ATSR Long Term Stability Project

Minimize Objectives

The 'Long Term Stability of the ATSR Instrument Series: SWIR Calibration, Cloud Masking and SAA' project, commonly known as the ATSR Long Term Stability (or ALTS) project is designed to explore new and innovative ways of enhancing the (A)ATSR data set, making use of the complete time-series from ATSR-1, ATSR-2 and AATSR and contributing to the preservation and improved understanding of this valuable data set within the frame of the Long Term Data Preservation (LTDP) programme.


The main areas of investigation in Phase 1 (July 2013 to June 2015) have been:

  • Exploring a new approach to the assessment of Short Wave Infra-Red (SWIR) channel calibration.
  • The development, implementation and validation of a new method for Total Column Water Vapour (TCWV) retrieval.
  • A study of the South Atlantic Anomaly (SAA) using ATSR data.
  • Radiative Transfer (RT) modelling in support of TCWV algorithm development.
  • Prototyping of a tool to provide AATSR observations with their location in the original instrument grid.

Topics for Phase 2 (July 2015 to June 2016) include:

  • The development of a strategy for the retrieval and archiving of historical ATSR mission documentation.
  • An extension of the ATSR calibration analysis work.
  • An extension of the work on TCWV to include retrievals over land.
  • The development of potential new methods for cloud masking.
  • Continuation of the development on the AATSR GBT-UBT-Tool and an extension to include the analysis of issues associated with the Sea and Land Surface Temperature Radiometer (SLSTR) on the Sentinel-3 platform.

This work is being undertaken under ESA Contract No. 4000108531/13/I-NB

Minimize ATSR Document Retrieval



The curation of mission documentation has gained significant interest from numerous user communities and space agencies in recent years.  This interest follows growing concerns that historic missions are now threatened by the permanent loss of invaluable information/knowledge that once proved fundamental to a missions' success and evolution (e.g. ATSR-1, ATSR-2 and AATSR).  


The ATSR Quality Working Group (QWG) has expressed their concerns for the early ATSR missions as they begin to face a number of obstacles (e.g. the retirement of key personnel, the scatter of documentation across numerous locations/sites, the physical decay of documentation and the reading of old digital formats) that could threaten future ATSR curation and exploitation activities. As a result, ESA are supporting this new work package within the ALTS project that will investigate appropriate ATSR documentation retrieval and archiving strategies.


It is hoped that the successful implementation of these strategies will not only allow for the effective curation of ATSR documentation as well as the enhancement of the existing (A)ATSR data set (even though two decades has passed since the ending of ATSR-1 operations) but also, pave the way for similar curation activities for other ESA missions in the future.


Further Reading


Curation of Sensor Information for Climate Data Records from Space


Additional Information


Please do not hesitate to get in touch should you have any queries or questions regarding this particular work package, or if you have any information that you believe would be of value to the project.  The ALTS team can be contacted at the following: or

Minimize TCWV Retrieval

Current Activities

A new atmospheric total column water vapour (TCWV) retrieval algorithm, named AIRWAVE (Advanced InfraRed Water Vapour Estimator) has been developed using the measurements from the two thermal infra-red (TIR) channels of the ATSR instrument series, namely the 10.8 and 12 µm channels. The use of the dual view capability of the ATSR instruments allows for the accurate and precise retrievals of cloud free TCWV over oceans for day-time and night-time illumination conditions.


AIRWAVE is based on the accurate knowledge of the instrument's physical characteristics, such as the spectral filter functions, in combination with advanced radiative transfer models and the sea surface spectral emissivity database produced by the University of Edinburgh. The retrievals, therefore, do not require algorithm tuning or adjustments to independent water vapour data sets. Additionally, AIRWAVE can be readily extended to the upcoming SLSTR instrument on the Sentinel-3 satellite series.


