Session: B.4 - Progress in Hydrological Applications

THEME: Enhancing & Extending the GRACE Data Record, Saturday, 08:40 - 10:05

(Will GRACE results continue to be useful after the mission ends?)
John Wahr

(Exploring the link between Earth's gravity field, rotation and geometry in order to extend the GRACE-determined terrestrial water storage changes to non-GRACE times)
Hans-Peter Plag

(Pros and Cons of GPS for determining variability in continental water storage)
Tonie van Dam

(Calibration analysis of the global hydrological model WGHM with water mass variations from GRACE gravity data)
A. Guentner

(Using ancillary measurements to extend the GRACE-derived record of global freshwater discharge)
J. Famiglietti

Part - 2, Saturday 10:25 - 12:15

(Improvement of JLG terrestrial water storage model using GRACE satellite gravity data)
Keiko Yamamoto

(HYDROGRAV : First results: Southern Africa temporal gravity field changes from custom designed GRACE Mascons and a hydrological model)
Pernille E. Krogh

(What is GRACE Telling us About the Hydrology of the Nubian Aquifer?)
Mohamed Sultan

(Dynamics of surface water in Amazon inferred from measurements of inter-satellite distance change)
Shin-Chan Han

(Application of GRACE Water Storage for Water Resources Management: Case Study, High Plains Aquifer, US)
Bridget Scanlon

(Understanding extreme climate events using GRACE and climate models)
Jianli Chen

(Evaluating the Temporal Variations of Terrestrial Water Storage Components Using GRACE Data and Land Surface Modeling in Global River Basins)
Hyungjun KIM

(Temporal and spatial multiscale assessment of mass transport by combination of gravity observations from GRACE and terrestrial stations)
Corinna Kroner

(Developing the Global Geodetic Observing System into a Monitoring System for the Global Water Cycle (IGCP 565 Project))
Hans-Peter Plag

Abstract: The moderated portion of the Hydrology Session focuses on this question: Can we combine GRACE-based hydrology with other data in order to extend the time series and enhance its value in addressing climatological questions? I will provide a brief introduction to the topic, describing the issues and some ideas for how they might be addressed.

Back to top

Abstract: GRACE is presently expected to return data through 2012. If there is a GRACE follow-on mission, it is unlikely to be launched until well after that date. This talk will discuss ways in which GRACE results might continue to be useful even after GRACE is gone. The possibility of extrapolating between GRACE and a GRACE follow-on mission will be addressed.

Back to top

Abstract: Geodetic observations of Earth's time-variable gravity field, rotation and shape are intrinsically interlinked by that fact that the time variations of these quantities are caused by processes in one and the same unique Earth system. Although the signals induced by the Earth system processes in each of these three main fields of geodesy appear with different spatial characteristics and against a different noise spectrum, in principle, all three areas should allow the extraction of information on the Earth system processes. On time scales of weeks to decades, main signals in all three areas originate in mass relocation in atmosphere, ocean, cryosphere, and terrestrial hydrosphere, i.e. the 'fluid' envelop of the solid Earth. At these time scales, interaction of the solid Earth with its fluid envelop is to a high degree linear (the main non-linearities are in atmosphere and ocean and not impacting the interaction with the solid Earth), and signals in one area should be proportional to similar signals in the two other areas (although the proportionality may depend on spatial and temporal scales). Therefore, the recent period of highly accurate simultaneous observations of all three areas provides the unique basis for calibration of forward models and inversion methods which then could be used to extrapolate the models and/or inversions to the period before GRACE, when only accurate observations of time-variable shape and rotation were available. Likewise, the calibrated models and inversions could be used to extrapolate to post-GRACE times in case an immediate follow-on mission would not be available. The simultaneous period can also be use to establish realistic error estimates for models and inversions based on observations of time-variable rotation and shape only.

Considering the complex interactions of surfical mass relocation with the three geodetic areas and the requirement of mass conservation in the water cycle, we expect that only sufficiently sophisticated and comprehensive Earth system models will be able to make full use of the geodetic observations as constraints for the modeled mass relocation in the global water cycle. Such models, once calibrated with the full observational database of the last six years, will also provide the basis to assess the degree to which inversions of the geodetic observations can provide realistic estimates of mass relocation in the water cycle. In our presentation we will consider theoretical aspects of the model development and outline a road map towards a system model with assimilation of the geodetic observations.

Back to top

Abstract: In this talk, I will review recent research results which validate the use of GPS station positions for determining the long spatial scale continental water storage variability. I will also review research results which indicate that for shorter spatial scales that the inversion of GPS data for water storage information may be unreliable.

