller and lighter satellites, yet few studies have evaluated their effectiveness in TC monitoring. This paper employs an algorithm for automatic TC center location, involving three stages: coarse estimation from a whole SAR image; precise estimation from a sub-SAR image; and final identification of the center using the lowest Normalized Radar Cross-Section (NRCS) value within a smaller sub-SAR image. Using three wide-swath miniaturized SAR images of TC Noru (2022), and TCs Doksuri and Koinu (2023), the algorithm’s accuracy was validated by comparing estimated TC center positions with visually located data. For TC Noru, the distances for the three stages were 21.42 km, 14.39 km, and 8.19 km; for TC Doksuri—14.36 km, 20.48 km, and 17.10 km; and for TC Koinu—47.82 km, 31.59 km, and 5.42 km. The results demonstrate the potential of miniaturized SAR in TC monitoring.