The atmospheric TCWV is retrieved at the spatial resolution of the ATSR Level 1 products, i.e. around 1 km2 for specific studies on restricted areas, and at a coarser resolution (0.25°x0.25°) for long term data analysis on global scale.


One of the main features of AIRWAVE is the enhanced sensitivity to ATSR radiometric calibration. This implies that any possible drift affecting the ATSR instruments will be significantly affecting the retrieved TCWV too. Far from being a drawback, this extreme sensitivity will be extremely useful for the assessment of the calibration of the ATSRs for the current and future calibration correction schemes.


Preliminary inter-comparisons with the SSM/I and ERA-Interim TCWV products have been carried out for a set of test cases. The initial results show that the AIRWAVE data at coarse resolutions have no bias and a precision of 4% for the ATSR-2 and AATSR instruments and 15% for the ATSR-1 instrument.

Future developments


The next phase of the ALTS project will include:

  • Validation of the entire ATSR data set using data from balloon soundings.
  • The refinement of retrieval parameters, including e.g. the possible seasonal and/or latitudinal dependence
  • The retrieval of TCWV for cloud-free land surfaces



Some examples of AIRWAVE products are shown in this section.

  • Figure 1 shows the TCWV global field averaged over the >20 years long data set, while Figure 2 shows the average standard deviation TCWV for the same time period.
  • Figure 3 shows the details that can be depicted from the analysis of the AIRWAVE TCWV at 1x1km2 resolution (night-time passage over Sardinia, April 14, 2008).
  •  Figure 4 shows the AIRWAVE-SSMI/ECMWF inter comparison results.

Figure 1: 1991-2012 average TCWV global field at coarse resolution



Figure 2: 1991-2012 average σTCWV (standard deviation) global field at coarse resolution



Figure 3: Example of TCWV retrieval at 1x1km2 resolution (2008-04-14, night-time)


Figure 4: AIRWAVE-SSMI/ECMWF inter-comparison (preliminary) results. Upper plots: ATSR-1; middle plots ATSR-2; lower plots: AATSR.


Minimize RT Modelling

Current Activities

A dedicated Radiative transfer Model (RTM) Forward Model (FM) has been developed to reproduce the ATSR-1, ATSR-2 and AATSR responses, and validated against RTTOVv11 and used to generate a preliminary set of coefficients for the TCWV retrieval. These coefficients were used to produce the Advanced Infra-Red Water Vapour Estimator (AIRWAVE) v1 data set.


Within the ALTS project, the RTM was used for:

  • Simulation of ATSR radiances under different atmospheric conditions:
    To understand which species influence ATSR measured radiances and which atmospheric parameters should be taken into account in the TCWV retrieval. Sensitivity function calculations have been also performed in order to identify the altitude levels that mainly contribute to TCWV retrieved by AIRWAVE.
  • End-to-End simulation of TCWV retrieval using ATSR synthetic radiances:
    To validate the used procedure and to quantify errors due to the used assumptions.
  • Computation of theoretical parameters for H2O retrieval:
    Computation of all the parameters needed by the AIRWAVE algorithm. A new parametrisation scheme has been developed. 
  • Study of the impact of aerosol of different types on ATSR radiances (for H2O retrieval):
    Study to analyse possible biases on retrieved TCWV introduced by aerosol presence.
  • Creation of final look up tables for H2O retrieval processing chain:
    A new set of parameters were calculated. these parameters also account for different latitudes and seasons. We also calculate preliminary parameters for the SLSTR instrument on board the Sentinel-3 satellite series.

Future Developments

  •  Implementation of new parameters and generation of AIRWAVE v2 data set
  •  Feasibility tests for the extension of the TCWV retrievals to land surfaces
  •  Analysis for the development of a new cloud screening using the (A)ATSR Thermal  Infrared channels and the dual view (feasibility)
Minimize Calibration Analysis

Current Activities

The primary objective was to understand the radiometric calibration of the 1.6 micrometre channel of the (A)ATSR sensor series and develop the means to determine drift in the on-board calibration system using data driven methods. A secondary objective within the initial phase of the study was also to understand the radiometric calibration of the other reflectance channels using the same data driven techniques (figure 1).