Back to top

Abstract: Time series of variable gravity observed by the GRACE (Gravity Recovery and Climate Experiment) satellite mission can be transformed to temporal variations of continental water storage. Therefore, these data sets inspire hydrological modelers to use them as an input for validation and calibration of large scale hydrological models. In this study, we continue our work to incorporate GRACE data for the first time directly into the parameter tuning process of a global hydrological model to improve simulations of the continental water cycle. For the WaterGAP Global Hydrology Model (WGHM), we adopt an efficient multi-objective calibration framework to constrain model predictions by both measured river discharge and water storage variations from GRACE. The calibration was applied to the 22 largest river basins worldwide and is illustrated by the examples of several river basins. The approach leads to improved simulation results with regard to both objectives. In case of monthly total water storage variations we obtained a RMSE reduction of about 25 mm for the Amazon and 6 mm for the Mississippi, for instance.

In addition, the calibration results are analyzed for a physical interpretation of the calibration results. We intend to improve the understanding of hydrological processes and the knowledge on continental water transfers on a seasonal time- and a large regional-scale. Therefore, we analyze river basins with various process characteristics and their calibration performance on single processes related to water storages of groundwater, canopy, snow, soil and surface water. Furthermore, a comparison of calibration results for different GRACE-data extractions is presented.

The results highlight the valuable nature of GRACE data when merged into large-scale hydrological modeling and depict methods to improve large-scale hydrological models.

Back to top

Abstract: Recent work by Syed et al. has shown how GRACE data can be used in a coupled land-atmosphere water balance to solve for freshwater discharge from the large river basin to continental scales. One important constraint on the method is that it is limited to the time period of the GRACE mission. In this presentation we show how we extend the discharge record back to 1994 and the beginning of the TOPEX/Jason era. We first compare GRACE estimates of ocean mass change to altimetry estimates of global mean sea level rise (GMSLR) minus its steric contribution. The steric contribution is derived from ARGO floats for the recent past, and from the longer-term Ishii and Ingleby-Huddleston global ocean temperature datasets. The good agreement during the three-year overlap period (2003-2006) gives us confidence to use the Ishii and Ingleby-Huddleston data to estimate ocean mass change for nearly the entire TOPEX/Jason time period. Given a time series of ocean mass change, we solve the ocean mass balance for a single time series of freshwater discharge into the global ocean, using ancillary precipitation and evaporation datasets constructed as part o the NASA Energy and Water Cycle Studies (NEWS) effort. Results for the entire (1994-2006) study period demonstrate a significant trend in global discharge (4.6 km3/month) that is largely attributed to increasing global ocean evaporation (5.1 km3/month).

Back to top

Abstract: For the future improvement of JRA-JCDAS LDA and GRiveT Terrestrial Water Storage (JLG) model, the phases and amplitudes of the annual components of mass variations of GRACE (Gravity Recovery and Climate Experiment) and JLG model are compared for 70 major river basins in the world. The phases of the annual components of GRACE and JLG model show good correspondence in most of the river basins, but about 1 to 2 month discrepancies were shown in Lena, Changjiang, Mackenzie, Orinoco, Yukon and Kolyma basins. Because GRACE data represent actual mass variations of terrestrial water storage including ground-water, these discrepancies mean that the current version of JLG model does not well represent the annual components in these basins. Thus the modelÕs phases can be improved using the GRACE result as constraints. In some basins with large signals, the amplitudes of the annual components of the GRACE mass variations show about 2 to 4 times large values compared with the modelÕs ones. The discrepancies can be explained by underestimation of GRACE errors, underestimation of the JLG model amplitudes and/or overestimation of the GRACE amplitudes, although we cannot conclude which is the main reason, at this stage.

Back to top

Abstract: Newly developed space-borne and ground-based methods for observing time-lapse gravity have recently provided a new kind of data for water balance monitoring and hydrological model calibration. The HYDROGRAV project aim at applying time-lapse gravity surveys to hydrological model-calibration and the monitoring of terrestrial water storage, using remote sensing data from the GRACE satellites, ground-based gravity surveys and satellite altimetry.

Available 4x4 degree Mass Concentration blocks (MASCONS) yield non-smoothed gravity solutions with a 10 day temporal resolution. However, comparison with GLDAS water storage estimates in the Zambezi basin shows large phase offsets for i.e., the annual signal.

To improve the spatial resolution, a system of custom designed MASCONS, with a resolution of 1.5 by 1.25/1.5 degrees, has been designed over Southern Africa, covering an area of 9 mill km2 and including the four largest river basins of the area (Zambezi, Okavango, Limpopo and Orange). These mascons have been spatially constrained in hydrological regions corresponding to local drainage basins.

