THE FIRST NASA/LTER WORKSHOP ON SUNPHOTOMETRY/ATMOSPHERIC CORRECTION OF REMOTELY SENSED DATA
John R. Vande Castle - University of Washington
Eric F. Vermote - University of Maryland
This workshop was funded in part by NASA grant#NAGW-4565 and held at facilities of the Long-Term Ecological Research Network Office in the College of Forest Resources, University of Washington
1. ATMOSPHERIC EFFECTS AND REMOTE SENSING DATA
1.1 LOGISTICS, INTRODUCTION, OBJECTIVES, BACKGROUND
1.2 ATMOSPHERIC EFFECT ON REMOTE SENSING DATA: THEORY
1.3 WHAT ARE THE SUN PHOTOMETER DATA ?
1.4 ATMOSPHERIC EFFECTS ON REMOTE SENSING DATA: APPLICATIONS
1.5 OVERVIEW OF THE ATMOSPHERIC PROCESSING SOFTWARE FOR TM
2. DATA INTERFACE TO PHOTOMETER DATA
2.1 LTER SATELLITE AND SUN PHOTOMETER DATA ARCHIVE.
2.2 HOW TO ACCESS THE SUN PHOTOMETER DATA -
2.3 HOW TO EVALUATE THE QUALITY OF SUN PHOTOMETER DATA
2.4 PRESENTATION AND USE OF THE 6S CODE
2.5 FIRST TM TEST DATASET PROPOSAL
2.6 SITE PRESENTATION BY PARTICIPANTS:
3.1 TECHNICAL, LOGISTICAL ISSUES AND WORK PLAN
3.2 PRESENT AND FUTURES DATASETS
3.3 SUN PHOTOMETER NETWORK STATUS AND PLAN
3.4 LTER DISCUSSION FOR DATASET EVALUATION PLAN
4. BIBLIOGRAPHY
6. LTER/NASA WORKSHOP PARTICIPANTS
1.1 Logistics, Introduction, Objectives, Background
- John Vande Castle
Operational aspects were covered for this workshop at facilities of the Long-Term Ecological Research Network Office in the College of Forest Resources at the University of Washington. The primary objective of this workshop was to involve LTER scientists with use of sun photometer data for operational atmospheric correction of satellite remote sensing data. The main objective is to use automated sun photometer data to generate atmospheric corrections for satellite data, in particular for Landsat-Thematic Mapper (TM) data. As a secondary objective, operational approaches involving NOAA-Advance Very High Resolution Radiometer (AVHRR) satellite data, hand-held sun photometer data, and airborne datasets such as AVIRIS were also included. There was a focus on operational aspects of atmospheric data corrections. More specific research applications were not ignored for their importance of more general applications of the sun photometer data.
1.2 Atmospheric effect on remote sensing data: theory
- prepared by E. F. Vermote and J.C. Roger, presented by E. F. Vermote
A background presentation was given for a description of how the atmosphere influences the earth signal measured from space in the 0.4-2.4mic region of the solar spectrum. The focus was to present the formulation of the atmospheric effect as well as to describe the atmospheric parameters one needs to properly simulate the signal (pressure, aerosol concentration and type, water vapor, ozone). The general equation of the signal in the case of a lambertian, uniform target was presented and each of the terms in the equation analyzed in detail. The order of magnitude of the effect of the atmospheric effects on the signal was discussed in the case of Landsat-Thematic Mapper data.
Some more advanced modeling efforts like the influence of the BRDF of the surface and the heterogeneity of the target background were also discussed but in less detail. This presentation was an introduction to the 6S radiative transfer code (Second Simulation of the Satellite Signal in the Solar Spectrum, Vermote et al, 1995). This code is recommended for use by the LTER investigators in the framework of this project. The last version of this code has been successfully implemented at the LTERnet computing facility (see section 2.4)
