Contributors: B. Calmettes (CLS), L. Carrea (University of Reading), C.J. Merchant (University of Reading)
Issued by: B. Calmettes
Date: 21/04/2021
Ref:C3S_312b_Lot4.D3.LK.6-v3.0_LWL_Product_User_Guide_and_Specification_i1.0.docx
Official reference number service contract: 2018/C3S_312b_Lot4_EODC/SC2
History of modifications
List of datasets covered by this document
Related documents
Acronyms
General definitions
'Orbit' is one revolution around the Earth by the satellite.
A satellite 'Pass' is half a revolution of the Earth by the satellite from one extreme latitude to the opposite extreme latitude.
'Repeat Cycle' is the time period that elapses until the satellite flies over the same location again.
'Range' is the satellite-to-surface pseudo distance given from the 2-way travel time of the radar pulse from the satellite to the reflecting water body
'Retracking' is the algorithm which computes the altimetric parameters (range, backscatter coefficient...) from the radar echoes
'Altitude' is considered as the height over the reference ellipsoid (orthometric height).
'Mean Profile' is an average over several years of the surface heights with respect to geoid, computed following the satellite's repetitive tracks.
Scope of the document
The Product User Guide and Specification (PUGS) is the primary document that users need to read before handling the products. It gives an overview of the product characteristics, in terms of algorithm, technical characteristics, and main validation results. The Appendix A - Specifications contains the specification of the product, based on the statements of the lake ECV from GCOS.
Executive summary
The C3S Lake production system (C3S ECV LK) provides an operational service, generating lake surface water temperature and lake water level climate datasets for a wide variety of users within the climate change community. The present document covers the lake water level system. The water level products in the C3S aim at providing users with water level time series on lakes located beneath satellites tracks. Since satellite altimetry is not an imagery technique (measurements being only possible at nadir), both the ground coverage and the temporal repeatability is strongly dependent on the orbital characteristics of the altimeters.
This Product User Guide and Specification (PUGS) is a self-contained document which gathers all necessary information to use the Water Level products in an efficient and reliable way.
Product Change Log
The following Table, Table 1, provides an overview of the differences between different versions of the product up-to, and including, the current version.
Table 1: Changes in the product between versions.
Version | Product Changes |
V3.0 | The spatial coverage (number of lakes) has increased. 166 lakes are now being monitored and data from Sentinel-3D is used |
v2.1 | The spatial coverage (number of lakes) has increased – 94 lakes are now being monitored. Envisat reprocessed data will be available in 2020. New datasets for intercomparison. New tests on absolute assessment: mainly the missing values as a function of the altimetric mission. |
v1.0 | First release of the dataset. CDR produced until 2018-12-31. |
Data description
Hydrological information is traditionally obtained via ground-based observation systems and networks that suffer from well-known inherent problems: high cost, sparse coverage (often limited to political local/national instead of geographical/hydrological boundaries), slow dissemination of data, heterogeneous temporal coverage, destruction of the stations during floods, absence of stations in remote areas, absence of management strategy, etc.
Although originally conceived to study open ocean processes, the radar altimeter satellites have nevertheless acquired numerous useful measurements over lakes. This technique, that can complement ground-based observations systems, potentially provides a major improvement in the field of continental hydrology, due to the global coverage (however limited to Earth's portion at nadir of the orbital ground tracks), regular temporal sampling and short delivery delays.
Early studies rapidly demonstrated the great potential of satellite altimetry to monitor large continental water bodies such as the Caspian Sea or the East African lakes (Birkett, 1995; Birkett et al, 1998; Mercier et al., 2002). However, these studies also raised many important issues related to the need of a dedicated treatment and interpretation of the measurements acquired over continental waters. More specifically, the radar echoes (waveforms) returned by inland waters can strongly differ from those returned by the ocean (because of the contamination of the signal by reflections over emerged lands), thus altering the accuracy of the height retrievals.
However, the recognition of these systems by the hydrological community is increasing, and new altimeter sensors are currently being designed following the requirements expressed by hydrologists for scientific use but also for the water resources management operational services that are emerging. For the moment, the scientific use is preponderant (Crétaux et al, 2011) and mainly concerns the global scale: monitoring of large climatically sensitive lakes or the study of the temporal water masses redistribution within a large basin.