Preliminary evaluation of these new MASCON data has been carried out by comparing MASCON gravity changes to 3D computed gravity changes from the output of a hydrological model build over the Zambezi river basin using ArcSWAT.

Back to top

Abstract: The Nubian Aquifer is one of the largest (~50,000 km3) freshwater aquifer systems in NE Africa; it lies beneath the surface of the hyperarid Western Desert of Egypt and the adjacent portions of eastern Libya, northeastern Chad, and northwestern Sudan covering an area of about 2,000,000 km2.

We analyzed equivalent water thickness grids (0.5x 0.5 degrees) of GRACE data for 63 months for the period (4/2002 to 10/2007). Processing was accomplished as follows: (1) the temporal mean was removed from each grid point measurement, (2) Sean Swenson's destriping method was applied to reduce the noise and enhance the signal to noise ratio, (3) results were smoothed using a Gaussian smoothing function (radius: 250-km), (4) monthly water storage predictions from Goddard's GLDAS/Noah model were removed, (5) the Red Sea signal was fitted and removed. Because, the water storage model was forced with precipitation only, anthropogenic effects if any were not accounted for. Surface water (with the exception of snow) was not accounted for.

The spatial distribution of GRACE data was compared to precipitation data extracted from 3-hourly TRMM data, digital elevation data (extracted from SRTM data: resolution: 1 km), stream networks (extracted from SRTM), and basement uplifts (extracted from regional geologic maps). Comparisons made in a web-based GIS that was generated for this purpose shows the following: (a) large positive anomalies (standard deviation > 6 cm) observed on standard deviation images generated over periods of one, two, three, four, five, and six years, were persistent over the same areas and increased with the increase in sample population (number of observations), (b) anomalous areas correlated spatially with the slopes and foothills of mountains over which the largest cumulative (4/2002 to 10/2007) precipitation (>3000 mm) was observed. These observations raise the intriguing possibility that we are examining elements of recharge, and/or surface runoff and/or groundwater flow. To further test these preliminary findings, we are expanding the examined area to include all of Africa and the Arabian Peninsula as well. Preliminary results indicate similar findings over the examined expanded area.

Back to top

Abstract: Terrestrial water storage in the Amazon basin and its surrounding areas is studied by exploiting the instantaneous measurements of distance changes between two satellites of Gravity Recovery and Climate Experiment (GRACE) mission. We found that the surface water in the channels and floodplains can be significant in weighing total water storage. Its magnitude can be as large as soil moisture perturbing the motions of the GRACE satellites to a detectable amount by the on-board radar-ranging instrument. The simulations with variable routing velocities indicate that the most reasonable velocity throughout the basins and throughout the years is about 30 cm/s with significant seasonal change. The lower velocity during rising stages and peak water season, and the faster velocity during falling stages is delineated by the observations. The backwater effects mainly caused by the southern tributaries reconcile with such changes in the overall routing velocity. Direct assimilation of the radio tracking data will be followed to construct more realistic models for surface water dynamics.

Back to top

Abstract: The goal of this project is to apply GRACE time-variable gravity data to quantify water storage variations in major terrestrial aquifers. The High Plains Aquifer in the central United States is our initial focus. This Aquifer was specifically mentioned in the 1997 National Research Council report ÒSatellite Gravity and the GeosphereÓ as one suitable for application of space-borne gravity measurements due to its large size. Standard processing of GRACE data resulted in large seasonal terrestrial water storage changes that were highly correlated with measured groundwater and soil moisture storage changes (R>0.8). Although the area of the High Plains aquifer is sufficiently large (450,000-km2) for GRACE analysis, reliable estimates of water storage variations require a balance among noise reduction, maximum spatial resolution, and minimum spatial leakage. Non-uniform aquifer discharge, primarily from irrigation and recharge from precipitation, make the spatial resolution problem even more challenging. Furthermore, a greater temporal resolution (~ 1 Ð 10 d) is of interest to assimilate GRACE data into a numerical water storage model of the High Plains. We will evaluate different approaches to data processing to minimize noise, increase temporal resolution by using available monthly and 10 day solutions, and examine different approaches of increasing spatial resolution. The different processing approaches will be applied to the High Plains where detailed monitoring data are available and then transferred to other aquifers where less data are available, such as the Nubian Aquifer and the North China Plain.