1.3 What are the Sun Photometer Data ?
- Brent Holben
Data obtained by a network of automated sun photometer instruments was established in 1994 for some LTER sites after experimental acquisitions at Harvard Forest and Virginia Coast Reserve LTER sites in 1993. The LTER instruments are part of a larger effort to provide sun photometer information on a global scale. Background information leading to this work can be found in previous workshop reports on-line on the LTERnet.edu computer. Background information on the sunphotometer work can be found in the bibliography. These instruments provide data on atmospheric aerosol and water information that are needed for precise atmospheric correction of satellite data. The raw sun photometer data are transmitted via GOES satellite and processed at NASA/Goddard. The data are maintained on-line which is also mirrored on the LTER computer servers (LTERnet.edu, time.lternet.washington.edu). The raw data are processed by a specialized program written at NASA/Goddard called "demonstrat". This program is a key component of the sun photometer data since it organizes the data, and process them with accepted algorithms to provide a wide variety of information needed for various aspects of atmospheric correction of satellite data. As an example, the program provides a view of the time and location for all sun photometer data collected within the network. (figure 1). These data are critical for the atmospheric correction work in this effort since they provide the information on atmospheric aerosols needed for atmospheric correction of satellite data. Further information regarding this is given in section 2.3
1.4 Atmospheric effects on remote sensing data: applications
- prepared by E. F. Vermote and J.C. Roger, presented by E. F. Vermote
Eric Vermote covered operational implementation aspects of the atmospheric correction problem. For both TM and AVHRR sensors a description of the processing steps needed to get from raw data to useful physical quantities (i.e. the surface reflectance) was given. The different stages of processing were described, including navigation/geometric correction, cloud screening, absolute calibration, ozone correction using TOMS data, Rayleigh correction using Digital Elevation Models, stratospheric aerosol correction, water vapor database inversion and correction, and aerosol inversion and correction. The limitations of the atmospheric corrections in general, were also discussed.
During the presentation, the version 0 of the atmospheric correction software that processes TM (fast format) data was run remotely at GSFC on two TM full scenes. The corrected images were transferred to the LTER computers for evaluation on day 3.
1.5 Overview of the atmospheric processing software for TM
- Eric Vermote
Version 0 of the atmospheric correction software for TM data was described in more detail. The effects of other atmospheric constituents and their relation to atmospheric corrections was also presented. The changes envisioned for version 1 were discussed. These involve the inclusion of size distribution input, aerosol optical thickness (AOT) at 550nm, input and adjacency effect correction, and correction for elevation at 50km by 50km resolution.
2.1 Introduction to LTER satellite and sun photometer data archive.
- John Vande Castle
The sun photometer data for LTER work are transferred to the LTER computer servers automatically each day. The LTER Network Office also routinely archives these data for long-term storage. The sun photometer processing software has been ported to the LTER computer systems. The sun photometer data stored on the LTER computer systems is limited to data important to the LTER research. This reduces confusion with other datasets and permits the processing software to run much more quickly. LTER researchers have the advantage in working with the data from their personal LTER computer accounts, as well as access to other software maintained at the Network Office. Accessing the data and software are similar to access on the computer at NASA/Goddard as describe in the next section.
2.2 How to access the sun photometer data -
-Brent Holben
The sun photometer data are accessed though a specialize program named "demonstrat". This program was written specifically for processing the automated sun photometer, or similar photometer/radiometer data. A general user account for LTER investigators is available on a machine at NASA/Goddard, currently called "spamer.gsfc.nasa.gov". The demonstrat program can be accessed on the Internet and uses "X11" display capabilities for the graphical interface of the program. The "X11" access from local personal computers will vary with the local configuration, but the machine "spamer.gsfc.nasa.gov" is able to display to most remote "X11" displays. This whole process can be easily automated. An example of how to access the program in a non-automated procedure is as follows:
This is normally done with the command "xhost spamer.gsfc.nasa.gov", or by entering this machine name in the local xhost table.
This is normally done with the command "telnet spamer.gsfc.nasa.gov". A login prompt will be given, at which the general user account and password are entered.
This is normally done with the command (on spamer.gsfc.nasa.gov) "setenv DISPLAY computer:0.0", where "computer" is the name or Internet IP number of the local machine.
