MODIS Land Validation: Background Information

The information below provides an overview of the validation strategy being implemented by the MODLAND team and a description of the three stages of a "Validation Hierarchy".

(return to MODLAND validation page)

(go to MODLAND page)

Background information: Overall Approach

MODLAND product quality will be ensured by Calibration, Quality Assurance (QA) and Validation.

MODLAND will use several validation techniques to develop uncertainty information on its products. The methods include comparisons with in situ data collected over a distributed set of validation test sites, comparisons with data and products from other airborne and spaceborne sensors (e.g., SeaWiFS, AVHRR, MISR, TM/ETM+, ASTER), inter-comparison of trends derived from independently obtained reference data and MODLAND products, and analysis of process model results (including EOS IDS models) which are driven or constrained by MODLAND products.

Our primary validation techniques include collection of and comparison with field and aircraft data, and comparison with data and products from other satellites. Field data collection will include instantaneous measures of spectral reflectances and TIR radiance at various test sites, first as a calibration activity, and second, as validation of our radiometric variables. Second will be establishment of a semi-permanent array of test sites, usually including a flux tower, for extended temporal measurement of terrestrial biophysical dynamics over a range of landcover types. These field data will be archived in cooperation with the Oak Ridge DAAC. Results of all validation activities will be conveyed to the end-user through both published literature and MODLAND and EOS web pages.

To adequately cover the broad range of surface-atmosphere systems that will be encountered around the world, multiple validation methods applicable to different temporal and spatial scales will be implemented. Nevertheless, the highly discretized nature of land systems, in contrast to most atmospheric and ocean systems, makes in situ measurement of coarse resolution parameters extremely difficult. Essentially, this increases the uncertainty of the validation data itself. These concerns further necessitate a validation approach that is comprehensive and at times redundant.

The work will involve

Summarizing validation work within a three-stage validation hierarchy.
commitment to the EOS land validation core sites
product-specific activity and validation protocols (primarily by MODLAND PIs),
close cooperation with EOS validation investigators to meet specific product validation needs,
interaction with established data networks (e.g. FLUXNET, AERONET),
participation in community field campaigns (LBA, SAFARI 2000, GCIP),
developing novel validation instrumentation and approaches (e.g. MQUALS, CIMEL with BRDF),
and collaboration with the data providers (PIs, DAACs, ESIPs).

These activities are essential to MODLAND product validation and together provide the foundation for operational product validation. Details of product specific validation plans may be obtained from individual team members.

Validation Hierarchy

Stage 1 Validation: Product accuracy has been estimated using a small number of independent measurements obtained from selected locations and time periods and ground-truth/field program effort.
Stage 2 Validation: Product accuracy has been assessed over a widely distributed set of locations and time periods via several ground-truth and validation efforts.
Stage 3 Validation: Product accuracy has been assessed and the uncertainties in the product well established via independent measurements in a systematic and statistically robust way representing global conditions.

Background information: Measures of Success

Successful validation will have been accomplished if timely and accurate product uncertainty information becomes routinely available to the product users within two years after launch. During this period, MODLAND will adjust algorithms as necessary to provide better performance and consistency with validation data. Previously derived products will be reprocessed in a timely manner following changes to the operational algorithms.

Background information:

Validation Update for Terra and Aqua,
released December, 2000, available as pdf.

The MODLAND Validation plan is available as a pdf.

The validation plan describes

coordination with EOS Validation Investigators
validation sites
pre-launch activities
post-launch activities
implementation of validation results

Background information: Additional References

Additional references for MODIS validation include:
ORNL DAAC's "Best Practices for Preparing Ecological and Ground-Based Data Sets to Share and Archive", published in October, 2000.
"EOS Land Validation Coordination: An Update" by Chris Justice, Dave Starr, Diane Wickland, Jeff Privette, and Tim Suttles
"EOS Global Land Validation Network", Jeffrey L. Privette, Jeffrey T. Morisette, Chris Justice, and David Starr, IGARSS'99 Proceedings.

Go to MODIS Land Team page...

Page designed and maintained by Dr. Jeffrey T. Morisette, with assistance from Michael Hohner, MAST, and John Owens, UMCP

phone: (301) 614-6676

e-mail jeff.morisette@gsfc.nasa.gov
Responsible NASA official: Dr. Jeffrey T. Morisette, Code 923, Goddard Space Flight Center

Security and Privacy Policy