Figure 1: Shows the new method results (blue) compared to the traditional methods (orange)


As part of the study, a full literature review covering the instruments and their calibration and exceptions review (documents that should be available but are not) will be provided.


Within the scope of the project, EOSense will examine the corrections required to convert data from a Level 0 to Level 1 product in radiometric terms (without a geometric correction) and develop methods to estimate the non-linearity in the Level 1 data products.

Future Objectives 

  • To extend the radiometric calibration of the 1.6 micrometre band to cover both ATSR-2 and ATSR-1, providing a complete calibration history across the entire series, developing modified algorithms of the data driven approach for ATSR-1 which does not have an on-board calibrator.
  • To extend the radiometric calibration of the other reflectance channels to cover ATSR-2.
  • To extend the methodology to cover the thermal channels for all three sensors in the series and provide an unbroken chain in thermal calibration.
  • To develop a methodology to allow monthly estimates of calibration drift, rather than one based on annual averages.
  • To investigate the use of a novel algorithm for assessing relative calibration in providing information on Level 0 non-linearity. This may be compared to non-linearity models developed for the 1.6 micrometre band.
  • To develop a full understanding of the low level artefacts present in the imagery and their causes.
Minimize GBT-UBT Tool

Current Activities


The overarching objective is to design and implement a prototype software tool to reconstruct (A)ATSR measurement point geolocation prior to a gridding (quantisation) process. Further requirements for additional software functionality continue to evolve during the activity. Where possible, these requirements are translated into additional features to complement the primary objective.


The motivation of the activity relates to the perceived need amongst ATSR data users to work with measurement geolocations in their native position for a variety of applications. The primary objective is enabled through a Technical Note "Instrument Pixel Co-ordinates and Measurement Times from AATSR Products" written in 2010 by A. Birks (RAL), available at The Technical Note establishes the requisite methodology for "un-gridding" measurement geolocations. This process is a transform from Gridded Brightness Temperatures (GBT) to Un-gridded Brightness Temperatures (UBT) for Level 1B products in Envisat format, hence GBT-UBT.


The tool has undergone a validation campaign and is scheduled for use within the wider ALTS team. In addition, the source code for the tool is released to the public as open source (GPL v3) and is available (for free) via GitHub ( Pre-compiled binary executables (Java) are also available at the above URL. Ongoing external work will integrate the tool into an upcoming version of the Sentinel-3 Toolbox.


A full description covering the tool features, usage and output is also available via GitHub at the above URL (see A short list of features of the tool are as follows:

  • Parses ATSR-1, ATSR-2 and AATSR TOA 1P (Level 1B) files.
  • Computes UBT geolocations for measurement pixels.
  • Orthorectifies geolocations taking into account an external Digital Elevation Model.
  • Computes the acquisition time of each measurement
  • Computes the extent/size of AATSR pixel Field Of Views projected onto a spherical Earth surface in along and across track directions.
  • Outputs array datasets in either netCDF4-CF or HDF5 format files.
  • Computations are multi-threaded to take advantage of multiple CPU environments.
  • Is written in Java to facilitate simple deployment on various Operating Systems.
  • Is licensed under the terms of the GNU General Public License (GPL3), which implies that it is completely free, including the source code.

Future Objectives


The primary tasks remaining in this activity includes:

  • Investigate overall AATSR geolocation accuracy using 6.05 dataset gas-flare detection database referenced to North Sea Oil Platform locations.
  • Investigate, design and prototype SLSTR format GBT-UBT transform.
  • Support Sentinel-3 Toolbox wrapper development.
Minimize Schedule

Phase 1 - July 2013 to June 2015

Phase 2 -  July 2015 to June 2016