G31B-0649
Variability in land water storage from GRACE and ENVISAT, and rainfall in South American river basins
Previous work has demonstrated the capability of GRACE to capture important aspects of the hydrological cycle, in particular seasonal and interannual fluctuations in land water storage of large river basins. Part of this behaviour can be immediately assigned to seasonal/interannual fluctuations of precipitation. In this study, we investigate existing correlations between GRACE water storage (two GRACE products are used and compared, the GRGS and GSFC/Mascons solutions), ENVISAT-based surface water levels and precipitation data over four large river basins of South America (Orinoco, Amazon, Tocantins and Parana). At the seasonal time scale, precipitation and total water storage correlate well in the Parana basin, with a few weeks lag of storage with respect to forcing. Over the Amazon, Tocantins and Orinoco, the two variables also correlate well. But in some years, storage response to forcing is enhanced, suggesting that other terms of the water balance (e.g., runoff) play a significant role. To investigate this, discharge data at the most downstream stations in these river basins are analysed, while the water balance is studied using outputs of global hydrological models available over the same time span as GRACE data. We also analyse water level data from ENVISAT altimetry over the main rivers. Finally, we study the interannual connection between rainfall and water storage, using among others, Empirical Orthogonal Functions (EOF). Compared to the seasonal cycle, the interannual signal displays larger regional variability both in precipitation and water storage.
G31B-0650
Change in land water storage in the East Africa region inferred from GRACE and altimetry data.
Drought can be regarded as one of the most damaging of natural disasters in human, environmental, and economic terms. It occurs as a result of extremes in climate that are driven by natural variability but may be exacerbated or dampened by anthropogenic influences. In East Africa rainfall exhibits a great spatial and temporal variability. Such events have impact on the water budget of this region. But water use and more generally anthropogenic forcing also affect regional hydrology. In this study, we investigate water storage change (surface and ground) using in synergy satellite radar altimetry, GRACE satellite gravity and other data to quantify recent change in surface waters and total land storage in East Africa over the recent years. Water levels of most East African lakes display significant decrease since the strong ENSO event of 1997- 1998. GRACE data available since 2002 also show decrease in total water storage over this region. The volume of water stored within lakes and reservoirs is a sensitive proxy for precipitation and may be used to study the combined impact of climate change and water-resource management. We also combine GRACE, altimetry and precipitation data sets to explore the relative contributions of the source term to the seasonal and interannual hydrological balance of this area and its link with the western Indian Ocean thermal change
G31B-0651
Using InSAR to Evaluate Pumping-Related Aquifer-System Response Between 1992 and 2007 in Ground-Water Basins of Eastern Nevada
Interferometric synthetic aperture radar (InSAR) has become a commonly used tool to detect and measure the magnitude and spatial variation of aquifer-system response, specifically subsidence, in groundwater basins in Nevada. Previous work has included InSAR studies of Las Vegas, which has a well-documented history of water-level changes and subsidence. The purpose of this study was to extend InSAR studies from Las Vegas to the north to Ely, Nevada in order to evaluate on a reconnaissance level the present groundwater system response to pumping in 35 hydrographic basins. These data will form a baseline for the proposed Southern Nevada Water Authority's In-state Groundwater Project which is designed to provide as much as 202 hm3/yr (164,000 acre-ft/yr) from 6 of these basins to the Las Vegas metropolitan area. We have processed more than 100 interferometric pairs using ERS and Envisat data from the WInSAR and GeoEarthscope archives covering the time period 1992-2007. Results were analyzed in time series in order to identify potential atmospheric and topographic artifacts, and to verify the occurrence of the groundwater signal. The preliminary results show that some principal groundwater basins currently undergoing moderate levels of groundwater pumping exhibit small, localized subsidence signals of a few centimeters for one or more years. These basins include Lake, Patterson, Butte, and White River Valleys, where annual pumping rates are on the order of 2.5-16.0 hm3/yr (2000-13,000 acre-ft/yr). Spring and Cave Valleys have localized subsidence signals away from known pumping centers. Localized signals are also located near the towns of McGill and Ely, Nevada in Steptoe Valley. These small amplitude signals are consistent with the low to moderate levels of pumping presently occurring in these valleys. The InSAR results also showed that some other basins undergoing moderate levels of pumping do not exhibit any visible evidence of aquifer-system impact. Due to the proximity of proposed new production, the geometry and magnitude of existing production, the Muddy Springs area was targeted for close examination. The results showed that although the area produces about 8.7 hm3/yr (7,000 acre-ft/yr) of pumped water from both the carbonate and alluvial aquifers and has experienced minor water-level declines, no visible subsidence was detected over a multi-year period. Similarly, basins containing earth fissures, common indicators of aquifer-system compaction and/or pumping strain, were analyzed for evidence of subsidence. Fissures are present in Garnet, Dry Lake, and Delamar Valleys, but InSAR results revealed no visible subsidence signals in these valleys, indicating that the fissures are not groundwater related. Of particular interest, InSAR revealed a subsidence signal associated with oil production in Railroad Valley. Results show that up to several centimeters of subsidence occurred in the Trap Spring field during the early to mid-1990's when the field initially went into major production, and that the signal was controlled by faults bounding the reservoir.
