SMEX02 Landsat 5 and 7 Thematic Mapper Land Surface Brightness Temperatures

Summary

This data set consists of land surface brightness temperatures (T_Bs) derived from Landsat 5 Thematic Mapper (TM) and Landsat 7 Enhanced Thematic Mapper+ (ETM+) thermal bands. The Landsat data coincided with the SMEX02 campaign and were acquired once per day on 23 June, 1 July, 8 July, 16 July, and 17 July 2002 over the Walnut Creek watershed, Iowa, USA. The data collectively cover 41.69° N to 42.73° N and 93.16° W to 93.84° N. Landsat 7 ETM+ data have 60 m spatial resolution. Landsat 5 TM data have 30 m spatial resolution. The investigators used the MODTRAN 4.1 radiative transfer model to correct the original Level-1G radiances for atmospheric effects; then they converted radiances to T_Bs. Data are unsigned integer, band-interleaved-by-line (BIL), binary values, scaled by 100. Data are available via FTP.

These data were collected as part of a validation study for the Advanced Microwave Scanning Radiometer - Earth Observing System (AMSR-E). AMSR-E is a mission instrument launched aboard NASA's Aqua Satellite on 04 May 2002. AMSR-E validation studies linked to SMEX are designed to evaluate the accuracy of AMSR-E soil moisture data. Specific validation objectives include assessing and refining soil moisture algorithm performance; verifying soil moisture estimation accuracy; investigating the effects of vegetation, surface temperature, topography, and soil texture on soil moisture accuracy; and determining the regions that are useful for AMSR-E soil moisture measurements.

Citing These Data

Jackson, T., F. Li, and W. Kustas. 2005. SMEX02 Landsat 5 and 7 Thematic Mapper land surface brightness temperatures. Boulder, CO, USA: National Snow and Ice Data Center. Digital media.

Overview Table

Category	Description
Data format	Data are unsigned integer, band-interleaved-by-line (BIL), binary values, scaled by 100. Pixel values of "0" indicate missing data. Data are in a PC byte order (little endian). SGI and Sun users should byte-swap these data before using them.
Spatial coverage and resolution	Data collectively cover 41.69° N to 42.73° N and 93.16° W to 93.84° N (Walnut Creek watershed, Iowa, USA). Landsat 7 ETM+ data have 60 m pixel resolution. Landsat 5 TM data have 30 m pixel resolution.
Temporal coverage and resolution	Landsat data were acquired once per day on 23 June, 1 July, 8 July, 16 July, and 17 July 2002.
Tools for accessing data	Users can read these data with any remote sensing or image processing software.
File naming convention	File names are "ddmmyy_btemp.bil", where "dd" is the two-digit day, "mm" is the two-digit month, "yy" is the two-digit year, and "btemp" is brightness temperature (TB).
File size	File sizes range from 3.33 MB to 13.85 MB.
Parameter(s)	Data consist of brightness temperatures (TBs) in Kelvins.
Procedures for obtaining data	Data are available via FTP.

Table of Contents

1. Contacts and Acknowledgments
2. Detailed Data Description
3. Data Access and Tools
4. Data Acquisition and Processing
5. References and Related Publications
6. Document Information

1. Contacts and Acknowledgments

Investigator(s) Name and Title

Tom Jackson, Fuqin Li, and William Kustas
United States Department of Agriculture
Agricultural Research Service
Hydrology and Remote Sensing Laboratory
Beltsville, MD 20705-0000

Technical Contact

Acknowledgements

This research was supported by the NASA Earth Observing System (EOS) Aqua AMSR, Terrestrial Hydrology, and Global Water and Energy Cycle programs.

2. Detailed Data Description

Land surface temperatures (T_s) and brightness temperatures (T_Bs) are key boundary conditions in many remote sensing-based land surface modeling schemes. Currently available satellite thermal infrared sensors provide data with different spatial resolutions and temporal coverage to estimate land surface temperature. The Geostationary Operational Environmental Satellite (GOES) has 4 km spatial resolution in the thermal infrared channel, while the NOAA-Advanced Very High Resolution Radiometer (AVHRR) and the NASA Terra- and Aqua-Moderate Resolution Imaging Spectroradiometer (MODIS) sensors have 1 km spatial resolution. The Terra-Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) provides significantly higher spatial resolution at 90 m. The Landsat 5 TM has 30 m resolution, and Landsat 7 ETM+, which has 60 m resolution. However, these instruments have a repeat cycle of 16 days. Some T_s standard products are available from NOAA AVHRR and MODIS data with a 1 km resolution. These low-resolution data are limited in some applications.

As part of the SMEX02 campaign, two Landsat 5 TM scenes and three Landsat 7 ETM+ scenes were acquired during the primary study period. These data were used to produce high-resolution (30 m to 60 m) T_B data sets. Users should be aware that these satellite data represent a single point in time, while ground-based estimates were taken before, during, and after the SMEX02 campaign.

The following table summarizes characteristics of the five images in this data set.

