ABSTRACT Building on a recent suite of work that has demonstrated theoretical feasibility and ope... more ABSTRACT Building on a recent suite of work that has demonstrated theoretical feasibility and operational readiness of a satellite altimeter based flood forecasting system, we recently put a progressively designed altimeter based transboundary flood forecasting system to the ultimate test of real-time operational delivery in Bangladesh. The JASON-2 satellite altimeter, which was in orbit at the time of writing this manuscript, was used as the flagship altimeter mission. This paper summarizes the entire process of designing the system, customizing the workflow, and putting the system in place for complete ownership by the Bangladesh stakeholder agency for a 100 day operational skill test spanning the period of June 1 2013 through Sept. 9, 2013. Correlation for most of the flood warning stations ranged between 0.95 to 0.80 during the 1 day to 8 days lead time range. The RMSE of forecast typically ranged between 0.75m to 1.5m at locations where the danger level relative to the river bed was more than an order higher (i.e., >20m). The RMSE of forecast at the 8 days lead time did not exceed 2m for upstream and mid-stream rivers inside Bangladesh. The RMSE of forecast at the 8 days lead time exceeded 2m at a few estuarine river locations affected by tidal effects, where danger level relative to river bed was smaller (i.e., <;20m). Such a satellite altimeter system, such as one based on the JASON-2 altimeter, is now poised to serve the entire inhabitants of the Ganges-Brahmaputra-Meghna river basins as well as 30 or more flood-prone downstream nations currently deprived of real-time flow data from upstream nations.
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2010
The proposed Surface Water and Ocean Topography (SWOT) mission will provide global, space-based e... more The proposed Surface Water and Ocean Topography (SWOT) mission will provide global, space-based estimates of water elevation, its temporal change, and its spatial slope for terrestrial water bodies. Using derivations of water slope from the Shuttle Radar Topography Mission (SRTM) elevation data, river bathymetry and Manning's equation, the potential of SWOT for discharge estimation of large braided rivers in humid climates, such as the Brahmaputra river, was found to be promising (Jung , in Earth Surface Processes and Landforms, 2010). In this study we extend the work on assessing SWOT for braided rivers to understand the sensitivity of two river hydraulic parameters to discharge estimation: 1) section factor (AR2/3) derived from land-water classification and in situ river bathymetry and 2) Manning's roughness coefficient. For braided rivers, the first parameter, is intimately dependent on how braided rivers are classified of the multiple channels (water) and in-stream braided bars (land) that consequently dictates the accuracy of wetted perimeter and area of flow estimation from water elevation data. We show that the use of the minimum water elevation data at a river cross section minimizes estimation of section factor which consequently minimizes outlier discharge estimation reported in the Jung study. We also show that by treating roughness coefficient “flexible” as a calibration parameter, discharge estimation from SRTM elevation data can be further improved through trial and error manual optimization. Our sensitivity study illustrates the value of treating section factor and roughness coefficient as calibration parameters for data assimilation systems that use SWOT observables to estimate river discharge in braided rivers.
ABSTRACT Building on a recent suite of work that has demonstrated theoretical feasibility and ope... more ABSTRACT Building on a recent suite of work that has demonstrated theoretical feasibility and operational readiness of a satellite altimeter based flood forecasting system, we recently put a progressively designed altimeter based transboundary flood forecasting system to the ultimate test of real-time operational delivery in Bangladesh. The JASON-2 satellite altimeter, which was in orbit at the time of writing this manuscript, was used as the flagship altimeter mission. This paper summarizes the entire process of designing the system, customizing the workflow, and putting the system in place for complete ownership by the Bangladesh stakeholder agency for a 100 day operational skill test spanning the period of June 1 2013 through Sept. 9, 2013. Correlation for most of the flood warning stations ranged between 0.95 to 0.80 during the 1 day to 8 days lead time range. The RMSE of forecast typically ranged between 0.75m to 1.5m at locations where the danger level relative to the river bed was more than an order higher (i.e., >20m). The RMSE of forecast at the 8 days lead time did not exceed 2m for upstream and mid-stream rivers inside Bangladesh. The RMSE of forecast at the 8 days lead time exceeded 2m at a few estuarine river locations affected by tidal effects, where danger level relative to river bed was smaller (i.e., <;20m). Such a satellite altimeter system, such as one based on the JASON-2 altimeter, is now poised to serve the entire inhabitants of the Ganges-Brahmaputra-Meghna river basins as well as 30 or more flood-prone downstream nations currently deprived of real-time flow data from upstream nations.
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2010
The proposed Surface Water and Ocean Topography (SWOT) mission will provide global, space-based e... more The proposed Surface Water and Ocean Topography (SWOT) mission will provide global, space-based estimates of water elevation, its temporal change, and its spatial slope for terrestrial water bodies. Using derivations of water slope from the Shuttle Radar Topography Mission (SRTM) elevation data, river bathymetry and Manning's equation, the potential of SWOT for discharge estimation of large braided rivers in humid climates, such as the Brahmaputra river, was found to be promising (Jung , in Earth Surface Processes and Landforms, 2010). In this study we extend the work on assessing SWOT for braided rivers to understand the sensitivity of two river hydraulic parameters to discharge estimation: 1) section factor (AR2/3) derived from land-water classification and in situ river bathymetry and 2) Manning's roughness coefficient. For braided rivers, the first parameter, is intimately dependent on how braided rivers are classified of the multiple channels (water) and in-stream braided bars (land) that consequently dictates the accuracy of wetted perimeter and area of flow estimation from water elevation data. We show that the use of the minimum water elevation data at a river cross section minimizes estimation of section factor which consequently minimizes outlier discharge estimation reported in the Jung study. We also show that by treating roughness coefficient “flexible” as a calibration parameter, discharge estimation from SRTM elevation data can be further improved through trial and error manual optimization. Our sensitivity study illustrates the value of treating section factor and roughness coefficient as calibration parameters for data assimilation systems that use SWOT observables to estimate river discharge in braided rivers.
Uploads
Papers