Objectives The spaceborne global navigation satellite system-reflectometry (GNSS-R) technology has been extensively employed in monitoring surface hydrology in recent years. However, research on the key factor of precipitation in the surface hydrological cycle, especially its spatiotemporal relationship with cyclone global navigation satellite system (CYGNSS) signals, remains limited. Therefore, this paper conducts a detailed and comprehensive analysis of the direct relationship between CYGNSS reflectivity and precipitation, taking into account multiple influencing factors and scale effects.

Methods This study utilizes monthly CYGNSS data and precipitation data from 2019 to 2020, taking Guangdong Province, China as the study area. It analyzes the spatiotemporal response relationship between CYGNSS signals and precipitation, while comprehensively considering the influence of different driving factors such as digital elevation model (DEM), slope, land cover type, soil type, and normalized difference vegetation index (NDVI) on the CYGNSS signal-precipitation response. The study also explores scale effects, and uses a factor detector model in the geographical detector to investigate the degree of influence of each driving factor.

Results From 2019 to 2020, the monthly average CYGNSS reflectivity and precipitation show significant seasonal differences and nearly synchronous variations, validating the potential of CYGNSS signals for precipitation monitoring. The positive correlation between CYGNSS signals and precipitation strengthens with increasing spatial scale. For example, at small scale (1 km), the positive correlation between CYGNSS signals and precipitation is 21.00% (significant at the 0.05 level), while at medium scale (3 km) and large scale (9 km), it reaches 26.84% and 40.36% respectively, highlighting the advantage of CYGNSS in monitoring precipitation at larger scales. Additionally, the study found that at different spatial scales, regions with lower DEM and slope (indicating relatively flat terrain), areas with land cover type of cropland, and regions with lower to moderate vegetation coverage exhibited a stronger positive correlation between CYGNSS signals and precipitation. However, the influence of soil type on the relationship between CYGNSS signals and precipitation response varied across different scales. The driving factors' influence on the relationship between CYGNSS signals and precipitation is ranked as follows: DEM>NDVI>slope>land cover type>soil type, and their influence becomes stronger with increasing scale.

Conclusions This study elaborates on the response relationship between CYGNSS signals and regional precipitation, providing useful references for further utilizing GNSS-R technology to monitor precipitation and facilitating accurate assessment of surface hydrological changes. The results demonstrate that certain spatiotemporal response relationships exist between CYGNSS signals and precipitation across different scales, and various driving factors have certain influences on their response relationships.