Global Reservoirs and Lakes Monitor (G-REALM)


Near Real Time products with datum based on a single satellite overpass (1 day)
Status products with datum based on a multi-year mean
Top

Specific Notes

  • Regarding surface elevation products:
  1. It is important to consider the location of the satellite ground track AND the section of track used in complex or drought-prone regions where height variability may be location dependent. In such cases check the specific geographical location in the text product header.
  2. The more recent radar altimeters are using an on-board DEM to more rapidly acquire the water surface. There are cases where this DEM setting is incorrect resulting in either a) the seasonality not being captured or b) the observed seasonality being dampened. End users must check the product summary Tables (column H) which highlight historical or current potential DEM issues for the lake/reservoir.
  3. The smoothed height variation (lower graphs) products are not available. These can vary according to the smoothing function applied and so are provided as a visualization aid only.
  4. The products are routinely upgraded. End users should note the product version number in the graph legend and table header, and avoid utilizing products of mixed heritage.
Top

Recent Updates

  • 20th December 2023: The 27-day resolution lake product updates are in hold until March 2024 while the project adapts to the upgraded (Version 5) datasets available from the Sentinel-3A,-3B missions.
  • August 2023: All lake products in the State of Alaska have been slightly modified to reflect a change in atmospheric correction.
    The 10-day resolution time series graph products now include a zoomed-in snapshot of the last 3 years of water level variation.
  • July 17th 2023: Products for the Kakhovskoye Reservoir (000873, 100873, 200873) have been revised. Note that severe water loss due to a dam breach in June 2023 will result in erroneous elevation values as the dry lake floor emerges.
  • October 2022: The 10-day product for 004103.Manwan was incorrect, it has now been corrected.
  • August 2022:The 10-day resolution water level products are once again operational. Data from the Sentinel-6A Michael Freilich mission is now offering near real time lake monitoring. The existing TOPEX/Jason 10-day products are being extended using the Sentinel-6A data. Cross-checking of the updated time series by the team will continue until end of October 2022.
  • March 21st 2022: On April 7th 2022 the Jason-3 satellite will be moved to its new end-of-life orbit. The Sentinel-6 Michael Freilich satellite (previously called “Jason Continuity of Service” or Jason-CS) will take over the monitoring of the lake and reservoir surface heights at 10-day resolution. Routine product updates will be halted in April as the old (Jason-3) and new (Sentinel-6) time series are merged and validated. We hope to resume operational updates by the summer of 2022.
  • January 27th 2022:
    a. Revised water level products for 004852.Saslaj and 001597.Tekeze are now available. Please reject previous versions.
    b. The following lake products have had a change of ID number - Lake Amboseli (Kenya) are now ID 110159 and 010159. Lake Eyre (Australia) has changed to ID 010068, and Lake Torrens (Australia) has changed to ID 010065.
  • January 2022: The product tables now include column H which highlights yes/no as to whether the water product has been, or is being, affected by on-board DEM issues. See the "Important – Read Me" link for details.
  • August 25th 2021: Target ID numbers have changed from 4 digits to 6 digits. In most cases IDs remain the same except that two leading zeros '00' are now inserted at the beginning. Example, 0314.Victoria is now 000314.Victoria
  • April 5th 2021: Operational lake product updates have resumed, however checks on January to March 2021 height estimates are ongoing. Minor edits may follow.
  • February 2021: The updating of the operational products is on hold while a recent change in Jason-3 satellite data format is being investigated.
  • January 2021: The Lake Product generating system has recently moved to a new server. During the process, the time series were manually inspected. While keeping to the same product Version Number, please note:
    a. Lake Batman (Turkey) has changed lake ID number to 3910.
    b. 0498.Mallery has been upgraded across its entire time series.
    c. The following lakes have been minimally edited across their Jason-3 period (Lake ID’s 0091,0152,0156,0341,0380,0398,0406,0425,0443,0509,0514,0519,0744, 1000,1259,1723,2282,5620,8001,9150 and notably 0277.Aral_Sea_2, 0342.Neuseidl, 0368.Salton_Sea, and 0958.Vallabhsagar.