This is done with the command "demonstrat". When the command is entered, the opening display will start on the local machine. This initial display can take a while to start since all the data in the archive is scanned by default. There are other options which can be used, including entering the instrument number in the "demonstrat" command to access data for only one machine.
2.3 How to evaluate the quality of sun photometer data
-Brent Holben
The raw sun photometer data are processed and evaluated using the "demonstrat" software. The initial display (figure1) allows the user to select a given dataset for evaluation. A dataset is selected by scrolling through a "timeline" or by drawing a box around a selected region of the earth. Once a dataset is selected, it is processed by selecting the dataset name. A specific portion of the timeline can also be selected. The dataset selected will, by default, be displayed in a graphical window showing a plot of optical aerosol thickness vs. time for each channel of the instrument (figure2). Options in this window allow the user to filter the raw data, and remove data the user does not need. Individual segments of the data can be selected, as shown in an example of data from the Sevilleta LTER site (figure3). A single day can be selected (figure4) as well as other parameters including atmospheric water vapor (figure5). Information derived by processing the data through known algorithms, for instance estimated particle size distribution (figure6), can be selected as well. The data can be estimated through a "Bird model" (figure7) for any time of day, and these data can be fitted to estimates of Aerosol Optical Thickness (AOT) for sensors of the Cimel instrument as well as satellite sensor systems such as Landsat-TM (figure8).
2.4 Presentation and Use of the 6S code
- Eric Vermote
A working session was conducted on the use of the 6S for satellite signal simulation. The latest version of the code (3.2 alpha) has been successfully installed at the LTERnet computing facility. An example of a data file was distributed the working group. The code was run by the group and both data and results were extensively commented on by Eric Vermote. Finally, the working group was asked to modify the example data file and run the code again for TM band 2. The Graphical Interface (GUI) to the 6S program was run remotely at GSFC and presented to the working group. This GUI interface will be implemented later at LTERnet computing facility.
2.5 First TM test dataset proposal
-John Vande Castle
John Vande Castle presented the first TM dataset selected for the project. This dataset contains scenes already acquired by NASA as part of the Global Change data acquisition. Initial data for the project included those in conjunction with sun photometer measurements. For more data in 1994 and 1995, additional scenes in the archive as well as to be acquired will be added for the BNZ, SEV, AND, and NTL sites. AVHRR test datasets were also discussed for the MCM, NTL, BNZ sites.
2.6 Site presentation by participants:
2.6.1 MCM: McMurdo - Taylor Polar Dry Valley
Biome: Polar desert oases
Hydrologist, Gayle Dana studies glaciers at the MCM site, and conducts research in the evolution of the inter-annual variation in the melting process for global warming/cooling related study. She has used measurements of the amount of melting by various processes including the monitoring of the thickness of the ice sheet using poles. The site of interest is about 400km2 Remote sensing interest is in the visible and the thermal data, to assess the energy exchange at this LTER site. Gayle underlined the difficulty of getting remote sensing because of a lack of TM coverage (no receiving station right now), the cost of SPOT-HRV data, and the difficulty of getting AVHRR 1km data.
Automatic sun photometer measurements were acquired from January 10, 1995 to January 24, 1995 at the MCM site (-77°37';162°52') along with BRDF (4° azimuth, 7° view angle, several solar angles) and nadir measurement of different surface type spectra (Goetz instrument, 18° FOV, 0.4-2.5mic spectrum). Data collection will continue again in December of 1995. AVHRR data may be available from the McMurdo HRPT receiving station.
2.6.2 AND: H.J. Andrews Experimental Forest
Biome: Temperate coniferous forest.
Warren Cohen presented remote sensing research related to establish land cover mapping of the biome and change detection using high spatial resolution data, primarily Landsat-TM. In the broader perspective, this site is closely associated with the EOS Global Land Cover test site validation concept. In terms of remote sensing, this site is problematic because of the large topographic effects it exhibits.
