Improved river level retrieval in South Asia by Saral/Altika RaDAR altimetry

The Ganges-Brahmaputra-Meghna (GBM) basin in South Asia, which mainly covers Bangladesh and India territories, has periodic and flash floods every year due to Indian monsoon from Bay of Bengal and melting glacier/snow water from the Himalayas. These three main rivers with countless tributaries not only have seasonal water variation but also occasionally high-elevated waters. However, due to the remoteness and insufficient local infrastructure, river gauges is limited in this area and the timeliness of data availability is hardly achievable. Therefore, using satellite observation to monitor rivers and streams becomes a promising and necessary technique as the data processing algorithm has been developed and the data is more accessible. In the past two decades, altimetry satellites that were used to observe open ocean had been studied for investigating inland water bodies. The altimetry waveform needs to be carefully retracked in order to identify the water signal in a series of radar echoes. From a chain of well-known altimetry satellites, e.g., ERS-1/-2, TOPEX/Poseidon, Jason-1/-2, and Envisat, the sampling rate is mainly at 18-20 Hz after a smoothing from its nominal pulse repetition frequency. However, the sampling rate corresponding to about 350 m on the ground is sometimes insufficient for narrow rivers. Hence, in this study we utilized SARAL/AltiKa, an Indian altimetry satellite operated in Ka band radar, with a sampling rate at 40 Hz, to investigate its applicability in the GBM basin and compare with other independent satellites. Our preliminary result in Yaluzangbu River, the upstream of Brahmaputra, shows the uncertainty of measurements in each cycle has reduced from meters level to below 50 cm.