    d. 1992.Xiaowan, 2292.Indrawati, 2323.Gilgel_Gibe_III, 4034.Tehri, 6446.Jinping_I, 6871.Huandeng have been edited to a greater degree to remove poor height values introduced due to incorrect on-board DEM settings.
    e. A recent change in on-board DEM has also created errors for the following lakes and correct monitoring has halted, 2282.Xiaolangdi, 2292.Indrawati, 2323.Gilgel_Gibe_III, 4034.Tehri, 6446.Jinping_I, 6871.Huandeng.
    f. The seasonality of the newly created 6871.Huandeng reservoir is also not being correctly represented. Again, this is likely due to an incorrect on-board DEM setting.
    g. All DEM issues have been reported and will likely be remedied at the next Jason-3 DEM upgrade in 2021.
  • 27th October 2020: Recent water level variations (post Sept 2020) for 2292.Indrawati, 2323.Gilgel_Gibe_III, 4034.Tehri and 6871.Huangdeng are being investigated.
  • 20th October 2020: The 10-day resolution product for 1500.Srisailam has been revised.
  • September 2020: The time series product for the Caspian Sea has been revised. The 2016 to present day elevation variations have been notably modified.
  • June 2020: Recently added 35-day and 27-day products have been revised to version 2.1. Product formats have also been modified to match those of the 10-day products.
  • March 2020: Upgraded 35day resolution water level products and new 27-day resolution products are now additionally available from the top global map.
  • Feb 24 2020: Minor upgrades of the water level products have been performed and product filenames have been modified.
  • December 20th 2019: A revised water level product for 0504.Manicouagan is now available. Please reject previous versions.
  • May 10th, 2019: The majority of the water level products have undergone an upgrade from version 2.3 to version 2.4. The upgrade includes improved accuracy and an extending back in time to 1992 where possible. Users are asked to i) note the product graph legend and product text file header for the version number, and ii) not to mix 2.4 and 2.3 product results.
  • August 18th 2017: Operational updates of the lake water level products has resumed.
  • July 2017: Weekly updates to the lake products have been halted due to technical issues. Operations will resume in August.
  • December 2016: Operations have resumed with data from the NASA/CNES Jason-3 mission. New operational products (e.g. lake0314.TPJOJ.2.txt) are now available. The corresponding status products (e.g. lake0314.TPJOJ.1.txt) will follow shortly. Note the product filename change to "TPJOJ". The merger of Jason-3 results to Jason-2 may be poor where there is little overlapping Jason-3 data, and such products will be reassessed in 2017. The G-REALM team have noted that for a few lakes/reservoirs (e.g. 1619.Three Gorges Dam) there has been no Jason-3 data available so far, and this has been due to a change in surface tracking algorithm. Both the team and the mission agency's are striving to rectify this as soon as possible.
  • A table depicting which lake/reservoir products are available, and over what time period, can now be viewed. Look to the clickable link within the Project Information section below-left.
  • End users must consider the location of the satellite ground track AND the section of track used to create the products. This is particularly important for complex and drought-prone lakes and reservoirs, where height variability may be location dependent. See the product headers.
  • End users must also note that the graphs/products associated with the filtered time series are provided as a visualization aid only.
Top

Background

The U.S. Department of Agriculture's Foreign Agricultural Service (USDA-FAS), in co-operation with the National Aeronautics and Space Administration are routinely monitoring lake and reservoir height variations for many large lakes around the world. The program utilizes NASA/CNES/ESA/ISRO radar altimeter data over inland water bodies in an operational manner. The surface elevation products are produced via a semi-automated process and placed at this web site for USDA and public viewing. Monitoring height variations will greatly assist the USDA/FAS Office of Global Analysis to quickly locate regional droughts, as well as improve crop production estimates for irrigated regions located downstream from lakes and reservoirs. Reservoir and Lake height variations may be viewed in graphical and text format by placing the cursor on and clicking the continent and lake of interest.
Top