G31B-0652
Accuracy Assessment of GPS Buoy Sea Level Measurements for Coastal Applications
The GPS buoy in this study contains a geodetic antenna and a compact floater with the GPS receiver and power supply tethered to a boat. The coastal applications using GPS include monitoring of sea level and its change, calibration of satellite altimeters, hydrological or geophysical parameters modeling, seafloor geodesy, and others. Among these applications, in order to understand the overall data or model quality, it is required to gain the knowledge of position accuracy of GPS buoys or GPS-equipped vessels. Despite different new GPS data processing techniques, e.g., Precise Point Positioning (PPP) and virtual reference station (VRS), that require a prioir information obtained from the a regional GPS network. While the required a prioir information can be implemented on land, it may not be available on the sea. Hence, in this study, the GPS buoy was positioned with respect to a onshore GPS reference station using the traditional double- difference technique. Since the atmosphere starts to decorrelate as the baseline, the distance between the buoy and the reference station, increases, the positioning accuracy consequently decreases. Therefore, this study aims to assess the buoy position accuracy as the baseline increases and in order to quantify the upper limit of sea level measured by the GPS buoy. A GPS buoy campaign was conducted by National Chung Cheng University in An Ping, Taiwan with a 8- hour GPS buoy data collection. In addition, a GPS network contains 4 Continuous GPS (CGPS) stations in Taiwan was established with the goal to enable baselines in different range for buoy data processing. A vector relation from the network was utilized in order to find the correct ambiguities, which were applied to the long-baseline solution to eliminate the position error caused by incorrect ambiguities. After this procedure, a 3.6-cm discrepancy was found in the mean sea level solution between the long (~80 km) and the short (~1.5 km) baselines. The discrepancy between a 40-km and the 1.5-km baselines was around 1 mm. These discrepancies were caused, in part, by the atmospheric delays, mutlipath and others. Suppose the short- baseline solution is the truth, the discrepancies are therefore regarded as the accuracy upper limit for this type of GPS buoy solution.
G31B-0653
Mexico City Subsidence
Many parts of Mexico City are undergoing rapid subsidence due to extraction of ground water in excess of natural recharge. We use PSInSAR (Persistent Scatterer InSAR) to investigate the link between surface deformation and ground water withdrawal. From 23 Envisat acquisitions, we obtain a time-series of surface deformation between January 2004 and July 2006 showing that the eastern part of Mexico City presents a high subsidence rate as high as 30 cm/year. This subsidence is clearly tied to declining ground water levels. We use TU Delft's Doris and PSI Toolbox for interferogram generation and PS processing. Assuming constant subsidence rates, we have obtained a dense, coherent deformation map throughout the area. Comparison between GPS and PSInSAR inferred rates show agreement in the longer term signal; we will also show initial results of GPS/PSInSAR comparison on shorter time scales. The high density of persistent scatterers allows us to detect differential deformation of urban infrastructure. Our analysis indicates that in some cases the rate of an individual persistent scatterer can differ about 3-4 cm/yr from local deformation. While some bigger buildings show slower subsidence, others have faster rates. It is important to note that in some large buildings we detected differential vertical motion, suggesting that these buildings are subject to shear stresses associated with local deformation.
