H31I-01 INVITED
Continental and Global-Scale Terrestrial Water and Energy Budgets Using Remote Sensing Observations
Documenting the global water and energy cycle through modeling and observations is fundamental to achieving the goals of the World Climate Research Programme's Global Energy and Water Cycle Experiment (GEWEX), and similar national programs like NASA's Energy and Water Cycle Study (NEWS) that support GEWEX research. This documentation is needed to acquire enhanced knowledge of Earth's climate, including characterizing the memories, pathways and feedbacks between key water, energy and biogeochemical cycles through observations, and so address the outstanding NEWS and GEWEX goal "to document and enable improved, observationally-based, predictions of (the) energy and water cycles..". To date, some components of the terrestrial water and energy budgets have been estimated. For example, there are a number of precipitation estimates based on polar orbiting passive microwave sensors from both experimental (e.g. AMSR-E on board NASA's Aqua Earth Observing platform) and operational platforms (e.g. the polar orbiting SSM/I sensors). Surface hydrologic soil moisture states are available using TRMM Microwave Imager or AMSR-E X-band (~10.7 GHz) measurements. Furthermore, variations in land water storage can be derived from GRACE gravimetric measurements that aggregate changes in snowpack, soil moisture and ground water. More recently, large scale evapotranspiration products are becoming available that, for example, use NASA EOS radiation, surface temperature and land characteristics within energy balance algorithms or Penman type equations. In this presentation, the terrestrial water and energy budgets are computed over the Mississippi River basin for recent years using remote sensing measurements for all components except for river discharge and are compared to land surface model based terrestrial budgets, and North American Regional Reanalysis (NARR) model outputs. We also use atmospheric water budgets from remote sensing and NARR as a further constraint on the terrestrial budgets. Analyses of the seasonal cycle and non-closure of the remote sensing based budgets provide indications of the consistency of the budget estimates. Based on the Mississippi basin study, the strategy for a global budget analysis will be discussed, and examples of initial global remote sensing products will be provided.
H31I-02 INVITED
Advances and challenges in large-basin climate and hydrology studies with the integration of models and data assimilation
In recent years, researchers have begun to focus on scientific studies of the impact of environmental changes on various hydrologic processes in large river basins. These studies have relied on the development and application of a model, coupling both climate and hydrologic systems. This presentation will address advances in designing interactive climate-hydrologic systems to examine explicit responses of rivers, lakes, wetlands and groundwater tables across various scales. Remotely sensed data and GIS data sets were integrated with observed and analyzed measured data to form a base for studying the response of hydrologic systems to environmental changes at different scales. Numerical experiments with different setups were conducted to evaluate the spatial distribution of hydrologic components in the basin. The results would shed the light on how best the variation in the small-scale hydrologic processes can be represented in the large-scale climate and hydrologic responses. This continuous direction of ongoing research will be at the nexus of related fields of geoinformatics, cyberinfastructure in watersheds and remote measurements.
H31I-03
What Will SWOT Measurements and Products Look Like?
The Surface Water Ocean Topography (SWOT) satellite mission is designed to measure water elevations
across the world's rivers, lakes, wetlands, and oceans. The mission is designed to provide global, high-
spatial resolution, ~weekly products of river discharge and changes in stored water volumes. Already, some
satellite missions such as GRACE, various altimeters, SRTM, and imaging systems, provide a glimpse of what
these products might look like. For example, despite GRACE's hundreds of kilometers of spatial resolution, it
allows a first-order assessment of Amazon floodplain storage anomalies – measurements that are key for
understanding the hydraulic flux and subsequent delivery of sediment and exchange of nutrients. Various
altimeters have shown that spot elevations of water surfaces are possible, albeit without a fine resolution on
the order of less than 100m. In contrast, repeat-pass interferometric SAR measurements show a high-spatial
resolution (~30m) but poor temporal resolution (~monthly) and are restricted to dh/dt mappings (i.e., water
elevation changes with time). SRTM is perhaps the most spatially detailed mapping of water surface
elevations globally, but is restricted to one temporal map (i.e., February 2000). SWOT is designed to
overcome these various limitations by using a Ka-band radar interferometer which is essentially a higher-
accuracy and more compact version of SRTM. Its look angle is near-nadir which maximizes the returned
radar energy and improves the height accuracy to +/- 0.5m (an order of magnitude better than SRTM).
Because SWOT uses this two-antennae SAR system, the smallest spatial resolution theoretically possible is
2m (in azimuth) by 0.75m (in range) whereas, after multi-looking, SWOT will have a nominal resolution of
around 30m x 30m. The mission will produce terabytes of data per month, thus ground systems need to be
designed to rapidly convert water elevation measurements to products of discharge and storage change.
SWOT is in the design phase and is being considered for launch around 2015.
http://bprc.osu.edu/water
H31I-04 INVITED
Integration of Satellite Data Toward a Refined Depiction of the Global Water Cycle
Numerous streams of satellite observations are now providing the raw materials to enhance our understanding of the global water cycle. At the same time, powerful yet economic computers and scientific innovations are enabling land surface models to simulate physical processes more realistically and at higher resolutions than ever before. Such models provide a platform for synthesizing the data streams, which is a necessary step toward the development of a physically consistent and realistic representation of terrestrial hydrospheric processes. This is the premise behind the Global Land Data Assimilation System (GLDAS) project, which is supported by NASA's Energy- and Water Cycle Sponsored Research (NEWS) program and which uses and contributes to the development of Land Information System (LIS) software. Here we describe some recent LIS/GLDAS modeling and data assimilation innovations, ongoing water cycle research, and new applications.
