Retrieving High Precision River Stages and Slopes from Space
Ernesto Rodriguez(1) and delwyn moller(2)
Jet Propulsion Laboratory,
4800 Oak Grove Drive,
Pasadena, CA 91109,
(2) jet propulsion laboratory, 4800 Oak Grove Drive, Pasadena, 91109, United States
Conventional radar altimetry has been successful in retrieving water level measurements at altimeter crossings with an accuracy of 10cm to 20cm. Although promising, this measurement accuracy is insufficient to provide global monitoring of fresh water bodies, as has been proposed by the WatER mission, for instance. In this paper we examine in detail the error sources that a near-nadir synthetic aperture radar interferometer, such as the KaRIN instrument proposed to meet the WatER requirements, will be subject to and demonstrate that with appropriate calibration techniques, measurements of river stage with an accuracy of approximately 5 cm and river slope with an accuracy of 1 cm/1km can be obtained. In the first part, we examine the main error contributors to the height measurements and quantify the expected magnitude of the errors. The errors sources examined include tropospheric effects, spacecraft orbit and attitude stability, the effect of vegetation, and the effect of topographic lay-over. These sources of error are examined analytically and also with the help of an instrument simulation which includes all error sources to generate simulated measurements. Simulated performance results will be presented for the Ohio river basin and for the Amazon basin at the Solimoes/Puros confluence. In the second part, we examine calibration techniques to mitigate the errors mentioned above and demonstrate the feasibility of achieving the height and slope performance given in the first paragraph. Simulated calibration results will be presented for both Ohio and Amazon basins. Finally, we propose a method for processing the interferometer data to optimally filter random measurement noise and provide high precision estimates of river stage and slope which can be assimilated simply into hydrologic models or used in conjunction with ancillary data or physical assumptions to provide estimates of river discharge.