G31B-0654
One Dimensional Bedrock-Aquifer Subsidence Model Based on InSAR-Observed Mine Subsidence in North-Central Nevada
Bedrock-Aquifer response from groundwater pumping is a rarely observed and, therefore, poorly understood phenomenon. Typically, the magnitude of drawdown necessary to create effective stresses high enough to induce significant and, therefore, observable surface subsidence is not reached. This study documents a case where mine-dewatering induced drawdown of greater than 500 m led to bedrock subsidence magnitudes of approximately 50 cm in an 8 year period. Ground subsidence related to mine-dewatering is a common occurrence due to the large volumes of water that are pumped in order to lower the local groundwater table to facilitate open pit and underground mining operations. Several mines located along the Carlin Trend of Central Nevada have produced InSAR-identified subsidence signals of greater aerial extent and magnitude than most municipal groundwater signals. The dewatering signal at one such mine shows a minimum of 45.8 cm of cumulative subsidence between June 1, 1992 and September 21, 2000. This study has created many (>50) interferograms, allowing a better understanding of how the subsidence signal evolved in response to varied pumping rates from dewatering wells. The deformation signal correlates well with the observed groundwater drawdown characteristics. However, since the spatial resolution of the InSAR is much better than that of the monitoring well locations, the complexity of the signal is better delineated. The maximum aerial extent of the subsidence feature extends as far as 20 km away from the location of the extraction wells used for dewatering. The InSAR results were utilized to test a new, simple, one dimensional bedrock-aquifer subsidence model. This model utilizes easily obtained rock strength and rock mass parameters along with drawdown magnitudes to estimate vertical strain within the bedrock aquifer. Comparison of InSAR-identified subsidence magnitudes with estimated subsidence magnitudes from the new model agree quite well, particularly in areas of maximum drawdown.
G31B-0655
The Potential for Groundwater Monitoring Using GRACE, InSAR and GPS
The task of managing groundwater resources globally is becoming increasingly more difficult due to exponential population growth and climate change. Remote sensing will provide the tools needed to monitor changes in groundwater on regional and global scales, allowing water storage observations to be made in areas of the world without proper infrastructure for ground-based monitoring systems, and leading to a greater understanding of the global water cycle. Currently, remote sensing of water resources is available via the Gravity Recovery And Climate Experiment (GRACE), Interferometric Synthetic Aperture Radar (InSAR), and other GPS networks, each with inherent strengths and limitations. Separating groundwater from other components of the water balance equation in the vertically integrated gravity signal provided by GRACE requires ancillary observation of precipitation, surface water, soil moisture, and evaporation. InSAR has greater capability of separating observations of different water storage components, yet is limited by issues of temporal decorrelation caused by changes in the land surface that do not affect GRACE. We present a preliminary proposal to integrate GRACE, InSAR, and other GPS networks to provide a comprehensive method of monitoring groundwater resources remotely.
G31B-0656
A comparison between GRACE observations, hydrology model predictions and ground gravity measurements by the superconducting gravimeters of the Global Geodynamics Project (GGP) in Europe
We estimate the total water storage variations for the period 2002-2007 integrated over different groups of river basins in Central Europe from the GRACE solutions of time variable gravity field. These observations are compared to soil moisture estimates from global hydrology models such as GLDAS and LaD World, as well as with precipitation fields. We focus on extreme events such as the summer 2003 and autumn 2006 European heat waves. An estimate of the time delay for the closure of the budget of total water with respect to the winter 2002-2003 level is drawn from the GRACE observations. Gravity residuals due to the total water storage variations are computed from the corrected measurements of the superconducting gravimeters of the Global Geodynamics Project (GGP) in Europe and are decomposed into principal components by an EOF analysis in order to extract the common signal from the different stations. The same analysis is also done on the GRACE solutions of the gravity field over the same area and for the same period as for the ground gravity residuals. We show a good agreement between the phases of the first principal components but discrepancies between the amplitudes are found. Satellite-derived gravity differs from ground gravity because of the free-air effect of the vertical displacement of the Earth's surface. We discuss about the best way to remove this geometrical effect when comparing both types of gravity data and compute the corrected gravity time series.
