Report of the
Regional Science Topic
Workshop on
\ Remote Sensing
* and Landscape
Characterization
November 1 - 3, 2005
Chicago, Illinois
REGION
DEVELOPMENT
-------
Report of the Regional Science Topic Workshop on
Remote Sensing and Landscape Characterization
November 1-3, 2005
Chicago, Illinois
Table of Contents
NOVEMBER 1,2005
Bharat Mathur, Deputy Regional Administrator, U.S.EPA Region 5
David Klauder, EPA-ORD Office of Science Policy
Gary Foley, Director EPA-ORD National Center for Environmental
Research
David Macarus, Planning Committee Chair, U.S. EPA Region 5
Stephen Gotanson, Chief of Information Services, U.S. EPA Region 5
Welcome to the Workshop 1
NOVEMBER 1,2005, Afternoon Session
Moderator. Steve Goranson, EPA Region 5, Office of Information Systems
Remote Sensing Overview: EPA Capabilities, Priority Agency Applications,
Sensor/Aircraft Capabilities, Cost Considerations, Spectral and Spatial
Resolutions, and Temporal Considerations
Ross Lunetta, EPA-ORD National Exposure Research Laboratory,
Environmental Science Division, Research Triangle Park, NC 2
National Land Cover Database
James Wickham, EPA-ORD National Exposure Research Laboratory,
Environmental Science Division, Research Triangle Park, NC 3
National Wetland Inventory and Remote Sensing Applications in Wetlands
Evaluation
Brian Huberty, US Fish and Wildlife Service, Ft. Snelling, MN 3
Activities in Great Lakes Landscape Characterization and Urban Energy
Use Sustainability
Bert Guindon, Canada Centre for Remote Sensing, Ottawa, Ontario,
Canada. 4
NOVEMBER 2,2005, Morning Session
Moderator: Pranas Pranckevicius, EPA Region 5, Great Lakes National
Program Office
Using Remote Sensing, GIS, and Field-Based Techniques to Assess Ecological
Conditions in the Great Lakes Basin
Ric Lopez, EPA-ORD, National Exposure Research Laboratory,
Environmental Science Division, Las Vegas, NV 5
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
APR 1| 2D06
OFFICE OF
RESEARCH AND DEVELOPMENT
MEMORANDUM
SUBJECT: Regiojj/paD^R^rno^ Ser)sing(Landscape Characterization Workshop
FROM:
ANSI'S fan i . Xdmoft ^
•--^X-.L •. ... -
TO: All Workshop Participants
Thank you for your participation in the Region/ORD Remote Sensing/Landscape
Characterization Workshop held November 1-3, 2005, at the Region 5 office in Chicago.
Attached is the final workshop report, which summarizes the workshop presentations and
discussions.
This ORD-sponsored workshop was jointly planned and implemented by staff
from EPA Regions 4 and 5, the Office of Research and Development, and the Office of
Environmental Information. The major objectives of the workshop were to provide
participants with a better understanding of the science and scientific applications relevant
to remote sensing and landscape characterization and establish a network of EPA
scientists who, after the workshop, will continue to exchange information and work
together on this science. This training workshop addressed: 1) how remote sensing
technologies are applicable to environmental work; 2) how remote sensing technologies
integrate with land-based data to assist Regions, States, Tribes, and local communities in
landscape characterization for environmental purposes; 3) how to access remote sensing
technologies, both inside and outside of EPA; and 4) how remote sensing relates to the
Global Earth Observation System of Systems.
For additional information on this workshop, please contact David Macarus,
Region 5 Science Liaison, at (312) 353-5814 or Tom Baugh, Region 4 Science Liaison,
at (404) 562-8275. Contact Dick Garnas, Office of Science Policy/ORD, at (202) 564-
6785 for information on other workshops in the Region/ORD Science Topic Workshop
Series.
Attachment
cc: Deputy Regional Administrators
ORD Executive Council
Regional Science Liaisons to ORD
Internet Address (URL) • http://www.epa.gov
Recycled/Recyclable « Printed with Vegetable Oil Based Inks on Recycled Paper (Minimum 20% Postconsumer)
-------
Large Area Monitoring for Pesticidal Transgenic Crops: How
Spectral Imaging May Play a Role
John G laser, EPA-ORD, National Risk Management Research
Laboratory, Cincinnati, OH 14
LD>AR for Leak Detection
Barry Feldman, EPA Region.6, Dallas, TX 15
Use of Remote Sensing To Assess Aquatic Systems
Robert Hall, EPA Region 9, San Francisco, CA 16
ROUND TABLE DISCUSSION: Regional, State,
Tribal, and Local Views
Panel Members: Ted Prescott, Illinois Environmental Protection
Agency; John Esch, Michigan Department of Environmental
Quality; Jeff Herter, New York State Department of State;
E.J. McNaughton, Indiana Department of Environmental
Management; and James Robb, Indiana Department of
Environmental Management
Panel Comments 17
General Discussion 18
Attachment 1: EPA Office of Science Policy; Regional Science
Workshop on Remote Sensing and Landscape Characterization
Agenda
Attachment 2: EPA Office of Science Policy; Regional Science
Workshop on Remote Sensing and Landscape Characterization
Participants List
Attachment 3: EPA Office of Science Policy; Regional Science
Workshop on Remote Sensing and Landscape Characterization
Evaluation Summary
-------
The Great Lakes Observing System (GLOS) Remote Sensing Subsystem Plan
Roger Gauthier, Great Lakes Commission, Ann Arbor, MI 6
The Midwest Spatial Decision Support System Partnership Update with
Emphasis on Remote Sensing/Landscape Characterization and Rural to
Urban: Using NASS Cropland Data Layers to Estimate Changes
Richard Farnsworth, Purdue University, West Lafayette, IN 7
Did the May 2004 Milwaukee Sewer Overflows Affect Chicago Beaches?
David Rockwell, Great Lakes National Program Office, Chicago, IL 8
An Advanced Geospatial Approach to Develop High-Resolution Population
Distribution for Night and Day
Budhendra Bhaduri, Oak Ridge National Laboratory, Oak Ridge, TN 8
NOVEMBER 2,2005, Afternoon Session
Moderator: Carmen Maso, EPA Region 5, Office of Science, Ecosystems,
and Communities
Data Integration: Making Data Accessible and Useful for Regions, States,
and Tribes
Steve Young, U.S.EPA, Office of Environmental Information,
Washington, DC 10
GEOSS and EPA
John Lyon, EPA-ORD, National Exposure Research Laboratory,
Environmental Science Division, Las Vegas, NV 11
Land-Cover Characterization and Change Detection Using Multi-
Temporal MODIS NDVI Data
Ross Lunetta, EPA-ORD, National Exposure Research Laboratory,
Environmental Science Division, Research Triangle Park, NC 11
Forecasting Environmental Change
Bruce Jones, EPA-ORD, National Exposure Research Laboratory,
Environmental Science Division, Las Vegas, NV 12
The Coastal Change Analysis Program (C-CAP): Land Cover and
Change Information for the Nation's Coasts
John McCombs, National Oceanic and Atmospheric Administration
Coastal Services Center, Charleston, SC.. 13
Regional Vulnerability Assessment (ReVA)
Betsy Smith, EPA-ORD, National Exposure Research Laboratory,
Environmental Science Division, Research Triangle Park, NC 14
NOVEMBER 3,2005, Morning Session
Moderator John Perreeone, HP A Region >, OffJee of Sei
Ecosystems, and Communities
-------
U.S. Environmental Protection Agency
Office of Science Policy
Regional Science Workshop on Remote Sensing and Landscape Characterization
U.S. EPA Region 5
Lake Michigan Room, 12th Floor
Valdas V. Adamkus Environmental Resource Center
77 W Jackson Boulevard
Chicago, IL 60604
November 1-3,2005
MEETING SUMMARY
November 1,2005
Welcome to the Workshop
Mr. Bharat Mathur, Deputy Regional Administrator, U.S. Environmental Protection Agency (EPA)
Region 5, welcomed the participants to Chicago and to the Regional Science Workshop on Remote
Sensing and Landscape Characterization. He explained that the general goal of Region 5 workshops is to
provide scientists with the tools to address topics such as emerging pollutants, pesticide exposures,
endocrine disrupters, and so forth. One goal of this workshop is to develop a network of Regional
scientists to collaborate and share ideas on remote sensing and landscape characterization to furthci the
topic.
Dr. David Klauder, from EPA's Office of Research and Development's (ORD) Office of Science Policy
(OSP), welcomed participants to the meeting and explained that OSP sponsors and co-hosts the Regional
workshops with the Regions. He introduced Dr. Gary Foley, Director of EPA's National Center for
Environmental Research via video.