2.6.3 SEV: Sevilleta National Wildlife Refuge
Biome: -semi-desert site
Greg Shore presented the research conducted on the Sevilleta site, where the study of hydrology and vegetation response to available precipitation is crucial for this arid region. The most important factor in this almost undisturbed site seems to be the ENSO phenomenon. Precipitation is actually monitoring by two indirect methods: (1) monitoring through analysis of lightning strike data, and correlation with response of vegetation through NDVI after precipitation. Some additional water availability measurements collected by Barbara Nolen are also used in the study. The studies uses primarily Landsat-TM data for calculation of NDVI. SEV needs images every month for efficient monitoring of the vegetation. An other important subject of research may be the study of natural fire in the area and the associated response of the ecosystem in terms of change in land-cover. A sun photometer has been permanently installed at the Sevilleta site since June, 1994.
2.6.4 NTL: North Temperate Lakes
Biome: North Temperate Forest/Lake/Agricultural/Urban
David W. Bolgrien and Randolph H. Wynne presented the activity at NTL/MADISON LTER sites. The two research areas there cover a wide range of studies related to atmosphere, fresh water pigment detection, remote sensing, atmospheric correction, and cloud screening analysis. A variety of airborne and satellite based sensor systems are used in this work, from high to low spatial and spectral resolution. The team appears to be a potentially strong support element to this program through its expertise in the research proposed. A sun photometer was installed at the northern lake site in 1994, but was moved to the roof of the Space Science and Engineering (SSEC) building at the University of Wisconsin campus where it can be maintained on a continuos basis.
2.6.5 BNZ: Bonanza Creek site
-Biome: Alaska / Polar Evergreen forest site
Phyllis Adams and Dave Verbyla presented their remote sensing work which includes classification of land cover at the BNZ LTER site of an area roughly 300km x 300km. Other research at the BNZ site includes work with the Fairbanks SAR facility at the University.
2.6.6 KNZ: Konza Prairie:
-Biome: Prairie
Doug Goodin presented research efforts which include the continuation of the FIFE program at the Konza prairie site. There was extensive sun photometer work done at KNZ within the FIFE program, but calibration of the data was problematic.
2.6.7 HFR: Harvard Forest
- Biome Deciduous Forest
Mary Martin presented the remote sensing studies at HFR concerning estimation of chemical content of the vegetation using remote sensing data. Much of this work has been done using data from AVIRIS. Although some sun photometer data was collected at the site in 1993, suitable remote sensing data concurrent with these data were not acquired.
3.1 Technical, logistical issues and work plan
- John Vande Castle
A couple of issues were brought up by John Vande Castle concerning the mechanism to evaluate the atmospheric correction effect on satellite data.
3.2 Present and futures datasets
-John Vande Castle
John Vande Castle summarized the datasets presently available for the project, both Landsat-TM data in the LTER archive (with no sun photometer data), those data already acquired with sun photometer data, and plans for future datasets. Future dataset descriptions have been identified and will be discussed further through the "spam@lternet.edu" email group. This acquisition effort will need to be coordinated with NASA Headquarters. The requested TM Path/Row for the different sites were discussed and include the following:
Site Path/Row
3.3 Sun photometer network status and plan
-John Vande Castle
The sun photometer network for LTER research has been running well. All sites currently participating are interested in continued acquisition of data. This situates sun photometer installation on a continuous basis for SEV and NTL sites. A sun photometer is installed from April through September at BNZ, and moved to MCM for use from December through February. An instrument is installed at the H.J. Andrews experimental Forest LTER (AND)from April through October and has been used in the polar summer for SeaWifs work at the PAL LTER site. Future plans call for polling interest by LTER sites for additional instrument installations. A "spare" sun photometer could also be maintained at the LTER Network Office to cover specific over-flights at LTER sites. Future work with EOS validation work is planned to include extensive use of the sun photometer data at LTER sites as well as expansion of work in this workshop to other sensor platforms.
3.4 LTER discussion for dataset evaluation plan
- all
The workshop participants discussed procedures on evaluation of the corrected data sets. John Vande Castle suggested that LTER investigator apply any currently used remote sensing algorithms to both corrected and uncorrected data. This could be in the form of image classification or even NDVI analysis. Some of the data sets (e.g. AVHRR) are relatively new for some of the LTER investigators, and may not yet be applicable to operational corrections. It was decided to give some more thought in the next few months to the evaluation process (see schedule). An Email group was created for use by sun photometer/atmospheric correction work called "spam@lternet.edu", and this will be used in the future by those associated with LTER sun photometer/atmospheric correction work.
