Abstract: Water level monitoring is an important part of the "third line of defense" in rain-water monitoring and forecasting, and accurate and real-time water level monitoring is very important in flood disasters. The most prevalent method for measuring water level is the use of water level gauges, which can be costly and have limited spatial resolution. In recent years, Global Navigation Satellite System-Reflectometry (GNSS-R) technology has emerged as a promising approach for water level monitoring due to its low cost and high spatiotemporal resolution. This paper introduces a novel real-time water level monitoring method based on GNSS dual-antenna attitude measurement and develop a model to invert water level based on baseline vector. This method uses double-difference observations to eliminate errors caused by various factors, such as satellite and receiver clock bias, ionospheric and tropospheric delay. In order to verify the stability and reliability of this method, field tests were carried out at Dongshahe Station in Beijing. Experimental results from June 5 to July 10, 2023 show that the method can achieve 1 Hz water level monitoring with a root mean square error (RMSE) of 2.88 cm compared to on-site water gauge data.