Dr. Foley explained that ORD, with nearly 2,000 employees, 14 laboratory and research facilities, and a
$55 million external grant budget, is tasked with providing credible, relevant, and timely research results
and technical support to inform EPA policy decisions. Making decisions with sound science requires
relevant and high-quality research that is properly characterized and appropriately used in the
decisionmaking process. EPA's high-priority research areas include human health, particulate matter,
drinking water, global change, endocrine disrupters, homeland security, and many others. The general
goals of the Regional workshops are to create cross-Ageacy science networks and collaborations, identify
opportunities to integrate EPA's science into Regional decisionmaking, and identify the most critical
science uncertainties. Topics of the workshops are identified by the Regions and anticipate future
environmental issues. Remote sensing and landscape characterization was identified as a topic because it
offers new tools to enhance the ability to monitor environmental changes and better focus on remediation
strategies.
Dr. David Macarus, the Planning Committee Chair from EPA Region 5 Office of Science, Ecosystems,
and Communities, thanked the collaborators of the meeting, including Thomas Baugh, Steve Goranson,
John Lyon, Ed Washbum, Ross Lunetta, and all of the speakers. He introduced Stephen Goranson, the
moderator for the afternoon session.
Stephen Goranson, Chief of the Information Services Branch at EPA Region 5, explained that the specific
workshop objectives were to discuss current and planned projects using remote sensing and landscape
characterization information and to identify real possibilities for collaborative projects on the gathering
and application of remote sensing and landscape thaructdizatioa infoniiailOfl.
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
.November 1.. 2005, Afternoon Session
Steve Goranson, EPA Region 5, Office of Information Systems
Remote Sensing Overview: EPA Capabilities, Priority Agency Applications, Sensor/Aircraft
Capabilities, Cost Considerations, Spectral and Spatial Resolutions, and Temporal Considerations
Ross Lunetta, U.S. Environmental Protection Agency, Office of Research and Development, National
Exposure Research Laboratory, Environmental Science Division, Research Triangle Park, NC
Within the Environmental Science Division (ESD) of EPA's National Exposure Research Laboratory
(NERL), there are two main branches that provide remote sensing support, the Landscape Ecology
Branch and the Landscape Characterization Branch. Data costs for remote sensing vary widely from no
cost to being cost prohibitive. U.S. government satellite data and archive aerial photography are available
for a nominal data reproduction and distribution charge, which is applied for both Landsai imagery and
archive aerial photography. Data costs generally increase proportionally with the increased spatial
resolution (geometric); aircraft data are approximately one order of magnitude more expensive than that
collected from satellite platforms, whereas commercial satellite imagery is approximately one order of
magnitude more costly than that acquired from U.S. government assets. In planning for a remote sensing
project budget, a guideline to follow is that data acquisition will account for 15 percent of the budget, data
processing for SO percent, and accuracy assessment for 35 percent. General data types for remote sensing
include aerial photography, multispectral imagery, and light detection and ranging (LIDAR); radar and
microwave data are rare. Tn formulating a remote sensing project, it is important to articulate the
information (data) needs required in as much detail as possible and to formulate the data quality objectives
required for specific data needs. The three basic principals of remote sensing that are relevant to this workshop
are thematic, spatial, and temporal resolutions.
Currently available satellites are able to provide increased spatial resolution, which provides increased
landscape patchiness, or increased temporal resolution, which provides increased biodiversity
information; however, no satellite currently exists that can provide both spatial and temporal resolutions
at high levels. Remote sensing can be used for water quality applications (e.g., mapping submerged
aquatic vegetation beds, monitoring turbidity plumes, real-time measurement of water quality parameters,
etc.); landscape (terrestrial) applications (e.g., wetland mapping); and atmospheric monitoring
applications (e.g., impact of smoke on regional air quality, determination of ammonium emission,
detection of paniculate matter exposure). In performing accuracy assessment of the gathered data, it is
important to consider three quality principals: radiometric quality, geometric quality, and thematic
quality. After data are collected, an accuracy assessment is performed to deteimine the source of any
errors by comparing the remote sensing data to reference data (e.g., field measurement data, interpreted
aerial photography, and other remote sensing-derived data). The current trends in remote sensing include:
(1) application of high-frequency remote sensor data collections to support multitemporal data analysis;
(2) development of a multistage approach that uses coarser resolution (high frequency) data for complete
area coverage supplemented with higher resolution data for identified areas of interest; (3) integration of
in situ monitoring networks with remote sensor data to extend the spatial distribution of field monitoring
observations; and (4) use of unmanned aerial vehicles with integrated sensor packages for both multistage
data collections and "telaplace" monitoring capabilities.
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
National Land Cover Database
James Wckham, U.S. Environmental Protection Agency, Office of Research and Development,
National Exposure Research Laboratory, Environmental Science Division, Research Triangle Park,
NC
The original 1992 National Land Cover Database (NLCD) was a bilateral initiative between ORD and the
Earth Resources Observation and Science (commonly known as EROS) Data Center. The Multi-
Resolution Land Characteristics (MRLC) Consortium is a group of Federal agencies that formed as a
consortium to acquire remote sensing data for environmental monitoring. Currently, the consortium
includes EPA, the U.S. Geological Survey (USGS), National Oceanic and Atmospheric Administration
(NOAA), U.S. Department of Agriculture (USDA), U.S. Forest Service, Bureau of Land Management,
National Park Service, U.S. Fish and Wildlife Service (USFWS), and National Aeronautics and Space
Administration (NASA). For the project, the continental United States is divided into 66 mapping
quadrants, and the metadata of the 2001 NLCD consists of image data (collected in the spring, summer,
and fall), ancillary digital elevation model (DEM) data, impervious surface and tree canopy derivatives,
and land cover data. The three primary components of the 2001 NLCD are land cover (both native pixel
and a convolved 4-pixel product), impervious surface, and canopy density and are supported by a node
map, a confidence map, and decision rules for node map interpretation. The impervious surface
component is an estimated percentage based on an extrapolation from the digital ortho quarter quads,
resulting in a crisper map when compared to the 1992 NLCD. The node map component is derived from
regression tree software, and each "terminal node" or "leaf is assigned a number that can be used with an
associated iext file to trace the history of classification. The confidence map is derived from the
regression tree in 1 percent increments with two different patterns of significance.
Currently, land cover, impervious surface, and canopy density data are available for various regions of the
United States. The 2001 data are freely available from the MRLC Consortium Web Site at
http://www.mrlc.gov/. In comparing the 1992 data with the 2001 data, however, a pixel-to-pixel
comparison is not possible because data are not in the same geometric space (i.e., 90-m DEM in 1992 vs.
30-m DEM in 2001), different classification methodologies were used, and slight changes in class
definitions occurred. The approach to compare the two and generate land cover change data that is low
cost, operationally fast, rigorous, robust, and applicable nationwide is to: (1) compare NLCD 1992 and
NLCD 2001 at Anderson Level 1 to establish areas of agreement; (2) use these areas of agreement as the
source of training pixels to develop a decision-tree classification of the 1992 image mosaic, as well as the
2001 image mosaic; and (3) use confidence data, spatial information, and spectral data to generate
Anderson Level 1 classifications that identity areas of probable change To request land cover change
imagery, send an e-mail to esrisupport@epa.gov and provide the tar file name(s) from the MRLC Web
Site ('http://www.mrlc.govv do wnload_data.asp) or provide path/row and date. The staff then will arrange
the best method of data transfer. Currently, NLCD data are being used in a project to address impaired
water sites in Illinois and their likelihood of recovery.
National Wetland Inventory and Remote Sensing Applications in Wetlands Evaluation
Brian Huberty, U.S. Fish and Wildlife Service, Ft. Snelling, MN
The National Wetland Inventory (NWI) is an environmental indicator tool that can identify the status and
trends of U.S. wetlands and also serves as a resource management tool, providing a standard
wetland/surface water map of the United States. The NWI is important for resource management, water
purification, fish and wildlife habitat identification and protection, flood prevention, erosion control,
recreation, biological productivity, and water supply protection. Previously, the national status of
wetlands was determined every 10 years but now is determined every 5 years. Although the race of
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
wetland loss over time has declined, this is a result of an increase in backyard ponds; actually, natural
wetlands are decreasing. Currently, NWI is correcting and updating features of wetland maps with digital
orthoimages. NWI is also creating a Master GeoDatabase (MOD), one of the top three largest civilian,
topologically structured datasets in the world. The MOD is a seamless, standardized dataset that allows
state and public access via a Web site.
Informational goals (e.g., spatial, spectral, or temporal) determine what resolution is necessary for data
interpretation and identification. Another key component of information needs is positional accuracy.
