[1] We studied 34 global reservoirs for which good quality surface elevation data could be obtained from a combination of five satellite altimeters for the period from 1992 to 2010. For each of these reservoirs, we used an unsupervised classification approach using the Moderate Resolution Imaging Spectroradiometer (MODIS) 16-day 250 m vegetation product to estimate the surface water areas over the MODIS period of record (2000 to 2010). We then derived elevation-area relationships for each of the reservoirs by combining the MODIS-based estimates with satellite altimeter-based estimates of reservoir water elevations. Through a combination of direct observations of elevation and surface area along with documented reservoir configurations at capacity, we estimated storage time histories for each reservoir from 1992 to 2010. We evaluated these satellite-based data products in comparison with gauge observations for the five largest reservoirs in the United States (Lakes Mead, Powell, Sakakawea, Oahe, and Fort Peck Reservoir). The storage estimates were highly correlated with observations (R = 0.92 to 0.99), with values for the normalized root mean square error (NRMSE) ranging from 3% to 15%. The storage mean absolute error (expressed as a percentage of reservoir capacity) for the reservoirs in this study was 4%. The multidecadal reconstructed reservoir storage variations are in accordance with known droughts and high flow periods on each of the five continents represented in the data set.Citation: Gao, H., C. Birkett, and D. P. Lettenmaier (2012), Global monitoring of large reservoir storage from satellite remote sensing, Water Resour. Res., 48, W09504,

Remote sensing and long‐term monitoring of closed and climatically sensitive open lakes can provide useful information for the study of climatic change. Satellite radar altimetry offers the advantages of day/night and all‐weather capability in the production of relative lake level changes on a global scale. A simple technique which derives relative lake level changes is described with specific relevance to the TOPEX/POSEIDON geophysical data record data set. An assessment of the coverage and global tracking performance of both the NASA radar altimeter and the solid state altimeter over these lakes is discussed, and results are presented for the first 1.75 years of the mission. Lake level time series were acquired for 12 closed lakes, six open lakes, and three reservoirs, providing information in many cases where ground gauge data are unobtainable or the lake is inaccessible. The results, accurate to ∼4 cm rms, mark the beginning of a very accurate lake level data set, showing that TOPEX/POSEIDON can successfully contribute to the long‐term global program.

Abstract. For certain major rivers and wetlands, hydrological information can often be difficult to obtain due to the inaccessibility of the region, the sparse distribution of gauge stations, or the slow dissemination of data. Satellite radar altimeters have the potential to monitor height variations over inland waters. Here, it is shown that the NASA radar altimeter (NRA), currently operating on board the TOPEX/POSEIDON satellite, can successfully track both large wetlands and rivers of > 1 km width. The coverage, performance, and limitations of the NRA altimeter are discussed with relevance to these regions, and the merits of utilizing the geophysical data records (GDRs) and the sensor data records (SDRs) are explored. Time series of relative water level variations, for the first -3.7 years of the mission, have been obtained for a selection of the world's largest rivers, and for major wetlands of international importance. Validation shows the results can be accurate to -11 cm rms, offering the potential to observe these regions as part of a long-term hydrological monitor!ng program. IntroductionMajor river systems are important topics of research covering a wide range of applications such as transport, flood hazards, water and food resource management, studies of the hydrological cycle, and addressing the impacts of land use and climate change [Leopoldo et al., 1985]. Wetlands (swamps, marshes, floodplains, internal deltas, etc.) cover 6% of the world's land surface. The extent of inundated area and depth of wetland water varies seasonally and annually according to inflow, precipitation, and evaporation rates. As their hydraulic response is relatively slow, they are good proxy indicators of local climate, but they are also important with regard to local environmental and economic concerns, being both a source of food and fresh water [Maltby, 1986].Although most major North American and European rivers are frequently (daily) and accurately (better than 1 cm at the point of measurement) monitored (with water level variations being converted to discharge rates and the data subsequently archived), water levels for many other regions can be difficult to obtain due to the inaccessibility of the region, the sparse distribution of gauge stations, or the slow dissemination of data. Although primarily aimed at ocean applications, satellite radar altimetry has demonstrated a potential to provide both lake and river/wetland height data. A radar altimeter emits a series of microwave pulses toward the surface, and by noting the two-way time delay between pulse emission and echo reception, estimates the altimetric range, that is, the distance between the antenna and the surface (see Rapley [1990] for full details on radar altimetry). In To summarize, some degree of success has been achieved in deriving altimetric water levels over rivers and wetlands. The limitations have been poor knowledge of the satellite orbit and instrumental inefficiencies in data gathering (variable according to each altimeter). As the radar echoes fr...

[1] Satellite radar altimetry has the ability to monitor variations in surface water height (stage) for large wetlands, rivers, and associated floodplains. A clear advantage is the provision of data where traditional gauges are absent. As part of an international program, a complete altimetric analysis of the Amazon Basin is being undertaken. Here, an updated and more rigorous evaluation of the TOPEX/POSEIDON (T/P) data set is presented for the first $7.5 years of the mission. With an initial study group of 230 targets, height variability at many ungauged locations can be observed for 30-50%, the range reflecting the clarity of the variations in lieu of instrument limitations. An assessment of the instrument performance confirms that the minimum river width attainable is $1 km in the presence of some inundated floodplain. This constraint does allow observation of the main stem (Solimões/Amazon) and the larger tributaries, but rugged terrain in the vicinity of the target additionally places severe limitations on data retrieval. First-order validation exercises with the deduced 1992-1999 time series of stage fluctuations reveal accuracies ranging from tens of centimeters to several meters (mean $1.1 m rms). Altimetric water levels in the Solimões and Amazon are particularly well defined with amplitudes <13 m and variations in peak-level timing from May to July. The water-surface gradient of the main stem is found to vary both spatially and temporally, with values ranging from 1.5 cm/km downstream to 4.0 cm/km for more upstream reaches. In agreement with ground-based estimates, the seasonal variability of the gradients reveals that the hysteresis characteristic of the flood wave varies along the main stem and the derived altimetric velocity of this flood wave is estimated to be $0.35 m/s. Overall, the altimetric results demonstrate that the T/P mission is successfully monitoring the transient flood waves of this continental-scale river basin.