The retrieval methodology
Background
Satellite Radar Altimetry
Radar altimetry from space consists of vertical range measurements between the satellite and water level. The difference between the satellite altitude above a reference surface (usually a conventional ellipsoid), determined through precise orbit computation, and satellite-water surface distance, provides measurements of water level above the reference. Placed onto a repeat orbit, the altimeter satellite overflies a given region at regular time intervals (called the orbital cycle), during which a complete coverage of the Earth is performed.
Water level measurement by satellite altimetry has been developed and optimised for open oceans. Nevertheless, the technique is now applied to obtain water levels of inland seas, lakes, rivers, floodplains and wetlands. Several satellite altimetry missions have been launched since the early 1990s: ERS-1/RA (1991-1996), TOPEX/Poseidon (1992-2006), ERS-2/RA (1995-2005), GFO (2000-2008), Jason-1 (2001-2012), ENVISAT/RA-2 (2002-2012), Jason-2 (2008-), CryoSat-2 (2010-), HY-2A (2011-), SARAL/AltiKa (2013-), Sentinel-3A (2016-), and Sentinel-3B (2018). ERS-1, ERS-2, ENVISAT and SARAL have a 35-day temporal resolution (duration of the orbital cycle) and 80 km inter-track spacing at the equator. TOPEX/Poseidon, Jason-1, Jason-2 and Jason-3 have a 10-day orbital cycle and 350 km equatorial inter-track spacing. GFO has a 17-day orbital cycle and 170 km equatorial intertrack spacing. Sentinel-3A orbit has a revisit time of 27 days, and its inter-tracking separation is 104 km. It is reduced to 52 km in the two-satellite configuration (Sentinel-3A and B).
Figure 1 shows the Jason passes over some Russian lakes. The combined global altimetry data set has more than 20 year-long history and is intended to be continuously updated in the coming decade. Combining altimetry data from several in-orbit altimetry missions increases the space-time resolution of the sensed hydrological variables.
Figure 1: TOPEX/Poseidon (Jason-1, Jason-2 and Jason-3) passes over the lakes Balkash, Yssik Koul, Alakol and Saysan. Note that in between those passes, no measurements are made by these satellites.
Computing surface height
Figure 2: Altimetry principle (Credit CNES – Mira production)
For all satellites, the following operation is done:
Surface Height = altitude – corrected_range
With:
Corrected_range = range + Wet_tropo + Dry_ tropo + iono + polar tide + solid earth tide
"Wet_tropo" is a correction to account for the delay of the radar signal due the wet part of the troposphere.
"Dry_tropo" is a correction to account for the delay of the radar signal due the dry part of the troposphere.
"Iono" is a correction to account for the delay of the radar signal due the electronic content of the ionosphere.
"Polar Tide" and "Solid Earth Tide" are two corrections that account for the temporal deformation at the Earth' surface.
Note that for SARAL/AltiKa, ionospheric correction is not taken into account due to the frequency used for this mission (Ka-band), which is not sensitive to the ionosphere electron content.
Finally, the water surface height is expressed with respect to the geoid:
Water Surface Height = Surface Height – geoid
The geoid value considered here is extracted from a mean profile file over large lakes.
Processing
Altimetry missions used are repetitive, i.e. the satellite overflies the same point at a given time interval (10, 27, 35 days… depending on the satellite). The satellite usually does not deviate from more than +/- 1 km across its track.
A given lake can be flown over by several satellites, with potentially several passes, depending on its surface area.
In the current version of the product, 160 lakes are monitored. Among them, 8 lakes are monitored with data from a single mission (Jason-3 data, Sentinel-3A or Sentinel-3B) 76 lakes with data from two missions and 2 lakes (Hottah and Iliamna) with the data from all three missions. Several passes can be used to derive the lakes' LWL timeseries according to their surface areas. The term "lakes" refers to lakes or reservoirs.