The Landsat satellite program most likely is approaching its end, but the USDA's National Agriculture
Imagery Program, which acquires aerial imagery during growing seasons, makes its digital ortho
photography images available. Four major vendors are sources for remote sensing equipment: (1)
Airborne Data Systems, Inc.; (2) Leica Geosystems; (4) Intergraph Corporation (Z/IDMC* Digital
Mapping Camera); and (4) Vexcel (UltraCam™ Digital Aerial Camera). Another tool for remote sensing
is LIDAR, which uses lasers to emit light pulses that strike the ground and reflect back to the airborne
sensor. Because the precise altitude and position of the sensing aircraft is known, the elevation of surface
points can be determined based on the amount of time it takes for the pulse to return to the sensor.
Interferometric Synthetic Aperture Radar (known as IFSAR) is another tool to determine elevation
information. Currently, there are no comprehensive and current (statistical or mapped) estimates or
comprehensive and synoptic high-resolution images of the Great Lakes Basin wetlands. This is
significant because the Great Lakes account for 20 percent of the Earth's and 95 percent of North
America's surface freshwater. Geospatial infrastructure, in addition to additional satellites, airborne data,
and geospatial information, is needed for future remote sensing applications.
An upcoming event that may be of interest is the American Society for Photogrammetry and Remote
Sensing's Great Lakes Conference to be held October 28-November 1, 2007, in Ottawa, Canada.
Additionally, Dr. Huberty will be resurrecting the wetland remote sensing component of the Federal
Geographic Data Guidelines and asked participants to contact him if they were interested in this area.
Activities in Great Lakes Landscape Characterization and Urban Energy Use Sustainability
Bert Guindon, Canada Centre for Remote Sensing, Ottawa, Ontario, Canada
In the early 1990s, the Canada Centre for Remote Sensing (CCRS) and EPA compared expertise in image
processing, archival imagery, systems design and development, landscape ecology, water quality
assessment, and available datasets and determined that a collaboration between the two involving land
cover mapping of the Great Lakes watershed would be synergistic. From this collaboration, the
Composite Land Processing System (CoLaPS) was developed to merge together the multitude of
composite archival Landsat scenes from 1972-1992. CoLaPS does not take a traditional approach of
mosaicing images together to merge the scenes but instead composites land cover classifications.
Additionally, it provides a seamless linkage between its land cover production system and user
subsystem. CoLaPS offers the ability to study subareas (e.g., subwatersheds) and perform riparian
analysis. The system was used for forest fragmentation analysis in the Great Lakes sub-basins using an
algorithm that compared the relationship of ratios of forest pixels surrounded by other forest pixels and
then linking it to deforestation models to determine how sensitive the fragmentation characteristics were
to deforestation strategies. These models also can be used to forecast deforestation in developing
countries.
Another project of CCRS, mandated by the Canadian government, is to utilize remotely sensed land use
and land cover information and determine how it impacts energy consumption. The federal priorities for
this project include urban agenda development, transportation-related energy consumption, environment
issues, climate change, and land \i
-------
from transportation, two-thirds of which come from urban areas. The strategy to forge science-policy
links to manage the energy priorities included providing a historical perspective and also relevant
geospatial information (i.e., the Canadian Urban Land Use Survey, which integrates land cover/land use
information derived from Landsat images with census information), methods for indicator quantification
(e.g., urban compactness and the transport mode index), a survey of geospatial aspects of work-related
travel (i.e., the archetypal Canadian urban form), travel prediction models to support forecasting, and
direct access to geospatial information for policymakers. The ultimate goal of the project is to develop
forecasting capabilities that can be given to policymakers to predict energy consumption.
November 2, 2005. Morning Session
Moderator: Pranas Pranckevicius, EPA Region 5, Great Lakes National Program Office
Using Remote Sensing, GIS, and Field-Based Techniques To Assess Ecological Conditions in the
Great Lakes Basin
Ric Lopez, U.S, Environmental Protection Agency, Office of Research and Development, National
Exposure Research Laboratory, Environmental Science Division, Las Vegas, NV
The practical applications of remote sensing, geographic information systems (GIS), and field techniques
are being used to map wetlands and analyze ecological and watershed condition metrics in the Great
Lakes Basin and also to create new maps and improve the ecological functions (e.g., attenuating floods,
improving water quality, and increasing biological diversity/wildlife habitat) of palustrine wetlands of the
Gulf Coast Region. The Landscape Ecology Approach is an ecological approach at a broad scale. In
using this approach, the Landscape Ecology Branch of BSD produces land cover datasets, examines
changes in land use or cover (i.e., potential "drivers" of ecological change), and then determines changes
in landscape composition and pattern over time and space (e.g., forest fragmentation). These changes are
then linked with potential changes in ecological processes (i.e., "receptors") that affect terrestrial species.
These changes may cause changes in ecological goods and services, which in turn may drive
policymaking decisions. Performing the landscape ecology assessment of the Great Lakes Basin using
this approach presented challenges, such as mapping land cover for a vast area (i.e., drivers), calibrating
geospatial data with field information, and mapping land cover metrics that have potential as indicators of
ecological condition (i.e., receptors). The summary dataset from the assessment is available to the Great
Lakes research community in the Landscape Ecology Metric Browser (v2.0) and includes information
about general landscape characteristics, sediment flowing to coastal wetlands, sediment available for
coastal nourishment, urban density, land conversion, habitat adjacent to coastal wetlands, and habitat
fragmentation, which are linked with State of the Lakes Ecosystem Conference (SOLEC) indicators. The
initial landscape metrics findings indicate that increased wetland edge and wetland density are associated
with reduced plant community biodiversity.
The landscape ecology approach also was used to assess the ecological functions of depressional wetlands
in the Gulf Coast Region. Depressional wetlands were chosen because they: (1) have similar functions to
other wetlands; (2) are individually small and diffuse but cumulatively include a tremendous amount of
wetland; (3) are easier to identify because there is consensus on their size, shape, location, position, and
proximity; and (4) may be "isolated" in the landscape, leading to questions of how they function. Data
gathered about the depressed wetlands can be extrapolated to larger watersheds (e.g., the Great Lakes
Basin). Multiseason/multiyear data representative of the region from 1999-2003 was used in a regression
analysis. Photogrammetric accuracy assessment was done to determine if the technique was comparable
in accuracy to NWI, NLCD, and Gap Analysis Program (GAP) vegetation association datasets. The GAP
vegetation association dataset, however, was not useful for identifying depressional wetlands in coastal
Texas. The ultimate goal of the project is to produce a map product for the Region with the relative
accuracy with respect to the NLCD and NWI datasets.
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
The Great Lakes Observing System (GLOS) Remote Sensing Subsystem Plan
Roger Gauthier, Great Lakes Commission, Ann Arbor, MI
The Global Earth Observing System of Systems (GEOSS), coordinated by the United Nations and the
World Meteorological Organization, developed from an international desire to develop a framework of
interoperability between and within nations to share information about the environment at large.
Additionally, a Commission on Oceans Policy report called for a major investment in the Great Lakes,
oceans, and coastal areas. As a result, NOAA led legislative activities to form the Integrated Oceans
Observing System (IOOS). The objectives of IOOS are to: (1) facilitate safe and efficient maritime
operations; (2) mitigate the effects of natural hazards; (3) improve prediction of climate variability; (4)
reduce public health risks; (5) improve national security; (6) sustain and restore living resources; and (7)
preserve and restore healthy ecosystems. The IOOS approach is focused on defining research needs and
developing pilot activities for preoperational projects to create operational functionality with a linkage at
the local, regional, and global levels. Eleven regional associations are being developed to support IOOS,
many of which are led by academic institutions. The Great Lakes Commission, with Federal support, is
coordinating the Great Lakes Observing System (GLOS), one of the 11 regional nodes of IOOS. GLOS
supports 13 user communities and focal areas, including the important areas of safe drinking water and
commercial navigation.
A draft strategic business plan for GLOS was completed, and a regional data management and
communications (DMAQ node currently is being developed. Additionally, a pilot data integration
project is being developed for the St. Clair and Detroit rivers and the Lake St. Clair system. GLOS is
focusing on integrated data collection, analyses, modeling, and value-added product creation for a series
of meteorologic, hydrologic, hydraulic, chemical, and biological themes. The remote sensing subsystem
focal areas are lake-wide process monitoring, regional land cover assessments, high-resolution land use
changes, coastal wetlands monitoring, and emergency response products. Additionally, nearshore
observations of wind, waves, water levels, water chemistry, sediment, nutrient, and contaminant loading
to the system are important. The nearshore subsystem components include water level observation
system improvements, vertical control network improvements, directional wave metering, high-frequency
radar deployment, airborne LIDAR and hyperspectral mapping, tributary streamfiow station upgrades,
and sediment transport modeling. Components of the interconnecting waterway subsystem are permanent
flow metering, Lake St. Clair buoy deployment, hydrodynamic modeling, and field calibration. The
hydrodynamic modeling subsystem includes modeling improvements, data assimilation and analysis, and
operational product delivery. The information integration subsystem, a major focus of the project,
includes a regional DMAC clearinghouse node, Web partnerships, linkage to the Great Lakes Information
Network, and national DMAC interfaces. The regional emphasis for remote sensing is to implement a
coastal remote sensing program, conduct periodic technical workshops, and initiate focused education and
outreach.
