In this paper, we investigate the feasibility of using coherent and semi-coherent GNSS-R signals for land application such as land surface elevation and inland water body (IWB) surface gradient observations. Raw GPS L1 intermediate frequency (IF) datasets recorded by CYGNSS satellites in the low-Earth orbits are processed using an adaptive hybrid tracking (AHT) algorithm presented in . The AHT algorithm can effectively mitigate the cycle slips in the carrier phase estimations caused by signal amplitude fading while simultaneously reduce noise for both the pseudorange and carrier phase measurements through an adaptive filtering scheme implemented in the algorithm. Two examples from processing the CYGNSS raw IF data are presented in the paper. One example is a semi-coherent land-reflection event over Vietnam which demonstrates pseudorangebased land altimetry at meter-level accuracy as compared with the shuttle radar topography mission (SRTM) digital elevation map (DEM). The second example is an coherent reflection pass over the Orinoco River. This example demonstrates the potential of GNSS-R for IWB surface height retrieval at centimeter-level precision and surface gradient at cm/km level.