H31I-05
Estimating continental-scale water balances through remote sensing
The magnitude of the key terms in the terrestrial water balance at global and continental scales is surprisingly poorly known. Neither in situ observations nor land surface modeling is of sufficient accuracy to close the water balance through independent estimates of the terms. Remote sensing provides a basis for near-independent estimation of some of the key terms, and while strategies that combine in situ and remote sensing and modeling via data assimilation, an important baseline is a stand-alone approach to estimating the water cycle through remote sensing alone. Therefore, we address the science question – "How well can terrestrial water balance at continental scales be estimated from remote sensing observations alone?" In this study, all the major terrestrial water balance components (precipitation, ET, change of water storage, etc.) over the continental US are estimated from remote sensing observations from 2003 to 2007. The remotely sensed precipitation is from the TRMM 3B42 Version 6 product, which is further partitioned into rainfall and snow using surface air temperature as a threshold. A very simple snow scheme is used that ignores snowpack heat content and assumes that when the net radiation is positive, some of the accumulated snow will melt and become snow runoff. A remote sensing-based evapotranspiration (ET) product from Princeton University based on MODIS measurements is used. The method is based on 1:30 am and 1:30 pm (local time) overpasses, which are expanded to daily total ET by assuming a constant evaporative fraction. Precipitation, snow runoff, and ET are all calculated hourly, and are then upscaled to monthly values. These water cycle variables, together with total terrestrial water storage from the Gravity Recovery and Climate Experiment (GRACE, also monthly), are used to calculate the runoff for major basins across the continental U.S. To evaluate the accuracy of the methodology, the computed basin runoff is compared with observations at the outlets of major rivers.
H31I-06
HYDROGRAV - Hydrological model calibration and terrestrial water storage monitoring from GRACE gravimetry and satellite altimetry - First results
Space-borne and ground-based time-lapse gravity observations provide new data for water balance
monitoring and hydrological model calibration in the future. The HYDROGRAV project (www.hydrograv.dk) will
explore the utility of time-lapse gravity surveys for hydrological model calibration and terrestrial water
storage monitoring.
Merging remote sensing data from GRACE with other remote sensing data like satellite altimetry and also
ground based observations are important to hydrological model calibration and water balance monitoring of
large regions and can serve as either supplement or as vital information in un-gauged regions.
A system of GRACE custom designed Mass Concentration blocks (Mascons) have been designed to model
time-variable gravity changes for the largest basins in Southern Africa (Zambezi, Okavango, Limpopo and
Orange) covering an area of 9 mill km2 with a resolution of 1 by 1.25 degree. Satellite altimetry have been
used to derive high resolution point-wise river height in some of the un-gauged rivers in the region by using
dedicated retracking to recovers nearly un-interrupted time series over these rivers. First result from the
HYDROGRAV project analyzing GRACE derived mass change from 2002 to 2008 along with in-situ gravity
time-lapse observations and radar altimetry monitoring of surface water for the southern Africa river basins
will be presented.
http://www.hydrograv.dk
H31I-07
Science Requirements for Hydrologic Storage Change from the SWOT Mission
The Surface Water Ocean Topography (SWOT) satellite mission has been selected by the NRC Decadal Survey for launch between 2014 and 2016. NASA and CNES have jointly endorsed SWOT and provided encouragement via the formation of a Science Working Group. It is critical to study the Level 1 SWOT hydrologic science requirements, which drive the mission design. Hydrologic requirements include estimates of surface water discharge and changes in storage. This paper focuses on describing current results of simulation studies aimed at quantifying specific SWOT science requirements on global hydrologic storage changes. The objectives of the study include the optimal spatial and temporal sampling of storage changes and the related height accuracies and radar pixel sizes for the SWOT instrument. (1) Storage changes in the Amazon and Siberian Arctic have been estimated from existing satellite measurements, in-situ data, and model outputs. (2) The changes in water surface elevations and areas are calculated by dividing the storage changes from Task 1 by classifications indicating water body locations (i.e., water masks). (3) The desired level of storage change accuracy needed for hydrologic science descriptions of the Level 1 requirements have been incorporated. (4) Orbital tracks with differing spatial and temporal samplings have been studied using the various storage change maps (generated from Task 1) to determine percentages of the total that are or are not measured. We report the results of weekly, monthly, and seasonal variations in water surface elevations and areas (from Task 2) to determine the required SWOT instrument accuracies.
H31I-08
How accurately will SWOT measurements be able to characterize river discharge?
The Surface Water and Ocean Topography (SWOT) mission is a swath mapping radar altimeter that would provide new measurements of inland water surface elevation (WSE) for rivers, lakes, wetlands and reservoirs. SWOT has been recommended by the National Research Council Decadal Survey to measure ocean topography as well as WSE over land; the proposed launch date timeframe is between 2013 – 2016. SWOT WSE estimates would provide a source of information for characterizing streamflow globally. In this paper, we evaluate the accuracy of river discharge estimates obtained from SWOT measurements over the Ohio River and eight of its major tributaries within the context of a virtual mission (VM). SWOT VM measurements are obtained by simulation from the hydrodynamic model LISFLOOD, using USGS streamflow gages as boundary conditions and validation data. SWOT measurements are then input into an algorithm to obtain estimates of discharge variations. The algorithm is based on Manning's equation, in which river width and slope are obtained from SWOT, roughness is estimated a priori. Three different algorithms are used to estimate depth. SWOT discharge estimates are compared to the discharge simulated by LISFLOOD. In this way, we are able to characterize the accuracy of SWOT estimates of instantaneous discharge. More specifically, we characterize how SWOT accuracy varies as a function of the river characteristics and contributing area, such as Strahler order. More accurate depth and discharge estimates can be obtained by data assimilation, but will be more computationally expensive.