The Great Lakes Interagency Task Force draft plan calls for a $20 billion investment to restore ecological
balance to the system and is focused on supporting the U.S. contribution to GEOSS through IOOS/GLOS
and the Integrated Earth Observing System (IEOS), enhancing the coordination of monitoring activities,
implementing high-value SOLEC indicators, establishing a regional information management
infrastructure, creating a Great Lakes communications workgroup, and increasing funding for Great
Lakes research. Two critical near-term actions will be to enhance the physical and chemical observations
in the Lake Huron to Erie corridor through implementation of GLOS and convene information managers
under the Regional Data Exchange Initiative to develop inventories of data resources and initiate Web-
based integration among agencies.
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
The Midwest Spatial Decision Support System Partnership Update With Emphasis on
Remote Sensing/Landscape Characterization and Rural to Urban: Using NASS Cropland
Data Layers To Estimate Changes
Richard Famsworth, Purdue University, West Lafayette, IN
In Dr. Bartholic's unexpected absence, Dr. Famsworth presented information from both presentations as
part of an update on the Midwest Watershed Decisions Support Systems Partnership (MWDSSP).
At the April 2002 Midwest Web-Based Spatial Workshop in Chicago, Illinois, a group of organizations
and universities formed the MWDSSP to develop decision support tools and systems for decisionmakers,
including watershed planning groups, public officials, and development and planning groups in
government and the private sector, to assist them in developing and implementing economically viable,
ecologically sound plans. The specific objectives are to: (1) develop, promote, and use Web-based, user-
friendly, geospatial watershed management data and decision support tools; and (2) help set the standard
for other watershed management programs, such as promoting data initiatives, relating specific tools to
planning process phases and learning objectives, and creating systems where outputs of tools can be
plugged into other models. The tools developed for local officials, natural resources managers, and the
general public are Web-based, spatially based and scalable, science-based, accessible in the public
domain, intuitive, customizable, and freestanding.
In 2002, Michigan State University was working on a digital watershed project, a nationwide Web
application tool for effective watershed management, and Purdue University was working on long-term
hydrologic impact analysis model simulations for watersheds. A partnership formed with the goal of
making the digital watershed available to various stakeholders and the public. To meet that goal, the
digital watershed must be an information repository that has nationwide watershed coverage, multiple
forms of access, comprehensive datasets, a scaling function, online environmental modeling, and online
erosion and deposition modeling. The digital watershed provides enhanced usefulness of the Great Lakes
Basin landscape ecology metric. Additionally, another goal of the MWDSSP is to utilize LandSat for
sprawl assessment to facilitate wise land use deeisionmaktag. To this end, overlay methods are used in
wide-area landscape analysis to combine different datasets to help local decisionmakers in zoning and
other development determinations. MWDSSP also has enhanced the usability and functionality of EPA's
STORET database, which also is important for communities trying to make decisions.
Dr. Famsworth concluded the MWDSSP presentation and began his presentation on using the National
Agricultural Statistics Service's (NASS) cropland data layers to estimate change.
The challenge was to address the loss of agricultural and forest lands to urbanization, but existing
estimates of land use changes were based on random samples and only valid across the state. A previous
GIS project (i.e., the Indiana University-Purdue University Indianapolis [IUPUI] land use map) was
useful but too time-intensive and costly to repeat. Therefore, the project objectives were to: (1) use past
land cover spatial data and existing remotely sensed data (e.g., the IUPUI land use map and the USDA's
NASS cropland data layers from 2000-2003) to estimate land use change; (2) use ModelBuilder to
document the analysis; and (3) summarize the conclusions and make appropriate recommendations. The
advantages of using the NASS data was that it provided yearly data for 4 years and beyond, contained
major land use categories as part of the map layers, and offered high accuracy for major agricultural
crops. Some disadvantages included the presence of considerable variation across years in land cover as
well as the fact that pasture/rangeland is considered a "catchall" category, distributed throughout urban
and rural areas, and comprised approximately 30 percent of each year's data. The solution was to apply
rules to fill in missing data and remove cloud cover, use all available data (e.g., the IUPUI map layer and
data) to creat,? a hiw 2000 1«n.1 «-ov«r mop wiih a fosua on Urban tircits, and icly CAlcmJvcly to Hie
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
neighborhood analysis tool to reduce speckling and reclassify pixels in all categories. Following this
process, data were analyzed and it was determined that part of the urban growth in Indiana between 2000
and 2003 is capturing earlier urban growth and areas that were misclassified as pasture/rangeland. The
conclusions of the project were that: (1) NASS data layers contain valuable information, especially when
combined with other spatial data. (2) Incorporating new NASS layers and repeating the analysis will
begin to capture new growth. (3) ModelBuilder provides a cost-effective framework for incorporating
new data and running the analysis yearly. (4) The addition of other spatial data layers will improve model
results.
Did the May 2004 Milwaukee Sewer Overflows Affect Chicago Beaches?
David Rockwell, Great Lakes National Program Office, Chicago, IL
Initial reports from the Milwaukee Metropolitan Sewerage District indicated that heavy rains during the
month of May 2004 resulted in more than 4 billion gallons of storm and sewage overflows in Milwaukee,
Wisconsin. The estimated amount of the sewer overflows was later revised to 1.2 billion gallons. The
deep tunnel systems that Milwaukee utilized were overwhelmed by the amount of rainfall received and
the excess overflow entered Lake Michigan. Chicago beaches were temporarily closed in June 2004
because of high Escherichia coli levels in the water and the Milwaukee sewage overflow events were
blamed. EPA was mandated to examine the potential impact of the Milwaukee discharge on Chicago
beaches.
To accomplish this, hourly wind records were gathered from surface weather stations around Lake
Michigan for the period of May 1-June 20,2004; station observations were interpolated to create
continuous wind fields over the entire lake. The wind fields then were used to drive a lake circulation
model, which was calibrated and validated using the Lake Michigan Mass Balance datasets of 1994 and
1995. Currents from the lake circulation model were used to track the movement of water originating
from Milwaukee Harbor as it moved into southern Lake Michigan toward Chicago. Additionally,
extensive E. coli samples were collected between May 14-May 19,2004, in Milwaukee Harbor. The
highest concentrations were in the river water plume inside the harbor and within 1 km or the harbor
breakwall. As determined by conductivity measurements, concentrations dropped sharply outside of the
river water plume. Sampling surveys on subsequent days consistently demonstrated less than 200 E. coli
per 100 mL at distances 2-5 km from the harbor breakwall and less than 20 E. coli per 100 mL at
distances greater than 5 tan from the harbor. The lake circulation model, E. coli sampling, and Moderate
Resolution Imaging Spectroradiometer(MODIS) satellite imagery indicate that the high£. coli
concentration near Chicago beaches was not a result of the Milwaukee discharge. Investigating Chicago
precipitation levels, however, indicated that each spike in E. coli levels immediately followed a
precipitation event in Chicago. The recommendation is not to go swimming within 48 hours following a
rain event.
A brief videotape introducing Dr. George M. Gray, the newly appointed Assistant Administrator of
EPA's ORD, was shown.
An Advanced Geospatial Approach To Develop High-Resolution Population Distribution for Night
and Day
Budhendra Bhaduri, Oak Ridge National Laboratory, Oak Ridge, TN
Utilizing Census data for modeling and simulations causes challenges as a result of the Census data
limitations, including temporal resolution limitations (e.g., Census data is geared toward residential and
nighttime populations^ and spatial resolution limitations (e.g.. Census blocks often are too big).
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
Additionally, Census data assumes a uniform distribution of the population, which is not realistic. A
gridded approach, however, assumes a more realistic nonuniform distribution of population with
attributes associated with individual cells. Smart interpolation of the data ensures that there are many
layers in the decision process. The LandScan model, developed by the Oak Ridge National Laboratory, is
a population distribution model, database, and tool developed from Census and other spatial data using a
uniform regular grid. LandScan was developed via Dasymetric Spatial Modeling and distributes the best-
available Census counts to LandScan cells based on a likelihood coefficient calculated by this model.
The model structure is the same everywhere, but weights and scores for each variable (e.g., slope;
nighttime lights; land cover; road proximity; population data; and jail, hospital, and school locations) are
tailored to each region. The footprint of urban areas is not always captured correctly as a result of the
imprecise urban boundaries caused by the phenomenon of nonurban areas located inside the urban
boundary and urban areas located outside of the urban boundary. To improve urban boundaries, high-
resolution images are utilized to weight individual cells within an urban area relative to building type and
densities for a more realistic distribution within the urban area. Additionally, many different datasets are
used for LandScan, including Census block data for population, the Census Bureau's TIGER* Database
for roads, NLCD for land cover/land use, National Elevation Data for slope, and many others.