Semi-Automated Data Processing

The project currently utilizes near-real time data from the Jason-3 mission, and archive data from the Jason-2/OSTM, Jason-1, Topex/Poseidon, and ENVISAT missions. Data processing procedures closely follow methods developed by the NASA Ocean Altimeter Pathfinder Project (see references). When fully operational, updated products are delivered within 7-10 days after satellite overpass. The resulting time series of height variations are expected to be accurate to better than 10cm rms for the largest (and more open) bodies of water such as The Great Lakes, USA, Lakes Victoria and Tanganyika in Africa etc. Smaller lakes or those that experience more sheltered (from wind) conditions can expect to have accuracy's better than 20cm rms (e.g. Lake Chad, Africa). Satellite passes that cross over narrow reservoir extents in severe terrain will push the limits of the instruments with resulting rms values of many tens of centimeters. Despite limitations, satellite radar altimeters can potentially monitor the variation of surface water height for many large inland water bodies including lake, reservoir, wetland region and river channel. Full details and references can be found by clicking the associated links in the left-hand frame.
Top

Satellite Radar Altimetry

In General: A satellite radar altimeter is not an imaging device, but a nadir-pointing instrument continuously recording average surface `spot' heights directly below the satellite, as it transverses over the Earth's surface. Operating at 13.6 or 35.75 GHz, each altimeter emits a series of microwave pulses towards the surface. By noting the two-way time delay between pulse emission and echo reception, the surface height can be deduced. Each returned height value is an average of all surface heights found within the footprint of the altimeter. The diameter of the footprint depends on the surface roughness, but can typically range between 200m (for open pools of water in calm conditions) to a few kilometers (open water with surface waves). Each satellite is placed in a specific repeat orbit, so after a certain number of days the same point (to within 1km), on the Earth's surface is revisited. In this way, time series of surface height changes can be constructed for a particular location along the satellite ground track during the lifetime of the mission.
There have been a number of altimetric satellite missions to date and follow-on missions can be expected (see Figure and Table below).
General Timeline for Satellite Radar Altimeters
Instrument Summary
Satellite Operation Repeat Period
Topex/Poseidon 1992-2002 10 days
Jason-1 2002-2008 10 days
Jason-2/OSTM 2008-2016 10 days
Jason-3 2016-current 10 days
Sentinel-6 2020-current 10 days
HY-2A 2011-current 14 days
Seasat 1978 17 days
Geosat 1985-1989 17 days
GFO 2000-2008 17 days
SWOT-nadir 2022-current 21 days
Sentinel-3A 2016-current 27 days
Sentinel-3B 2018-current 27 days
ERS-1 1992-1993, 1995-1996 35 days
ERS-2 1996-2003 35 days
ENVISAT 2002-2010 35 days
SARAL 2013-2016 35 days
ICESat-1 (laser) 2003-2009 90 days
Cryosat-2 2010-current 365 days
ICESat-2 (laser) 2018-current variable
SWOT-swath 2022-current variable

Although their primary objectives are ocean and ice studies, altimeters have had considerable success in the monitoring of inland water bodies. In particular, the ability to remotely detect water surface level changes in lakes and inland seas has been demonstrated. Unhindered by time of day, weather, vegetation or canopy cover, the technique has further been applied to a number of rivers, wetlands and floodplains in several test-case studies. In particular, the results demonstrate how submonthly, seasonal, and interannual variations in height can be monitored.
For full details on Satellite Radar Altimetry and the application to inland water see the References section.
Top

Advantages

  • Day/night and all weather operation.
  • Generally unhindered by vegetation or canopy cover.
  • All determined surface heights are with respect to one common reference frame.
  • Satellites are placed in repeat orbits (up to 1km either side of a nominal ground track) enabling systematic monitoring of rivers, lakes, wetlands, inland seas and floodplains.
  • Has the potential to contribute height information for any target beneath the satellite overpass, thus contributing information where traditional gauge (stage) data may be absent.
  • Satellite altimetric instruments have been in continuous operation since 1991 and new missions are scheduled for the next decade. There is therefore the ability to monitor seasonal to interannual variations during the lifetime of these satellites.
  • Techniques have been validated and results published in peer-reviewed journals.