-John Vande Castle
Remer, L.A., Y.L. Kaufman and B.N. Holben. 1995. The size distribution of ambient aerosol particles: smoke vs. urban/industrial aerosol. Proceedings, Chapman Conference on Biomass Burning.
Holben, B.N., T.F. Eck, I Slutsker, D. Tanr, J.P. Buis, A. Setzer, E. Vermote, J.A. Reagan and Y.K. Kaufman. 1994. Multi-band automatic sun and sky scanning radiometer system for measurement of aerosols. Proceedings: Sixth International Symposium Physical Measurements and Signatures in Remote Sensing.
Hall, F.G., D.E. Stebel, J.E. Nickeson and S.J. Goetz. 1991. Radiometric rectification: toward a common radiometric response among multidate, multisensor images. Remote Sens. Environ: 35:11-27
Vande Castle, J. R. 1991. Remote sensing and modeling activities for long-term ecological research. Proceedings: GIS/LIS'91, Atlanta ACSM/ASPRS pp. 544-550
Day 1. - Sunday, June 25,1995
Atmospheric effects and remote sensing data: effects, theory and applications
8:15 Van departs Hotel
8:30 Room 261 Bloedel, Logistics, Introduction, Objectives - John Vande Castle
9:00: Atmospheric effect on remote sensing data: theory - Eric Vermote
10:30: Break
11:00: What are the Sun Photometer Data -Brent Holben
12:30: Lunch
1:30: Atmospheric effects on remote sensing data: applications - Eric Vermote
3:00: Break
3:30: Overview of the atmospheric processing software for TM
5:00- Adjourn - Salmon BBQ at John Vande Castle's / sunset discussion at Carkeek Park
Day 2. - Monday, June 26,1995
Hands-on use of data access and correction software
8:30am: Introduction to LTER"NET" satellite and sun photometer data archive, and interface with data correction software - John Vande Castle
9:00am: How to access the sun photometer data - Brent Holben
10:00am: Break
10:30am: How to evaluate the quality of sun photometer data - Brent Holben
12:30am: Lunch
1:30pm: Presentation and Use of the 6S code - Eric Vermote
2:15pm: first TM test dataset proposal -Eric Vermote
2:30pm: Site presentation by participants: 5:30pm: Adjourn
Day 3. - Tuesday, June 27,1995
Operational aspects
9:00am: Technical, logistical issues and work plan - Eric Vermote
9:30am: Present and futures datasets -Eric Vermote
10:00am: Sun photometer network status and plan - John Vande Castle
10:30am: LTER discussion for dataset evaluation plan - all
11:00am: Schedule -Eric Vermote
11:30am: adjourn
Phyllis Adams Site: BNZ
Institute of Northern Forestry
308 Tanana Drive
Fairbanks, AK 99775-5500
LTERnet address: PAdams@LTERnet.edu
E-Mail: pAdams@spruce.lter.alaska.edu
Phone: (907) 474-3310
FAX: (907) 474-3350
---------------------------------------------------------------
David W. Bolgrien Site: NTL
University of Wisconsin - Madison
1225 W. Dayton St. Rm. 1201
Madison, WI 53706-1695
LTERnet address: DBolgrien@LTERnet.edu
E-Mail: bolgrien@facstaff.wisc.edu
Phone: (608) 263-3266
FAX: (608) 262-5964
Organization(s): LTER
---------------------------------------------------------------
Warren B. Cohen Site: AND
USDA Forest Service
Forestry Sciences Lab
3200 SW Jefferson Way
Corvallis, OR 97331
LTERnet address: WCohen@LTERnet.edu
E-Mail: cohen@fsl.orst.edu
Phone: (503) 750-7322
FAX: (503) 750-7329
---------------------------------------------------------------
Gayle Dana Site: MCM
Desert Research Institute
University and Community College System of Nevada
7010 Dandini Boulevard (89512), PO Box 60220