Each Census block is evaluated individually. To determine the LandScan Nighttime Model, the.
population density of the block is compared with the number of houses in the block to determine if the
data are realistic. The information is then compared to eight Census block models based on percent
developed land, and additional data are analyzed (e.g., slope, land cover, road and railroad distances, area
landmarks, schools, airports, etc.) to determine actual nighttime population distribution. Once the
Nighttime Model has been prepared, the Daytime Model can be constructed utilizing tract-to-tract worker
flow, Bureau of Labor statistics, and InfoUSA data to determine the number of workers per Census block.
Again, land cover, road and railroad, airports, school, and other datasets are analyzed to help determine
population distribution. Additionally, nighttime population, tract-to-tract worker flow, and school age
children data are used to determine the size of the nonworking adult population. A subpopulation
(approximately 20%) of the adult population is the mobile "shoppers" that are not necessarily at home and
may be in nonresidential locations. Finally, stay-at-home, worker, shopper, ^ciiool, and prison
populations are analyzed to produce the daytime population distribution. Oak Ridge National Laboratory
has been tasked by the Department of Homeland Security to use this approach to build the baseline data
for daytime and nighttime populations for the entire Nation.
Validation and verification are performed to ensure that error in the model is minimized. Input data are
quality assured, and model refinements account for the spatial disparity of population density. Census
control totals are reconciled, the model is run with coarse data and compared to actual dmta, and data are
compared to imagery to validate and verify the model. Inaccurate or imprecise TIGER* data are
corrected when necessary. School databases and corresponding spatial locations are validated using
imagery. Currently available land cover data and imagery include NLCD, GeoCover, and Coastal
Change Analysis Program (C-CAP) data. Additionally, land use map data are more useful than land
cover data.
Space-time visualization is extremely critical for disaster preparedness; therefore, the more data that are
available, the better the models can be. It is a constant and ongoing effort to develop the models and data
free of charge for the Federal government and other users throughout the world. In addition to the
Daytime and Nighttime Population Models, models for lunchtime, rush hour, and special events (e.g.,
concerts, sports events) are being developed.
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
November 2, 20QS» Afternoon
Maso, EPA 5, Office of Science, Ecosystems, and Communities
Data Integration: Making Data Accessible and Useful for Regions, States, and Tribes
Steve Young, US. Environmental Protection Agency, Office of Environmental Information,
Washington, DC
Decisionmakers believe that they need more data, but this is not necessarily the case. From a historical
perspective, in die past, data were scarce, and the analysis of the scarcely available data was at a
premium. Time moved more slowly, the world was larger, and computers generally were for limited
applications such as payroll. Over time, however, computers proliferated and became smaller and less
expensive, increasing amounts of data became available, data systems proliferated, and the "elite" group
of computer programmers began to down, albeit slowly at first. Following this "renaissance," flic
approach to datasharing was to build a central database, provide a decision support system, generate
graphics, and produce primitive user interfaces. Currently, as technology has advanced, the approach to
datasharing is to build a centralized CIS; add satellite images or aerial photos; consider business
intelligence, indicators and metrics, expert systems, and so forth; and improve data management,
including registries and other shared resources. Datasharing was able to move from the old approach to
the new approach because of increased availability of personal computers, the World Wide Web, and
CIS; the development of networks, especially the Internet; the creation of the mouse and the graphical
user interface: heightened user expectations: and an increased desire for demonstrated return on
investment. The important idea to remember in gathering and analyzing date is to look at the big picture;
ask the question, "Is it only tunneling a subset of the information, or is it giving an accurate picture of the
larger world?"
Rapid remote sensing and the development of the portal concept (e.g., NOAA's Weather.gov Web Site)
of datasharing are major developments in making data accessible. The keys to making emerging
approaches effective are establishing Web services, service-oriented architectures, and (from GEOSS)
interoperability and "system of systems" thinking. Current approaches are lacking data integration; it is
necessary to go beyond simply gathering data and move toward data intelligence. Decisionmakers need
intelligent data to make decisions, instead of just receiving a colossal amount of raw data. The GEOSS
perspective, with an emphasis on Earth Observing Systems (EOS), recognizes that models (e.g., weather,
climate, atmosphere, etc.) play a large part in the big picture feedback loop. These predictive models then
are used to effect societal benefits through decision support systems. Additionally, decisionmakers need
interactive models, visualizations, scenario development, "What-if?" capabilities, relevant indicators, a
view of the larger picture (i.e., of life support systems such as water, oxygen, food, health), better GIS and
remote sensing, better interpretive tools (e.g., imagery feature extraction and georeferencing), and
intelligence. Multiple sensors of life support systems are necessary to see the larger picture, and
technology has increased these capabilities by an order of magnitude as cost has continued to decrease.
Ultimately, what is needed is real-time monitoring of vital Earth systems, information delivery to
decisionmakers, the ready ability to assess decision outcomes, early recognition of surprises, and support
for adaptive management. The adaptive management paradigm states that every action taken is
considered an experiment because the outcome is not guaranteed, and, therefore, flexibility is needed to
make course corrections after actions have been taken.
10 Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
GEOSS and EPA
John Lyon, U.S. Environmental Protection Agency, Office of Research and Development, National
Exposure Research Laboratory, Environmental Science Division, Las Vegas, NV
Through strong international and national cooperation, existing national and international monitoring
systems that will provide more complete, accurate, and accessible data and information to users and
decisionmakers can be improved. GEOSS is an important element of national and global strategies for
managing natural resources in a sustainable way. In Tokyo, Japan, April 2004, ministers, leaders, and
delegates from 45 nations discussed the GEOSS implementation plan to foster societal benefits and
determined that GEOSS would improve how the Earth system (i.e., its weather, climate, oceans, land,
geology, natural resources, ecosystems, and natural and human-induced hazards) is perceived and
understood. Such understanding is crucial to enhancing human health, safety, and welfare; alleviating
human suffering, including poverty; protecting the global environment; and achieving sustainable
development. EOS provides that understanding and, therefore, continual benefits toward sustainability.
GEOSS will potentially improve the current framework for environmental monitoring that exists largely
in support of regulatory imperatives but that does not yet comprehensively communicate the information
and models needed for decisionmakers and the public to understand and address complex questions. Data
management plans for policy and management decisions are a synthesis of multiple and often divergent
views, including the GEOSS strategic plan, the IEOS strategic plan, U.S. policy, agency implementation,
and IOOS DMAC, among others. Near-term opportunities that address current needs can be executed
using existing community standards and protocols that conform to an extensible, component-based
architecture that has a demonstrated use for decisionmaking.
It was determined that integrated data management, increased observations for disaster management,
global land observations, integrated draught observations, sea level observation systems, and air quality
monitoring are important to EEOS. The vision of the United States' contribution to GEOSS, via IEOS, is
to enable a healthy public, economy, and planet through an integrated, comprehensive, and sustained
Earth observation system. EPA's GEOSS tools may be found online at http://www.ep_3.Ac>\'..i^gpss/e.Q_s/
epa_eos.html. Within GEOSS, the objective of GLOS is to develop the organizational infrastructure for a
regional observing system node to coordinate data collection, modeling, and product development for the
Great Lakes, their connecting channels, and the St. Lawrence River with a cooperation between the
United States and Canada. Current progress on EOS and GEOSS can be found at
http://www.earthobservations.org/ and http://iwgeo.8sc.nasa.gov/.
Land-Cover Characterization and Change Detection Using Multi-Temporal MODIS NDVI Data
Ross Lunetta, U.S. Environmental Protection Agency, Office of Research and Development, National
Exposure Research Laboratory, Environmental Science Division, Research Triangle Park, NC
The Normalized Difference Vegetation Index (NDVT) is the normalized ratio of the red chlorophyll
absorption well (670 nm) and the Near-Infrared (NIR) foliar vegetation reflectance maximum (860 nm)
and gives the relative index of standing photosynthetieally active biomass (PAB). The MODIS NDVI 16-
day composite product (version 4.0) is provided by NASA, free of charge, on the Internet and is useful
data. MODIS multitemporal imagery has multiple resolutions. The objectives of utilizing MODIS
imagery are to: (1) develop change detection methods to append (change only) the NLCD 2001 baseline
land cover database on an annual basis; (2) provide continuous measures of PAB distributions to drive the
next generation of landscape process models; (3) evaluate the use of MODIS multitemporal data to
identify land cover change locations and patterns in near real-time; and (4) determine the utility of
MODIS to classify land cover to characterize the outcome of alteration (conversion) events. The study
area for the project is the major waters of the Albemarle/Pamlico Estuary. A land cover reference dataset
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization 11
-------
based on color infrared digital ortho quarter quads was created that was suitable for the accuracy
assessment of MODIS 250 m data products. To perform the accuracy assessment, a random sampling of
52 digital ortho quarter quads was obtained by stratifying by Bailey ecoregion; 13 quads were selected
and overlaid with the cell network and dot grid. Validation indicated that the Level I agreement was quite
good with agreement at a difference of 10 percent or less occurring in greater than 90 percent of the
quads.