Limitations

  • These instruments are primarily designed to operate over uniform surfaces such as oceans and ice-sheets. Highly undulating or complex topography may cause data loss or non-interpretation of data.
  • Retrieved heights are an "average" of all topography within the instrument footprint. Such values are further averaged in the direction of the satellite motion, giving, for example, one final height value every 580m (TOPEX/POSEIDON) or 350m (ERS) along the ground track. Altimetric values therefore differ from traditional gauge measurements which offer "spot" heights at specific locations.
  • The height accuracy is dominated by knowledge of the satellite orbit, the altimetric range (distance between antenna and target), the geophysical range corrections and the size and type of the target.
  • Unlike imaging instruments, altimeters only retrieve heights along a narrow swath determined by the instrument's footprint size. The effective footprint diameter can vary depending on the nature of the target, and can potentially range from several hundred meters to many kilometers.
  • Minimum target size is controlled by the instrument footprint size and the telemetry/data rates, and also on the surrounding topography and the target-tracking method used.
  • The satellite orbit scenario and target size also determine the spatial and temporal coverage. Improved temporal coverage is gained at the expense of spatial coverage for a single satellite mission.
  • Major wind events, heavy precipitation, tidal effects and the presence of ice will effect data quality and accuracy.
Top

Datasets

Several altimetric datasets are currently being exploited: 1) The Topex/Jason series in a 10-day repeat orbit and with global coverage extending to North/South latitude 66 degrees. This comprises T/P (1992-2002), Jason-1 (2002-2008), Jason-2 (2008 to 2016), and Jason-3 (2016 to present), and utilizes a mix of both IGDR and GDR data sets. 2) ENVISAT data with a 35-day resolution and extending to North/South latitude 81 degrees. Currently Jason-3 IGDR data are being used for near real time operational monitoring at 10-day resolution. The ENVISAT products will be extended in time with data from the historical ERS and SARAL missions, and they will be appended with Sentinel-3A measurements which will provide near real time monitoring at 27-day resolution.
Top

Products

The aim of this web site is to provide time-series of water level variations for some of the world's largest lakes and reservoirs. Currently, lakes ≥ 100km2 are included but future project phases will aim to include those in the 50-100km2 size range.
The main database products are graphs and associated information in tabular form. For the Graphs, changes in water level are real but the y-scale is arbitrary (relative) and given in meters. The x-axis refers to time with intervals of several months. The blue symbol represents results from the Topex/Poseidon satellite (the NASA-ALT and SSALT/Poseidon-1 altimeters), the red symbol denotes results from the Jason-1 mission (the Poseidon-2 altimeter), the purple symbol denotes the OSTM or Jason-2 mission (the Poseidon-3 altimeter). Additional graphs may also depict results (green symbol) from the GFO mission.
The Results Table gives heights, associated errors and date/time of the observation. Note that a geographical extent across the lake has been used to derive the time series - rather than a spot measurement which is more typical of a traditional gauge. A discussion on altimetric height accuracy can be found in the Accuracy+Validation section.
Top

For More Information Contact:

  • Dr. Charon Birkett
    Code 61A, NASA/GSFC
    Greenbelt, USA
    Charon.M.Birkett@nasa.gov
  • Dr. Curt Reynolds
    USDA, Foreign Agricultural Service
    Global Market Analysis, International Production Assessment Division (GMA-IPAD)
    1400 Independence Ave, SW
    South Building Stop 1051, Room 4609,
    Washington, DC 20250
    USA
    Curt.reynolds@fas.usda.gov
Top

Funding Acknowledgement

These lake products exist in the public domain and their creation was funded as part of the USDA/FAS/GMA and NASA Global Agriculture Monitoring (GLAM) Project and the NASA/Applied Sciences/Water Resources Program. The following general acknowledgement of this database should be made if the information presented here is used in publications for further scientific purposes and/or additional applications:
Lake products courtesy of the USDA/NASA G-REALM program at https://ipad.fas.usda.gov/cropexplorer/global_reservoir/.
Top

Disclaimer

Users of these datasets must carefully note the information given in the Accuracy+Validation and Advantages and Limitations sections.
This is an on-going project with elements that reside in the research domain. We therefore reserve the right to state the following liability disclaimer:
The USDA/NASA Project Investigators accept no responsibility for the accuracy and application of the lake level products held in this database.