Image Specifications

File name	062302_btemp.bil	070102_btemp.bil	070802_btemp.bil	071602_btemp.bil	071702_btemp.bil
Date	2002-06-23	2002-07-01	2002-07-08	2002-07-16	2002-07-17
Time (Fractional hours in GMT)	16.4943	16.7013	16.8083	16.5887	16.694
Landsat No.	5	7	7	5	7
Path	26	26	27	27	26
Row	31	31	31	31	31
Upper-left corner (UTM, meters)	431085.000 E 4731115.000 N	431064.092 E 4731115.913 N	431085.000 E 4731115.000 N	431085.000 E 4731115.000 N	431097.084 E 4731095.389 N
Upper-right corner (UTM, meters)	486616.000 E 4731115.000 N	486624.092 E 4731115.913 N	484450.871 E 4731102.466 N	486615.000 E 4731115.000 N	486657.084 E 4731095.389 N
Lower-left corner (UTM, meters)	431085.000 E 4616185.000 N	431064.092 E 4616155.913 N	431050.871 E 4616142.466 N	431085.000 E 4616185.000 N	431097.084 E 4616195.389 N
Lower-right corner (UTM, meters)	486615.000 E 4616185.000 N	486624.092 E 4616155.913 N	484450.871 E 4616142.466 N	486615.000 E 4616185.000 N	486657.084 E 4616195.389 N
Pixel size (meters)	30	60	60	30	60
Rows	3831	1916	1916	3831	1915
Columns	1851	926	890	1851	926

Format

Brightness temperature data are unsigned integer, band-interleaved-by-line (BIL), binary values, scaled by 100. Pixel values of "0" indicate missing data. Data are in a PC byte order (little endian). SGI and Sun users should byte-swap these data before using them.

File and Directory Structure

The FTP site contains five files. See the Image Specifications table above for characteristics of each file.

File Naming Convention

File names are "ddmmyy_btemp.bil", where "dd" is the two-digit day, "mm" is the two-digit month, "yy" is the two-digit year, and "btemp" is T_B.

File Size

062002_btemp.bil: 13.850 MB
070102_btemp.bil: 3.466 MB
070802_btemp.bil: 3.331 MB
071602_btemp.bil: 13.850 MB
071702_btemp.bil: 3.464 MB

Spatial Coverage

The Landsat data coincided with the SMEX02 experiment and were acquired over the Walnut Creek watershed in the Iowa study region. The data collectively cover the following area:

Southernmost Latitude: 41.69° N
Northernmost Latitude: 42.73° N
Westernmost Longitude: 93.84° W
Easternmost Longitude: 93.16° W

Spatial Resolution

Landsat ETM+ data have 60 m pixel resolution. The original Landsat 5 TM Level-1G data from the United States Geological Survey (USGS) have a 120 m footprint, but the USGS resampled the data to 30 m spatial resolution.

Projection

These Landsat data are in a Universal Transverse Mercator (UTM) Zone 15 projection and North America (NAD83) datum.

Temporal Coverage

The Landsat scenes in this data set were acquired on 23 June, 1 July, 8 July, 16 July, and 17 July 2002.

Parameter or Variable

Parameter Description

Pixel values represent T_Bs in Kelvins, scaled by 100.

Parameter Range

Data values range from 23044 (230.44 K) to 33553 (335.33 K).

Error Sources

Converting radiances to T_s and T_Bs involves a number of assumptions and approximations. Sensor properties represent one source of error. The calibration error is within ±0.6 K for ETM+ data obtained after December 2000, while the calibration error from TM is difficult to quantify; past studies indicate it is quite significant and requires recalibration with in situ data (Li et al. 2004).

A second source of error arises from the accuracy of water vapor measurements while using the MODTRAN atmospheric correction model. Schmugge, Hook, and Call (1998) estimate this as nominally 10%. If the target T_B is 300 K, this could lead to a T_B error of about 0.5 K for ETM+ data (Li et al. 2004).

A third source of error comes from estimating surface emissivity. The emissivity of the soil is relatively high at the Iowa study site, and the site is also vegetated. The investigators assumed a general emissivity of about 0.98. The emissivity error for this site should be less than 0.005, which could lead to a change of 0.2 K (ETM+ data) in the T_s when the target T_B is 300 K (Li et al. 2004).

The estimated overall accuracy of the T_Bs is approximately 1° K when comparing tower and satellite measurements. The TM T_Bs were recalibrated using low-altitude aircraft measurements (Li et al. 2004). Analysis revealed that the post-calibrated T_Bs from ETM+ still have a 0.98° C bias when compared to tower measurements. The T_Bs from TM have a bias of 1.47° C when compared to tower measurements (Li et al. 2004).

3. Data Access and Tools

Data Access

Data are available via FTP.

Volume

The entire data set is 37.1 MB.

Software and Tools

Users can read these data with any remote sensing or image processing software.

Related Data Collections

See Soil Moisture Experiment (SMEX) for more information.