Following the verification and validation process, data were input into the grid module, and bad data were
identified and eliminated by filtering pseudo hikes and drops followed by applying NDVI data flags for
acceptable data. The data gaps left by this process were managed by performing a Fourier transformation
of data into the frequency domain, separating out the cleaned signals, and performing an inverse Fourier
transformation into the time domain to estimate the missing data points. Following this process,
exploratory work with the data was performed and land features of interest were identified by end-
members (groups of relatively pure gpectra), it was found that §re§r«p follows a Ysiy predictive pattern,
and the NDVI was similar from year to year. For detecting change in land cover/land use, water and
agricultural masks were created to reduce false positives. Total NDVI from each year were calculated
from the data and compared for deviations that indicate change. Thresholds, standard deviations from the
mean, are utilized for mis comparison. Accuracy assessments indicated that using 2.5 standard deviations
from the mean yielded the best results.
In summary, two-date change detection using established analytical methods tends to be performance
limited in biologically complex ecosystems. Before extracting image end-members and performing
change detection, MODIS time series data must first be filtered and cleaned. Results indicate that
MODIS NDVI time series data analysis represents a substantial improvement in change detection
monitoring capabilities. Potential EPA applications for MODIS NDVI multitemporal products include
annual land cover change pattern products, updates to baseline land cover products, and continuous NDVI
datastreams for future environmental modeling efforts. Potential collaborations could be utilized to: (1)
determine the potential applications for annual change detection pattern products for regional scale
assessments; (2) integrate continuous NDVI datastreams with distributed landscape process models to
advance environmental monitoring and forecasting capabilities; and (3) develop land cover change
desktop alarm capabilities to provide a potential real-time monitoring and regulatory support capability.
Forecasting Environmental Change
Bruce Jones, U.S. Environmental Protection Agency, Office of Research and Development, National
Exposure Research Laboratory, Environmental Science Division, Las Vegas, NV
A forecast is the best estimate from a particular method, model, or individual given a set of specific
assumptions that may or may not turn out to be true. As such, it is imperative for forecasts to be
associated with estimates of uncertainty so that decisionmakers have information as to the likelihood of a
given forecast. Ecological forecasts offer decisionmakers estimates of ecological vulnerabilities and
potential outcomes given specific natural events and/or management or policy options. Ecological
forecasting is critical in understanding potential changes in ecological services before they happen and in
developing strategies to offset or avoid catastrophic losses of services and provisions. Three types of
ecological forecasts are vulnerability assessments based on current conditions, short-term forecasts, and
long-term forecasts.
For vulnerability assessments, the Genetic Algorithm for Rule-Set Prediction (GARP) is utilized. GARP
is a data-mining, inductive approach also utilized by NASA and USGS for their Invasive Species
Forecasting System fhttp://bp.gsfc.nasa.govA. EPA also has instituted the Regional Vulnerability
Assessment (ReVA) Program, which provides indices of relative condition and vulnerability Vwwed on
multiple datalayers and models for multiple environmental endpoints related to multiple stressors. Other
12 Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
examples of vulnerability assessment approaches include the USGS* GAP and ReGap analyses and the
Nature Conservancy's vulnerability assessments. Short-term forecasts use real- or near-real-time data
(onsite and/or remotely sensed) and base biophysical conditions that do not change (e.g., biophysical
characterizations and ecosystem resiliency). Short-term models relate conditions and species occurrences
to important drivers that change (e.g., ocean temperature, currents, etc.) and are the basis for mapping
forecasts onto remotely sensed and other spatially continuous data. NOAA's monitoring of harmful algal
blooms, the USGS/USFWS/NOAA bird migration monitoring, the Landscape Fire and Resource
Management Planning Tools Project, crop productivity forecasts, and drought and famine forecasts in
Africa are all examples of short-term forecasting programs. Long-term forecasts are scenario-based and
use base biophysical conditions that do not change to create models that relate conditions and species
occurrences to important drivers that do change. The goal of long-term forecasts is to develop decision
tools and Web-based applications. The EPA Corvallis Laboratory's Alternative Futures Analysis
Process, ReVa, and the Midwest Spatial Decision Support System Partnership are examples of long-term
forecasting projects.
New projects for forecasting include the National Ecological Observatory Network (NEON;
http://www.neoninc.org/) and the National Phenology Network. The NEON's mission is to identify and
understand critical variations and interactions in environmental drivers that will enable forecasting the
state of ecological systems for the advancement of science and the benefit of society. The NEON
Observatory Implementation Model will consist of 20 nodes, each with 3 fixed and 1 mobile observing
platform, representing climatic domains. The National Phenology Network is a continental-scale network
that observes regionally appropriate native plant species and is designed to complement remote sensing
observations. The data collected will be freely available to the research community and general public.
Gaps that still need to be addressed for ecological forecasting include retaining and upgrading remote
sensing platforms that measure land-surface, freshwater, and ocean conditions; improving biophysical
data, compatibility among data, models and linkages among models, and in situ monitors; providing
historical reconstruction (e.g., phylogenetics) and fixed, repeat biological data; and developing unproved
delivery systems to decisionmakers.
The Coastal Change Analysis Program (C-CAP): Land Cover and Change Information for the
Nation's Coasts
John McCombs, National Oceanic and Atmospheric Administration Coastal Services Center,
Charleston, SC
C-CAP, the objective of which is to improve scientific understanding of the linkages between coastal
wetland habitats, adjacent uplands, and living marine resources, was conceived in the late 1980s and
implemented as a NOAA program in the mid-1990s via grants and cooperative agreements. The C-CAP
Effectiveness Review Panel identified 13 recommendations to make the program more useful, including
the 7 following: leveraging other national efforts; focusing on applications; producing standard,
consistent, timely products; minimizing duplicate efforts; partnering with the private sector; executing
extensive outreach efforts; and collaborating with the USGS. The collaborative discussions with the
USGS led to acceptance of the MRLC database concept. NOAA's innovative approach utilizes
classification and regression tree analysis with standardized and accessible inputs and metadata tracking
of procedures, all of which has led to increased credibility and usefulness of the land cover data. In
addition to MRLC efforts, C-CAP also performs separate accuracy assessment and validation and has
changed the classification scheme by adding subclassifications to the woody and herbaceous wetlands
classifications. C-CAP has produced a digital map product line that includes land cover data for 2001,5-
year retrospective land cover data for 1996, retrospective change data (from 1996 to 2001), percent
impervious surface and percent tree canopy data, and Federal Geographic Data Committee metadata. The
data can be utilized to perform a very quick and simple land cover analysis snmmnry and pen??at
change. C-CAP has completed baseline data for the West and Gulf Coasts (minus Florida) of the United
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization 13
-------
States and the Great Lakes Basin. The Florida coastline and the East Coast contracts are in place. C-
CAP's priorities, however, changed following Hurricanes Katrina and Rita, resulting in the new priority
of remapping hurricane-affected areas.
Appropriate uses of remote sensing include large-scale and regional applications (e.g., watershed, county,
and state), resource inventories, population growth trends, and habitat fragmentation studies. Remote
sensing should not be utilized for jurisdictional wetlands legislation (1 acre minimum mapping unit may
overlook small, isolated wetlands or changes to wetlands), dock siting, parcel mapping, wetland
permitting, or small-scale studies. C-CAP is attempting to provide the high-resolution data that is in
demand. Available GIS tools to enhance the use of C-CAP data can be found online at
http://www.csc.noaa.gov/crs/lca.
Regional Vulnerability Assessment (ReVA)
Betsy Smith, U.S. Environmental Protection Agency, Office of Research and Development, National
Exposure Research Laboratory, Environmental Science Division, Research Triangle Park, NC
One task of the ReVA Program is to understand how to utilize existing data and make it concise and
meaningful for decisionmakers. Integrated science information is needed to meet ecosystems challenges.
Prior strategies included synthesizing existing information to improve understanding of the effects of
multistresses on the environment. ReVA synthesizes environmental data and model results to inform
decisionmaking. Some issues that need to be managed are discontinuity of valuables, highly skewed data.
unbalanced data (i.e., a large amount of terrestrial data but relatively little aquatic data), and the
interdependency of variables (i.e., values and metrics are correlated). Because there may be statistical
problems in putting together the data, many methods were examined to determine the sensitivity to data
issues. How each integration method assessed each data issue was examined. Multidecision criteria
requires multiple integration methods. ReVA is currently developing a toolkit that is accessible to states,
Regions, and so forth that is completely portable and can be used on any scale. The toolkit was externally
reviewed and received good reviews.