Back to top

Abstract: Satellite gravity measurements from the Gravity Recovery and Climate Experiment (GRACE) provide accurate quantitative measures of terrestrial water storage (TWS) change associated with extreme climate events, such as the 2005 extreme drought event in the Amazon river basin, regarded as the worst in over a century. Some locations in the central Amazon basin recorded lowest river water levels on record during that period. We use GRACE gravity data and climate models estimates to identify and quantify abnormal TWS changes associated with this major climate event. GRACE measurements show a significant TWS decrease in the central Amazon basin in the summer of 2005, relative to the average of the 5 other summer periods in the GRACE era (2002 Å 2007) analyzed in this study. In contrast, data assimilating climate and land surface models significantly underestimate the drought intensity. GRACE measurements are consistent with accumulated precipitation data from satellite remote sensing, and are also supported by in situ water level data from rive gauge stations. This study demonstrates the unique potential of satellite gravity measurements in monitoring large-scale severe drought and flooding events and in evaluating advanced climate and land surface models.

Back to top

Abstract: Terrestrial Water Storage Anomaly (TWSA) derived from the Gravity Recovery and Climate Experiment (GRACE) satellite is vertically evaluated over 48 selected global river basins with drainage areas larger than 200,000 km2. The components of terrestrial water storages including soil moisture, snow water equivalent, and river storage were simulated by the land surface model MATSIRO (Minimal Advanced Treatment of Surface Interaction and Runoff) and runoff routing scheme TRIP (Total Runoff Integrated Pathway) from 2002 to 2007. Atmospheric forcing was based on Japanese Meteorological Agency (JMA) Climate Data Assimilation System (JCDAS), and precipitation was corrected using various ground and satellite observations. The consistency in the time span between GRACE data and model integration is a unique advantage over previous similar studies in literature. It is found the annual cycle of each individual TWS component exhibit different amplitudes and phases for the basins in different climate regions. Overall the model-simulated total TWSA shows remarkable agreement in the phase of seasonal variations with the GRACE over most of the basins, but their amplitudes are not consistent with GRACE in some basins. Furthermore, the role of river channel storage is highlighted in this study. River channel storage explains up to 70% of the TWS variations depending on the basin sizes, and it is the dominant component of TWS variation over the largest river basins like the Amazon.

Back to top

Abstract: The network of superconducting gravimeters (SG) within the 'Global Geodynamics Project' (GGP) offers the unique opportunity to supplement and validate the gravity field variations derived from the GRACE satellite mission. Because of the different spatial and temporal resolution of the observations a combination of the data sets can be considered as the most straight forward approach to retrieve a maximum of information regarding mass transfers on different time and spatial scales esp. related to hydrology. For a consistent combination of the time series the gap in the resolutions has to be bridged.

In principle, the same reductions applied to the GRACE data have to be taken into account for the terrestrial observations. The separation of local, regional, and global hydrological effects in SG observations is crucial for the comparison and combination with satellite-derived gravity changes. It can be shown that even for stations with a hydrologically challenging situation such as Moxa/Germany local hydrologically induced effects on gravity can be successfully removed. The separation of the different spatial effects allows an examination of individual hydrological contributions as well as the influence of topography.

Initially, the study focuses on mid-Europe where a dense and long-term observation network exists, covering the period of the GRACE mission. From a first comparison of GRACE residuals, GRACE-derived gravity field changes, and gravity variations computed on the basis of hydrological models for several European stations a principle good agreement is found. From detailed analyses indications for the origin of deviations can be derived.

Back to top

Abstract: Geodetic observations of the Earth's time-variable gravity field, shape, and rotation capture the signals of variation in the entire fluid envelope of the solid Earth, including the terrestrial water storage. Therefore, the Global Geodetic Observing System (GGOS) has the capability to monitor mass transport particularly in the global water cycle. The IGCP 565 Project (http://geodesy.unr.edu/igcp565/) aims to utilize this potential and to develop GGOS into a monitoring system for the hydrological cycle on global to regional scales. Key scientific issues addressed are: (1) Development of an integrated dynamic model for the predictions of the geodetic signals induced by surface mass changes on daily to interannual scales; (2) Inversion algorithms for integrated geodetic observations for surface mass changes; (3) Assimilation of observed surface mass changes in hydrological models; and (4) Development of products relevant for regional water management. The research and development is carried out in a number of on-going and planned projects under the umbrella of the IGCP project. The project supports capacity building in space-geodetic data processing, modeling of the hydrological cycle, and interpretation of the observations in terms of terrestrial water storage. A focus is on products for regional water management, particularly in developing countries. Coordination of the research and capacity building is provided through a series of five annual workshops.

The first project workshop will be held on December 11, 2008 prior to the GRACE Science Team meeting. The focus of this workshop is on a review of the "Science of geodetic monitoring of the hydrological cycle" (http://geodesy.unr.edu/igcp565/workshops/). The presentation will briefly introduce the IGCP 565 project and report the main conclusion of the first Project Workshop.

Back to top