Reno, NV 89506-0220
LTERnet address: GDana@LTERnet.edu
E-Mail: gdana@maxey.unr.edu
Phone: (702) 674-7538
: (702) 674-7539
FAX: (702) 673-7397
Organization(s): LTER
---------------------------------------------------------------
Doug Goodin Site: KNZ
Kansas State University
Department of Geography
Manhattan, KS 66506
LTERnet address: DGoodin@LTERnet.edu
E-Mail: dgoodin@ksu.ksu.edu
Phone: (913) 532-6727
Organization(s): LTER
---------------------------------------------------------------
Joshua Greenberg Site: NET
College of Forest Resources, AR-10
University of Washington
Seattle, WA 98195
LTERnet address: JGreenberg@LTERnet.edu
E-Mail: josh@debris.lternet.washington.edu
Phone: (206) 543-4512
Organization(s): LTER
--------------------------------------------------------------
Brent N. Holben
NASA
Goddard Space Flight Center
Mailstop 923.0
Greenbelt, MD 20771
LTERnet address: BHolben@LTERnet.edu
E-Mail: brent@kratmos.gsfc.nasa.gov
Phone: (301) 286-2975
FAX: (301) 286-1757
Organization(s): NASA
---------------------------------------------------------------
Mary Martin Site: HFR
University of New Hampshire
Complex Systems Research Center
Morse Hall
Durham, NH 03824
LTERnet address: MMartin@LTERnet.edu
E-Mail: mem@woody.unh.edu
Phone: (603) 862-4098
: (603) 862-3033
FAX: (603) 862-0188
---------------------------------------------------------------
Barbara Nolen Site: JRN
New Mexico State University
Department of Biology
Box 30001, Dept. 3AF
Las Cruces, NM 88003
LTERnet address: BNolen@LTERnet.edu
E-Mail: bNolen@nmsu.edu
Phone: (505) 646-4465
FAX: (505) 646-5665
---------------------------------------------------------------
Gregory A. Shore Site: SEV
University of New Mexico
Department of Biology
Biology Annex
Albuquerque, NM 87131
LTERnet address: GShore@LTERnet.edu
E-Mail: gShore@sevilleta.unm.edu
Phone: (505) 277-2109
FAX: (505) 277-0304
--------------------------------------------------------------
John R. Vande Castle Site: NET
LTER Network Manager
University of Washington
LTER Network Office
College of Forest Resources, Box 352100
Seattle, WA 98195-2100
LTERnet address: JVandeCastle@LTERnet.edu
E-Mail: jvc@LTERnet.edu
Phone: (206) 543-6249
FAX: (206) 543-7295
---------------------------------------------------------------
Dave Verbyla Site: BNZ
University of Alaska, Fairbanks
Department of Forest Sciences
Fairbanks, AK 99775
LTERnet address: DVerbyla@LTERnet.edu
E-Mail: dVerbyla@merlin.salrm.alaska.edu
Phone: (907) 474-6927
Organization(s): LTER
---------------------------------------------------------------
NASA Goddard Space Flight Center
Code 923
Room N8, Building 16W
Greenbelt, MD 20771
LTERnet address: EVermote@LTERnet.edu
E-Mail: eric@kratmos.gsfc.nasa.gov
Phone: (301) 286-6232
FAX: (301) 286-3221
Organization(s): NASA
---------------------------------------------------------------
Sally Westmoreland
Department of Geography
San Diego State University
San Diego, CA 92182-4493
LTERnet address: SWestmoreland@LTERnet.edu
E-Mail: sjw@photon.com
Phone:
FAX:
---------------------------------------------------------------
Randolph H. Wynne Site: NTL
UW-Madison Environmental Remote Sensing Center
Atmospheric, Oceanic, and Space Sciences, Rm 1249
1225 West Dayton Street
Madison, WI 53706-1695
LTERnet address: RWynne@LTERnet.edu
E-Mail: rhw@albert.ersc.wisc.edu
Phone: (608) 263-6584
FAX: (608) 262-5964