Input data
Past lake levels are based on merged TOPEX/Poseidon, Jason-1, Jason-2, Envisat, SARAL, IceSat and GFO data provided by ESA, NASA and CNES data centres. Updates include Jason-3, provided by CNES, and Sentinel-3A /Sentinel-3B data provided by Copernicus Hub.
- Altimetry data consist of Short Time Critical (STC) and Non-Time Critical (NTC) data, and they are available within 2-3 days and 1 month respectively after measurement acquisition. We advise the user that NTC and STC nomenclature are used for Sentinel data (since these data are provided by Eumetsat), these denominations correspond respectively to GDR (Geophysical Data Record) and IGDR (Interim Geophysical Data Record) for Jason-3 (data provided by CNES, hence CNES nomenclature is used). These two types of files include the altimetric measurements as well as the necessary corrections for their exploitation (orbit, atmospheric corrections…). GDR data were used for time series initialisation, IGDR data are now used for operational processing. GDR are still used for the research products. The main differences between GDR and IGDR data reside in the orbit and radiometer derived values (GDR benefit from more precise values as well as more recent GDR calibrations). Differences between these corrections are of the order of a few cm.
- Mean along-track profile over the lakes
- Location information, including missions and track number for all the processed lakes
Method
The altimeter range measurements used for lakes consist of 1 Hz IGDR and GDR data. All classical corrections (orbit, ionospheric and tropospheric corrections, polar and solid Earth tides and sea state bias) are applied. Depending on the size of the lake, the satellite data may be averaged over very long distances. It is, thus, necessary to correct for the slope of the geoid (or equivalently, the mean lake level). Because the reference geoid provided with the altimetry measurements (e.g., EGM96 for T/P data or EGM2008 for Sentinel-3A) may not be accurate enough, we have computed a mean lake level, averaging over time the altimetry measurements themselves. Such mean lake level surface along each satellite track across the lake provides a better estimate of the model geoid. Mean profiles are ttherefore used for lake water level computations which are then referenced with respect to this estimate of the geoid. If different satellites cover the same lake, the lake level is computed in a 3-step process:
- Extraction of input data: including an editing to remove erroneous measurements
- Lake level computation and filtering
- Generation of output NetCDF files, including lake water level
Each satellite data is processed independently. Potential radar instrument biases between different satellites are removed using TOPEX/Poseidon data as reference. We generally observe an increased precision of lake levels when multi-satellite processing is applied.
Limitations of the product
The lake processing chain, initially developed at LEGOS, is based on the use of altimetry measurements from the CTOH database (Center for Topography of the Oceans of LEGOS), including Envisat GDRs. This database includes, in addition, some enhanced corrections that were not in the first data sets of institutional suppliers (Aviso, ESA), such as for the wet and dry tropospheric as well as for the ionospheric corrections. As part of the online version of HYSOPE (that serves Copernicus Climate Change Service), these tailored corrections were replaced by their corresponding standard corrections included in the operational input products. In recent mission products, the geophysical corrections have been regularity updated by agencies which improves the quality of the estimates and makes it unnecessary to replace them. The main source of uncertainty comes from the wet tropospheric correction, and depending on the regions varies from 2-3 cm. In addition, a module for calculating the height of the geoid has been developed to improve assessment of water levels. , . Moreover, the overall uncertainty on water surface height depends mainly on the size of the track over the lake and comes mainly from retracking. It varies from 8-10 cm for large lakes (with longer transects) to 1m for small and narrow lakes.
Product description
The product provides estimation of water level relative to a reference at the times that a given satellite overpasses the lake. It also contains the standard deviation associated to this measurement and the time of the overpass. The product also contains the metadata necessary for a good understanding.