4. Data Acquisition and Processing

Sensor or Instrument Description

TM is a multispectral scanning radiometer carried on Landsats 4 and 5. The TM has seven spectral bands, with a spatial resolution of 30 m for most bands.

ETM+, an improved version of TM, is carried on Landsat 7. The ETM+ has eight spectral bands with a spatial resolution of 60 m for most bands.

Data Source

The original TM and ETM+ radiance data from the USGS were Level-1G products, partly georegistered and radiometrically corrected. However, the data were not corrected for atmospheric effects.

Derivation Techniques and Algorithms

Atmospheric Correction

Radiance from a satellite platform is strongly affected by the presence of the atmosphere. So, atmospheric correction is needed to convert satellite-based radiance to an estimated surface T_B. The MODTRAN 4.1 radiative transfer model (Berk et al. 1998) was used to correct for atmospheric effects. See Li et al. (2004) for details of how the investigators derived land surface T_Bs and surface temperature data during SMEX02.

The sensor radiance (I_λ) is expressed as follows (Schmugge, Hook, and Col 1998):

I_λ = t_λI_λ(0) + d_λ

Where,

t_λ is atmospheric transmittance
I_λ(0) is the surface-leaving radiance
d_λ is the spectral radiance added by the atmosphere

A T_B is simply the temperature corresponding to a blackbody radiator emitting the same radiance. Radiance data can be converted into equivalent T_Bs as the following equation shows. The surface-leaving radiance, I_λ(0), can be expressed in terms of surface temperature in the following manner:

I_λ(0) = I_λ^B(T_B) = ε_λI_λ^B(T_s) + (1 - ε_λ)I_dλ

Where,

ε_λ is the wavelength-dependent surface emmisivity
I_λ^B(T_s) is the spectral radiance from a blackbody at surface temperature T_s
I_dλ is the downwelling sky radiance due to the atmosphere

Using MODTRAN, t_λ, d_λ, and I_dλ can be obtained from Landsat band response functions and radiosonde data.

Brightness Temperature Calculation

For Landsat thermal data, the investigators computed surface T_Bs in Kelvins for each pixel using the following equation:

T_B = (k₂ / ln[k₁ / I_λ(0) + 1])

Where,

I_λ(0) is the integrated band radiance (Wm^-2 sr^-1 µm^-1) from the first atmospheric correction equation above

k₁ and k₂ are calibration constants chosen to optimize the approximation for the band pass of the sensor. For Landsat 7, k₁ is 666.09 Wm^-2 sr^-1 µm^-1 and k₂ is 1282.71 K (GSFC 2005). For Landsat 5, k₁ is 607.76 Wm^-2 sr^-1 µm^-1 and k₂ is 1260.56 K (Schneider and Mauser 1996). When emissivity is known, all of the above equations can be used to estimate T_s.

5. References and Related Publications

Berk A., L.S. Bernstein, G.P. Anderson, P.K. Acharya, D.C. Robertson, J.H. Chetwynd, and S.M. Adler-Golden. 1998. MODTRAN cloud and multiple scattering upgrades with application to AVIRIS. Remote Sensing of the Environment 65: 367-375.

Li, F., T.J. Jackson, W.P. Kustas, T.J. Schmugge, A.N. French, M. Cosh, and R. Bindlish. 2004. Deriving land surface temperature from Landsat 5 and 7 during SMEX02/SMACEX. Remote Sensing of the Environment 92: 521-534.

NASA Goddard Space Flight Center (GSFC). "Landsat 7 Science Data Users Handbook." Updated 24 June 2005.
http://ltpwww.gsfc.nasa.gov/IAS/handbook/handbook_toc.html. Accessed August 2005.

Schmugge, T., S.J. Hook, and C. Coll. 1998. Recovering surface temperature and emissivity from thermal infrared multispectral data. Remote Sensing of the Environment 65: 121-131.

Schneider, K., and W. Mauser. 1996. Processing and accuracy of Landsat Thematic Mapper data for lake surface temperature measurement. International Journal of Remote Sensing 17: 2027-2041.

6. Document Information

Acronyms and Abbreviations

The following acronyms and abbreviations are used in this document.

AMSR	Advanced Microwave Scanning Radiometer
ASTER	Advanced Spaceborne Thermal Emission and Reflection Radiometer
AVHRR	Advanced Very High Resolution Radiometer
BIL	Band-Interleaved-by-Line
EOS	Earth Observing System
ETM+	Enhanced Thematic Mapper+
FTP	File Transfer Protocol
GMT	Greenwich Mean Time
GOES	Geostationary Operational Environmental Satellite
GSFC	Goddard Space Flight Center
NAD83	North American Datum of 1983
NOAA	National Oceanic and Atmospheric Administration
SMEX	Soil Moisture Experiment
TM	Thematic Mapper
USGS	United States Geological Survey
UTM	Universal Transverse Mercator

Document Creation Date

August 2005

Document Revision Date

August 2005

Document Review Date

August 2005

Document URL

http://nsidc.org/data/docs/daac/nsidc0241_smex_landsat_tbs.gd.html