November3. 2QOS. Morning Session
Moderator. Joiin Perreecme, £PA Region 5, Office of Science, Ecosystems, and Communities
Large Area Monitoring for Pesticidal Transgenic Crops: How Spectral Imaging May Play a Role
John Closer, U.S. Environmental Protection Agency, Office of Research and Development, National
Risk Management Research Laboratory, Cincinnati, OH
Of the 200 million acres of global transgenic crops, more than 50 percent are in the United States. Corn
bioengineered to produce its own insecticide, Bacillus thuringiensis delta endotoxin (Bt com), is viewed
as an environmental asset for both human health and ecosystems because of the possible avoidance of
pesticide applications; therefore, its lifetime is important to environmentally sustainable considerations.
As the Federal Insecticide, Fungicide, and Rodenticide Act (commonly known as FIFRA) requires all
pesticides to be registered, EPA is involved in the valuation process for Bt com. Growers using Bt corn
must have an integrated resistance management plan in place, and structured refuge, secondary pest
impacts, multiple crop pest impacts, and monitoring and surveillance are the responsibility of the
registrant. The existing monitoring strategy is to examine the development of resistance in insects and
pests that are supposed to be controlled. Current monitoring strategy concerns are: (1) Four limited
sections of com crop are used as the proxy for the entire crop. (2) Infestations for the com pests are
expected to begin as local phenomena. (3) The question of whether or not current proxy samplings
provide Adequate information for resistance detection. In developing a proactive tnoaiioriag approach,
14 Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
four items need to be considered: (1) There must be a representative sampling of all acreage that is not
physically possible to sample. (2) The current monitoring strategy provides insufficient warning of
resistance development. (3) No direct molecular technique is available currently to detect resistance.
(4) A comprehensive approach to reduce reporting time for resistance monitoring is needed.
Related spatial technologies are utilized to view leaf properties via a reflected light spectrum where
incoming light is preferentially absorbed (reflected) depending on plant physiology. The health of the
leaf, which is indicative of the health of the plant, can be determined via the light absorption patterh.
Transgenic crop lanes also are visible via infrared photography. The project investigated if transgenic
corn varieties could be distinguished from their nearest isolines (i.e., breeding pairs without transgenic
traits) by spectral reflectance imagery and if pest infestations in corn could be identified by reflectance
imagery. A Real-Time Data Acquisition Camera System-Hyperspectral, with a spectral resolution from
400-1000 nm and a spatial resolution of 0.5 m, was utilized for the project. The experimental design
included a complete random block design replicated five times, two separate infestations with European
Corn Borer (the first Infestation at V8-VIO growth stages), damage assessment via the Guthrie rating, and
all plots imaged in a 2-week schedule. Logistic regression was used to differentiate groups of hybrids
(i.e., a transgenic hybrid and its near isoline from other hybrids) with an overall accuracy of 92 percent.
Overall, there was 78 percent accuracy in separating transgenic from nontransgenic corn varieties.
Currently, the project is in the proof of concept stage, with the development of the analytical system and
initial ground training completed. The project will soon be moving into the proof of principle stage with
statistical testing development and data-mining. The proof of practice stage will follow with a working
system for compliance and/or resistance monitoring.
LIDAR for Leak Detection
Barry Feldman, U.S. Environmental Protection Agency, Region 6, Dallas, IX
The innovation process is different in technologies that help maintain the public good of a clean
environment. Private investment incentives are particularly weak, whereas the government role in
promoting innovation is relatively strong. Government affects innovation, both directly and indirectly,
via regulation, research funding, financial incentives (e.g., tax credits and other subsidies), and facilitation
of technology transfer. One current environmental challenge is focused on fugitive emissions of
hydrocarbons from industrial sources. Fugitive emissions are defined as routine leakage from valves,
pumps, flanges, connectors, and so forth that aggregate to significant tonnage. These emissions can be
hazardous air pollutants with direct health impacts; and can be controlled by equipment design or, more
commonly, leak detection and repair programs. The vast majority of equipment does not leak; 84 percent
of leaks come from 0.13 percent of equipment. Because of this trend, industry and EPA would like to
find a faster, less expensive way to find leaks. Because most leaks are at higher ppm, they can be
detected by remote sensing. Industrial process units have more than 10,000 different parts subject to
monitoring, some of which that are classified as "hard to monitor." Second generation gas finder infrared
thermal cameras, especially when combined with a telephoto lens, are extremely valuable for monitoring
areas previously classified as hard to monitor.
The basic principle of LIDAR is that solar energy heats materials on the Earth's surface, which is then
emitted back into the atmosphere at longer wavelengths. When this emitted energy passes through a
plume, some of the energy gets absorbed at specific wavelengths. The absorption signature is the
chemical identification. The next steps for utilizing LIDAR for leak detection include demonstrating
reliability in field use as well as detection limits and the subjectivity of interpretation, standardizing
procedures and recordkeeping, investigating the commercial availability of units, and developing
alternative technologies. EPA, the American Petroleum Institute, and title American Chemistry Council
are working together to develop testing procedures for detection limits and the range of chemicals that can
be detected. Verification will include laboratory and field testing. Currently, only two field-ready
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization 15
-------
imagers are known. The challenge will be to find additional commercially ready units. EPA's
Environmental Technology Council (ETC), the purpose of which is to achieve environmental results
through the application of innovative technology, has recommended approval of a project to address
improved compliance through remote sensing. Obstacles to remote sensing include the need to verify the
detection limits and range of chemicals the equipment can detect, and regulatory methodology needs to be
changed to allow the use of remote sensing use. Future applications of remote sensing include the
improvement of emissions inventories, determination of the cause of spikes in community monitoring
networks, and performance of compliance inspections to check stack emissions and compliance with
LIDAR.
Use of Remote Sensing To Assess Aquatic Systems
Robert Hall, U.S. Environmental Protection Agency, Region 9, San Francisco, CA
Structural stream physical habitat characterization includes measurements of stream size, gradient,
channel substrate type and size, riparian vegetation cover, structure and complexity, and anthropogenic
alterations. These physical characteristics strongly influence water quality and the capacity of a stream to
support a diverse biological community. Habitat complexity is the distribution of various types of
features providing fish concealment (e.g., large woody debris, undercut banks, overhanging vegetation,
boulders, and residual pools). Hyperspectral imagery can be used to detect and map the distribution of
large woody debris (i.e., greater than 15 cm diameter) and overhanging vegetation. LIDAR is used to
measure large boulders greater than 1 m and also residual pool depth, size distribution, and frequency.
Combining the two types of data provides a means to estimate residual pool volume, exposed mid-
channel gravel and/or sand bars, reach-scale indices of slackwater volume, and channel complexity.
Additionally, LIDAR data can measure quantitatively wetted width, bank slope, incised height greater
than 15 cm, bankfull height, and width at bankfull stage. The power of hyperspectral imagery is that a
large number of very narrow contiguous spectral bands of data can show spatial extent and allow for
quantitative summary and analysis of features. This is an automated method to measure wetted width and
to determine the number of hectares of riparian vegetation. Using near-infrared to determine wetted
width is impacted by streambed substrate composition and size, riparian vegetation, and shadows.
Riparian vegetation canopy cover is essential for moderating stream temperatures as well as providing
habitat and an indicator of the potential aquatic community present. Hyperspectral imagery integrated
with LIDAR data can determine vegetation type, structure, height, and distribution and also monitor
invasive species.
The objective of the Advanced Monitoring Initiative is to develop a three-dimensional image of aquatic
riparian habitat of the Humboldt River, Nevada, to classify watershed morphology and invasive
species. Spectral Imagery LIDAR Composite (SELC) technology was used to integrate hyperspectral
digital imagery elements with LIDAR surface data. The SILC process fuses hyperspectral pixels
pbotogrammetrically Wjtj1 individual X,Y,Z values. Each laser return is mathematically projected through
collinearity equations onto its proper position on the frame array. By using imagery acquired
simultaneously with the surface data, each surface point possesses an accurate spectral signature assigned
to its location, allowing accurate classification of features using conventional remote sensing techniques.
The SILC data allow urban terrain, forested terrain, agricultural lands, mountains, cliffs, ravines, and
wetlands to be subjectively masked and filtered, providing vastly improved bare-earth surfaces and
feature extraction. The objective of the Water Quality/Clarity project is to couple multispectral and
hyperspectral remote sensing data to measurable water quality parameters with the lake's physical
dynamics to develop a time series (i.e., seasonal) model of lake water quality/clarity. Methodology
includes calibrating the satellite imagery with current water quality parameters (e.g., total nitrogen,
phosphorus, Chl-a,b,c, turbidity, etc.), developing a time-series model of seasonal fluxes of each of the
parameters, and ultimately, developing a three-dimensional model of the lake by combining high-
resolution multibeam acoustic bathymetry and backscatter data.