Product content
The product contains all the descriptive metadata in the global attributes of the netCDF file (Table 2):
Table 2: Metadata included in the product files
Attribute | Value |
---|---|
title | Lake Water Level from satellite altimetry |
lake | Name of the lake (full name), in local language wherever possible (a choice might be done if the lake shores several different countries) |
basin | hydrological basin or catchment name (full name), in English |
country | names (full name) of the countries where the lake is located (several names possible) |
lake_barycentre_latitude | Decimal degrees north |
lake_barycentre_longitude | Decimal degrees east |
surface_reference | The surface reference for the lake water level |
institution | LEGOS, CLS, CNES |
source | Original data source |
history | Processing history of the dataset |
references | https://cds.climate.copernicus.eu |
convention | CF convention |
product_version | Product version of the data file |
summary | A paragraph describing the dataset |
id | File name |
naming_authority | fr.legos.cls |
keywords_vocabulary | GEMET, GCMD and iso19115 keywords |
gcmd_keywords | GCMD (Global Change Master Directory) keywords relative to the present product |
gemet_keywords | GEMET (GEneral Multilingual Environmental Thesaurus) keywords relative to the present product |
iso19115_topic_categories | ISO 19115 keywords relative to the present product |
cdm_data_type | vector |
comment | Miscellaneous information about the data |
date_created | The date on which the data was generated |
creator_name | LEGOS, CLS, CNES |
creator_url | |
creator_email | inlandwaterlevel@groupcls.com |
project | Copernicus Climate Change Service - C3S |
geospatial_lat_min | Decimal degrees north |
geospatial_lat_max | Decimal degrees north |
geospatial_lon_min | Decimal degrees east |
geospatial_lon_max | Decimal degrees east |
time_coverage_start | YYYY-MM-DD |
time_coverage_end | YYYY-MM-DD |
time_coverage_duration | ISO8601 duration string |
standard_name_vocabulary | NetCDF Climate and Forecast (CF) |
license | Data access and distribution |
platform | Satellite name |
sensor | Sensor name |
key_variables | Comma separate list of the key variables in the file |
processing_mode | Delayed Time |
processing_level | LEVEL3B |
Copyright | Copernicus Climate Change Service – C3S |
inspire_theme | Hydrography |
Credit | Lake and River Water Level products are generated by Theia-land program supported by CNES ( http://hydroweb.theia-land.fr). Cretaux J-F., Jelinski W., Calmant S., et al., 2011. SOLS: A lake database to monitor in the Near Real Time water level and storage variations from remote sensing data, Advances in space Research, 47, 1497-1507 |
The data is provided as a variable in the netCDF file as indicated in Table 3.
Table 3: Variables in the Lake Water Level product
time | Days since 1950-01-01 |
water_surface_height_above_reference_datum | water Surface height above surface of reference in meters |
water_surface_height_uncertainty | standard deviation from height of high frequency measurements used in the estimation |
Product characteristics
Projection and grid information
Longitude and Latitude values indicating the lake barycentre are expressed with respect to the coordinate system WGS84. Concerning the reference surface (Datum), for the timeseries using past missions (Topex/Poseidon, Jason1, ERS1, ERS2, Envisat), the water level was estimated with respect to the high-resolution global gravitational model GGM02C, released in 2004. For stability reasons, this reference is maintained when data for current missions (Jason 3, Sentinel-3A, Sentinel -3B) is used to extent these timeseries. However, for the timeseries starting after Jason2 mission in 2008, including Jason3 and Sentinel-3A/3B, the reference surface used is the Earth Gravitational Model 2008 (EGM2008). Data from this model, more up to date, is included in the agency's products.
It's very important to notice here, that for the lake water level, the C3S product contains the mean value of the water height over the lake and the variation of this level over the time has priority over an absolute value.
The reference surface (Datum) for the current data set is GGM02C (high-resolution global gravitational model).
Spatial information
As described above, the water level heights can only be measured right below the satellite's orbital ground tracks (nadir), if favourable measurement conditions are encountered. Unfortunately, it is not always possible to retrieve valuable measurements at each crossing between hydrographic networks and satellite ground tracks.
In addition, given the inclination of the orbit of some altimetry missions, such as Jason-3 that is limited to +/- 66° in latitude, some targets may not be flown over in the Arctic region.
The ground track coverage of all altimetry missions can be found on AVISO website: http://www.aviso.altimetry.fr/en/data/tools/pass-locator.html.
As an example, the offset between two ground tracks at the Equator is 315 km for Jason-3 tracks (10-days revisit time) and 80 km for Envisat (35-days revisit time).