G31B-0657
First Gravity and Geodetic Results from the GHYRAF (Gravity and HYdRology in AFrica) Experiment
Continental hydrology (soil moisture, aquifers) processes redistribute underground water and hence lead to alter both the gravity and shape of the Earth because of loading effects involving elastic deformation and Newtonian attraction contributions. We started recently a new experiment (2008-2010) in West Africa called GHYRAF (Gravity and Hydrology in Africa) ; one of the objectives is to retrieve the signature in gravity (ground and satellite-derived) and vertical GPS due to water storage changes. This research program is based on measurements mainly concentrating geographically on three distinct regions: the Sahara desert zone with almost no rainfall, the Sahelian zone (Niamey and Diffa in Niger) with moderate and highly variable rainfall and the equatorial monsoon region (Djougou, Benin Republic) with heavy rainfall. Our experiment consists in repeated ground gravity measurements with an absolute gravimeter (FG5) on a regular basis (3 to 4 months) of these base stations. All the gravity points are collocated with permanent GPS stations in order to independently estimate the gravity contribution due to vertical motions of hydrological origin. Since gravity is sensitive to the various length scales involved in hydrology, we rely on dense in-situ measurements (rain gauges, piezometers, soil moisture probes) to model local gravity effects. It is also planned to continuously monitor the time-variable gravity with a superconducting gravimeter (SG) at a fixed station in the monsoon zone. The primary goal is to better characterize the annual cycle of water storage in West Africa and to assess the predictions of global hydrology models (e.g. GLDAS, LadWorld) for this region. Our project will also allow us to validate GRACE satellite gravity observations with ground gravity and GPS observations in a region with large hydrological signal. The first results of the GHYRAF experiment will be shown as follows: 1/ the first AG measurements on the profile in Niger and Benin 2/ a preliminary treatment of the GPS data available in the region mainly originating from the AMMA (African Monsoon Multidisciplinary Analyses) follow-on program and their comparison with theoretical predictions for the continental scale hydrological loading 3/ an estimate of the GRACE derived gravity changes as observed on the Niger and Chad Basins.
G31B-0658
The 2005 Amazon drought as measured by GRACE and estimated by climate models
Satellite gravity measurements from the Gravity Recovery and Climate Experiment (GRACE) provide new quantitative measures of the 2005 extreme drought event in the Amazon river basin, regarded as the worst in over a century. GRACE measures a significant decrease in terrestrial water storage (TWS) in the central Amazon basin in the summer of 2005, relative to the average of the 5 other summer periods in the GRACE era. 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.
G31B-0659
Using Custom Designed GRACE Mass Concentration Blocks Over Southern Africa for Evaluating Terrestrial Water Storage Models
Hydrological models have traditionally been calibrated using head observations and discharge data. In many regions of the world, this kind of data is however scarce, and the need for better models has long ago been recognised. Our target region is Southern Africa, which has proved notoriously difficult to model due to the lack of river gauges. Comparisons between the hydrological model GLDAS, GRACE 4 by 4 deg Mascon blocks, and spherical harmonic products show large phase and amplitude discrepancies on seasonal signals throughout the region. We have set up 4 hydrological models of the 4 largest river basins in Southern Africa (Zambezi, Okavango, Limpopo and Orange) using SWAT (Soil and Water Assessment Tool), and analysed the gravity changes from GRACE through custom designed Mass Concentration blocks (Mascons). A system of 644 blocks, covering an area of approximately 37 by 35 deg has been set up. The Mass Concentrations have been solved at 10-day intervals using 1.25/1.5 by 1.5 deg blocks, constrained according to river basins and sub- basins matching the hydrological models. In this presentation we will compare and evaluate the new GRACE data against existing data, GLDAS, and our model results from SWAT.