Temporal information
Lake Water Level series are updated each time a given satellite crosses the lake or if several satellites fly over the lake. So, the update time does not depend on the duration of one orbital cycle (10 days for Jason-3, 35 days for Envisat, 28 days for Sentinel-3A/3B) it may be more frequent.
Target requirements
The target requirements and the gap with the current product characteristics are described in the Target Requirement and Gap Analysis Document [D5]. Table 4 shows target requirements. See the PQAR [D4] for an assessment of product quality.
Table 4: User requirements
Property | Target | Threshold |
Spatial Coverage | Global | Global |
Spatial Resolution | Lake area: 1 km2 | Lakes area: 1000 km2 |
Temporal Coverage | More than 25 years | 10 years |
Temporal Resolution | Daily | 1-10 days |
Standard uncertainty | 3 cm for large lakes, 10 cm for the remainder | 15 cm |
Stability | 1cm/decade | - |
Data usage information
File naming
Files are named:
C3S_LWL_<lakename>altimetry<version><time_coverage_start><time_coverage_end>_R<file_generation_date>.nc
Where
<lakename> is the name of the lake, in English or the Local name
<version> is the product version number.
File format
The actual format is netCDF-4 Classic Model.
Quicklook
For the lake Zhari-namco, in China, the time series has been computed since January 1993, starting with Topex/Poseidon until to 2002, Jason-1 between 2002 and 2008, Jason-2 between 2008 and 2016, and Jason-3 and Sentinel-3B since October 2016. Table 5 illustrates the content of the header for the netCDF file and Figure 3 the corresponding time series.
Table 5: Header of the netCDF file for the lake Zhari-namco
netcdf C3S_LWL_ZHARI-NAMCO_altimetry_V3.0_19930113_20210119_R20210120 { dimensions: time = UNLIMITED ; // (613 currently) variables: double time(time) ; time:standard_name = "time" ; time:long_name = "time" ; time:units = "days since 1950-01-01" ; time:calendar = "gregorian" ; double water_surface_height_above_reference_datum(time) ; water_surface_height_above_reference_datum:_FillValue = -32767. ; water_surface_height_above_reference_datum:standard_name = "water_surface_height_above_reference_datum" ; water_surface_height_above_reference_datum:long_name = "water surface height above geoid" ; water_surface_height_above_reference_datum:units = "m" ; water_surface_height_above_reference_datum:unit_long = "Meters" ; watrer_surface_height_above_reference_datnum:comment = "The reference surface (Datum) for the current data set is GGM02C (high-resolution global gravitational model). " double water_surface_height_uncertainty(time) ; water_surface_height_uncertainty:_FillValue = -32767. ; water_surface_height_uncertainty:standard_name = "water_surface_height_uncertainty" ; water_surface_height_uncertainty:long_name = "water surface height uncertainty" ; water_surface_height_uncertainty:units = "m" ; water_surface_height_uncertainty:unit_long = "Meters" ; // global attributes: :title = "Lake Water Level from satellite altimetry" ; :lake = "Zhari-Namco" ; :basin = "Mongolian" ; :country = "China" ; :lake_barycentre_latitude = "30.9 N" ; :lake_barycentre_longitude = "85.6 E" ; ;surface_reference = GGM02C (high-resolution Global Gravitational Model) :institution = "LEGOS, CLS, CNES" ; :source = "Copernicus Climate Change Service" ; :history = "Created on 2021-01-21" ; :references = "https://cds.climate.copernicus.eu/" ; :conventions = "CF-1.7" ; :product_version = "3.0" ; :summary = "Copernicus Lake Water Level database contains time series over water levels of large lakes around the world" ; :id = "C3S_LWL_ZHARI-NAMCO_altimetry_V3.0_19930113_20210119_R20210120.nc" ; :keywords_vocabulary = "GCMD, GEMET and is19115 keywords" ; :gcmd_keywords = "TERRESTRIAL HYDROSPHERE, SURFACE WATER" ; :gemet_keywords = "water level, water management, hydrometry, hydrology, climate, seasonal variation, environmental data, environmental monitoring, monitoring, remote sensing" ; :iso19115_topic_categories = "elevation,inlandWaters" ; :cdm_data_type = "vector" ; :comment = "" ; :creator_name = "LEGOS, CLS, CNES" ; :creator_url = "http://www.legos.obs-mip.fr, https://www.cls.fr/, https://cnes.fr/" ; :creator_email = "inlandwaterlevel@groupcls.com" ; :project = "Copernicus Climate Change Service - C3S" ; :geospatial_lat_min = -90. ; :geospatial_lat_max = 90. ; :geospatial_lon_min = -180. ; :geospatial_lon_max = 180. ; :time_coverage_start = "1993-01-13" ; :time_coverage_end = "2021-01-19" ; :time_coverage_duration = "P28Y4M26D" ; :standard_name_vocabulary = "NetCDF Climate and Forecast (CF)" ; :license = "Free after registration" ; :platform = "Topex/Poseidon, Jason-1, Jason-2, Jason-3, Envisat, GFO, SARAL, Sentinel-3A, Sentinel-3B" ; :sensor = "Poseidon Ku, Poseidon 2Ku, Poseidon 3Ku, Poseidon 3Bku, RA-2, RA Ku, Altika Ka, SRAL Ku" ; :key_variables = "water_surface_height_above_reference_datum, water_surface_height_uncertainty" ; :processing_mode = "Delayed Time" ; :processing_level = "LEVEL3B" ; :copyright = "Copernicus Climate Change Service" ; :inspire_theme = "Hydrography" ; :credit = "Lake Water Level products are generated by the Copernicus Climate Change Services, the Earth Observation Programme of the European Commission and the Theia-land program supported by CNES. The research leading to the current version of the product has received funding from CNES, LEGOS, IRD, CLS and ESA." ;
Figure 3: Water level series for the lake Zhari-namco
Appendix A - Specifications
The following table contains the User Requirements for Water Level as described in GCOS:
Content of the dataset | |
Content of the main file | The data file shall contain the following information on separate layers:
|
Spatial and temporal features | |
Spatial coverage | The product shall be distributed globally based on a harmonized identification of the products. The area of the lakes must be at least 1kmx1km |
Temporal coverage | Times series of 10 years minimum are required |
Temporal resolution |
|
Data uncertainties | |
Threshold | 15 cm |
Target | 3cm for large lakes, 10 cm for the remainder |
Format requirements | |
Format | NetCDF, CF Convention |
References
SOLS: A Lake database to monitor in Near Real Time water level and storage variations from remote sensing data, J. Adv. Space Res. Vol 47, 9, 1497-1507, doi:10.1016/j.asr.2011.01.004, 2011 Crétaux J-F , W. Jelinski , S. Calmant , A. Kouraev , V. Vuglinski , M. Bergé Nguyen , M-C. Gennero, F. Nino, R. Abarca Del Rio , A. Cazenave , P. Maisongrande
Lake studies from satellite altimetry, C R Geoscience, Vol 338, 14-15, 1098-1112, doi: 10.1016/j.crte.2006.08.002, 2006 Crétaux J-F and C. Birkett
A new Global Lakes Database for a remote sensing programme studying climatically sensitive lakes, J. Great Lakes Res., 21 (3), 307-318. Birkett, C.M., Mason, I.M., 1995.
Lake volume monitoring from space. Surveys in Geophysics, 37(2), 269-305. Crétaux, J. F., Abarca-del-Río, R., Berge-Nguyen, M., Arsen, A., Drolon, V., Clos, G., & Maisongrande, P., 2016.
C. Birkett 1995. The global remote sensing of lakes, wetlands and rivers fro hydrological and climate research. International Geoscience and Remote Sensing Symposium, IGARSS'95. Quantitative Remote Sensing for Science and Applications.
C. Birkett 1998. Contribution of the TOPEX NASA Radar Altimeter to the global monitoring of large rivers and wetlands. AGU1000. Doi: https://doi.org/10.1029/98WR00124
F. Mercier et Al. 2002. Interannual lake level fluctuations (1993-1999) in Africa from Topex/Poseidon: connections with ocean-atmosphere interactions over the Indian Ocean. Global and Planetary Change.