G31B-0660
Representation of Hydrological Information Within the Forthcoming new GDEM, ACE2
The ability to work on a continental scale (due to technological advances) increasingly allows the possibility of assessing climate and hydrological variables globally. This in turn has led to an increasing demand for Global Digital Elevation Models (GDEMs) to support hydrological investigations both by incorporating into their quality assessment those variables relevant to hydrology and by encoding additional hydrological information within their matrices. This paper presents the approach taken towards fulfilling this requirement for the forthcoming new GDEM, ACE2. ACE2 is being generated by fusing the results from the Shuttle Radar Topographic Mission (SRTM), with over 100 million height points derived from multi-mission retracked Satellite Radar Altimetry data, including the ERS-1 Geodetic mission. The results of a global assessment of the SRTM DEM over inland water are presented, demonstrating that from 800,000 points over lakes more than 90 percent of SRTM points lie within the SRTM stated accuracy of +/-16m with a modal difference of 2m. The success of the SRTM is geographically dependent with some areas displaying much higher correlation than others. In general, however, the inland water heights derived from radar altimeter data are shown to give more consistent and reliable heights than those within the SRTM dataset; accordingly, for ACE2 all lake heights are being replaced by altimeter derived values. Thanks to successful retracking, decadal timeseries of height variations from ERS-2/EnviSat and Topex/Jason1 are also available over a number of waterbodies globally. These allow the potential inclusion of additional hydrological information such as spatial extent, quantification of temporal variation and generic trends in the water levels. Results are presented showcasing how this additional information could be included within a GDEM and assessing the extent to which including these data could assist hydrological applications.
G31B-0661
InSAR time-series analysis of land subsidence due to groundwater overexploitation in groundwater basins of central and northeast Iran
We use Interferometric Synthetic Aperture (InSAR) observations acquired by the Envisat satellite during 2002-2008 to study land subsidence caused by groundwater over-exploitation in groundwater basins of central and northeast Iran. Earlier studies in these regions using a few interferograms showed rapid subsidence in the order of several centimetres per year. Here we use additional ENVISAT data and perform a time-series analysis using the small-baseline concept to better determine the temporal and spatial evolution of surface deformation.
G31B-0662
Improved Jason-2 Altimetry Products for Coastal Zones and Continental Waters (PISTACH Project)
As part of Jason-2 project, CNES is currently funding a dedicated study to improve satellite radar altimetry
products over coastal areas and continental waters.
The PISTACH (Prototype Innovant de Système de Traitement pour les Applications Côtières et l'Hydrologie)
project is organized around 3 phases:
*Phase 1: user needs and structure of coastal/hydrological products
*Phase 2: Development of new dedicated algorithms: retracking of the waveforms, wet and dry tropospheric
corrections, local models or high resolution global models for topography, geoid, land cover classification,
land water mask, data editing
*Phase 3: prototype implementation, validation and operations during Jason-2 CalVal phases
The input of the prototype will be constituted by Jason-2 Level 2 S-IGDR altimeter products, ECMWF meteo
fields, as well as several state of the art static auxiliary datasets. The first version of PISTACH products
should adopt the same format and structure as Jason-2 standard IGDR to facilitate their appropriation and
assessment by expert users. However, more simple and easy to-use products are envisaged for a wider
dissemination.
The implementation of the prototype will be completed in September 2008 and the first products should be
accessible in October.
The project, the prototype and the products will be presented at the meeting and the content of the products
will be illustrated and assessed over several test areas.
G31B-0663
Satellite-based Estimates of Global Freshwater Discharge
As a critical component of the global hydrologic cycle, freshwater discharged into the global ocean is imperative to the climate system and the human environment. To date, most global discharge assessments were based on the aggregated gauge-based streamflow measurements, which are severely limited spatially, temporally and conceptually. As a result, most prior assessments were restricted to reporting long- term averages global freshwater discharge estimates. Here we present month-to-month variations of global freshwater discharge, extending from 1994 to 2006, utilizing satellite and in-situ observations of the major hydrologic fluxes and states in a global ocean freshwater budget. The results presented here are the only available estimates of global discharge computed on a monthly basis over an extended period (1994-2006) from the recent past. The computed estimates of global freshwater discharge are subsequently analyzed in the context of trend attribution. Results for the entire study period indicate a significant trend in global discharge (4.6 km3/month) that is largely due to increasing global ocean evaporation (5.1 km3/month).