16 Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
In summary, quantitative structural and biological stream data acquired over a given spatial area and
temporal period are fundamental information needed to determine reference conditions, stream stability,
and the changing conditions of the stream channel and riparian zones over time. The need for most state
and tribal bioassessment programs is to maximize the return on investment of acquiring this data. There
are multiple advantages to hyperspectral and LIDAR data. The major advantage is the cost to benefit
relative to high-density ground sampling assessment to cover the same land area as it increases resolution
accuracy of more qualitative sampling programs and increases spatial interpolation and extrapolation of
point-based ground measurements.
Round Table Discussion: Regional, State, Tribal, and Local Views
Panel Members: Ted Prescott, Illinois Environmental Protection Agency; John Esch, Michigan
Department of Environmental Quality; JeffHerter, New fork State Department of State; E.J.
McNaughton, Indiana Department of Environmental Management; and James Robb, Indiana
Department of Environmental Management
Panel Comments
Ted Prescott stated that the State of Illinois needs remote sensing tools that can be applied to sites, the
ability to apply stream characteristics to point source and nonpoint source pollution, and the use of leak
detection to determine river pollutants. The Census information that illustrates population movement at
various times of the day would be particularly useful. States also would benefit from available imagery
being placed in a central clearinghouse/portal that each state could access. Currently, most states do not
have the budget to generate their own imagery. Assistance with brownfield redevelopment also would be
helpful.
John Esch stated that geologists have been utilizing geophysics for decades, and he was surprised that
geophysics, a form of remote sensing, was not mentioned during the workshop. He stated that it was hard
for him to visualize the application of the Great Lakes Basin information and tools to Superfund sites.
Additionally, access to historical data is just as valuable as access to new data to assist with ground
contamination cleanup, so that the history and cause of the ground contamination can be determined.
Better elevation data are needed, and the State of Michigan is interested in statewide LIDAR coverage.
There also is a great need in Michigan for thermal sensors (especially for naturally occurring methane).
Although it is not an environmental application, remote sensing could be utilized for investigating the
location of strategic minerals. Additionally, the State of Michigan has an immense interest in glacial
geology mapping. To assess water withdrawal in Michigan, aquifers need to be characterized,
Brownfield redevelopment also is an issue of concern, especially in the industrial cities of Flint, Lansing,
and Detroit.
JeffHerter explained that the New York State Division of Coastal Resources administers grants and
brownfield projects. The GIS branch of the division provides mapping for the use of all branches in the
division, which is currently utilizing remote sensing for land cover change analysis for Long Island, New
York. The division also has completed a static nonpoint pollution project utilizing modified C-CAP data,
as well as a nonpoint pollution model for Lake George, New York. An open space analysis of the south
shore of Long Island with modified C-CAP data has been completed. A wetland loss analysis using aerial
photography was completed for south shore marsh islands. A New York State digital orthophotography
dataset is available for public download at the New York State Web Site. It is a development tool for
local government decisionmakers in which to enter a storm event and pollutant and be provided with best
practices for watershed management. Additionally, New York has developed the New York State GIS
Clearinghouse (http://www pysgisjtate.riy.us/) in a coordinated effort with various partners, including
local governments, state agencies, universities, and industry. Data in the clearinghouse is for members
only. The partnership has several working groups, including a land use/land cover workgroup. The 22nd
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization 17
-------
New York State GIS Conference will be held October 23-24,2006, in Lake Placid, New York
(http://nvsgisconf.esf.edu/). Additionally, he found the free data download sites mentioned in the
workshop useful, as well as the mapping of invasive species via hyperspectral imagery.
E.J. McNaughton stated that the State of Indiana has flown the entire state to create a base product with a
resolution of 12 niches in most areas. Some counties added additional funding to have 6-inch resolution
of their areas. The photos encompass seven terabytes of information and include DEM data with 2 feet
contours. The state is attempting to release the data to all interested parties, including the public, and
Indiana University and Purdue University have the data available for public download on their Web sites.
Infrared data with 1 m resolution will be available on the Web site by March 2006. What the state would
like from EPA and other agencies is more data and a variety of sources to find data of critical importance.
The 2006 Indiana QIS Conference will be held in Indianapolis, Indiana, on March 7-8,2006
('htlp:/.;w.w\v..in^ov/ingisi/conference.''index.htmr),
James Robb stated that remote sensing information has been utilized to examine gross mitigation
compliance. IKONOS satellite data was acceptable for gross mitigation but was not as useful for wetland
impact assessment. The state wishes to utilize remote sensing data to identify those who do not report
wetland impact. Coarser resolution imagery is useful for obvious changes, but less obvious changes also
need to be investigated. It is necessary to prioritize spending for inspections, and the state would
appreciate help in determining what to prioritize. Currently, the State of Indiana is trying to determine
how many wetlands exist in the state and the percent increase or decrease from historical data. It would
be helpful if USFWS information could be scaled down to the state level. Orthophotography and infrared
band data would especially be helpful. Indiana counties currently are leading the way in GIS and have
started mapping land use and examining impervious surfaces. The Indiana Department of Natural
Resources is interested in advanced flood mapping. The Little Calumet River, arguably the most polluted
water body in the United States, has an immense restoration project currently underway. More expertise
and more opportunities to confer with remote sensing experts (e.g., at workshops such as this one) as well
as stable updates and increased sharing (i.e., willingness and ability) also are needed. Knowing for what
purpose data should and should not be used also is important. Historic data are needed to monitor why
and how land use has changed over time. Infrared data at an increased resolution are needed for water
bodies. Buffer mapping also is needed. Land use and land change data need to be standardized.
Validation and ground truthing information need to be shared as well.
General Discussion
Barry Feldmati of EPA Region 6 commented that EPA's ETC is creating a remote sensing database that
will be as good as the information that is being contributed. He encouraged workshop participants to
share their data and help build a quality database. Initially, the database will only be available within
EPA because of the EPA firewall, but work on changing the firewall requirements so that the database
may be shared outside EPA is underway. The hope is that it will be propagated by mid-2006.
Steve Young of EPA's Office of Environmental Information mentioned the National Environmental
Information Exchange Network (http://www.exchangenetwork.netA and its grant program. Sharing data
is consistent with the goals of the grant program, which offers opportunities for states to pursue shared
access for imagery.
Steve Goranson of EPA's Region 5 Office of Information Systems added that he was the Regional contact
for the National Environmental Information Exchange Network grant program, and he followed the
progress of Region 5 grantees. All states within Region 5 have nodes on the network. The highlight of
the grant program is the geospatial information generated. His office is cooperating with the State of
Wisconsin in a discovery and delivery relationship thai complies with EPA'a firewall by establishing an
extranet so that groups outside of EPA can view Region 5 holdings and metadata, and download data.
18 Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
EPA is working on ensuring that various catalogs can network with each other and are well populated;
scalability is a key factor. Brenda Smith of EPA's Office of Environmental Information is the chief
geographic information officer and is working on making geospatial information a one-stop program with
other Federal agencies. The EPA workgroup meets regularly to discuss data needs and other issues and is
very interested in keeping an open dialogue with the states. A data exchange network would open
opportunities for states to inform EPA about their data needs so that grant money could be aimed toward
relevant needs. Additionally, the USDA's Natural Resources Conservation Service has established an
extensive Web site. Jim Wickham of NERL is slowly populating data for public use on the Web. EPA
grants in this area are being promoted. The grant program is a great forum to formulate needs and make
connections.
Ronell Haney of EPA Region 5 found real-time monitoring to be of interest. As it is an intensive process
to rewrite real-time data (i.e., creates an increased load on the Web server), he asked if there was
particular equipment that was necessary and what would be the justification for the equipment. Dr.
Goranson responded that each division must prioritize internally what equipment is necessary for their
requirements. Real-time data are particularly useful in emergency response (Brian Cooper of EPA
Region 5 is working on a rapid assessment tools application). Mechanisms are hi place to capture data
that has been invested in, but a continual program to monitor water must rely on USGS servers and the
main EPA server. A national meeting is taking place in Las Vegas, Nevada, during the last week of
November and first part of December, and there will be a meeting session entitled "Extreme IT" that will
address such issues. ORD is well aware of technology needs. Tools are available, but it is a matter of
prioritization. EPA is working with other Federal agencies to address this issue.
Mike LaBatte of the Lake Traverse Reservation Tribe stated that his tribe is now utilizing GIS, and their
large database has grown dramatically. As GIS Manager, he would like to update land use and land cover
DEM for the 1,500 square mile, predominantly cropland reservation and asked what the best tool was for
his situation, which included producing the land use/land cover data for the tribe to plan for the EPA grant
and also producing the best land use/land cover data to provide tribal decisionmakers with a sufficient
knowledge of their environment. Ric Lopez of EPA's NERL ESD answered that ESD maintains a Web
site (http_^wjvi^eDiyjoAijierJesd_lJ that has links to find a variety of the tools discussed at the workshop.
The Analytical Tools Interface for Landscape Assessments (commonly known as ATtlLA) is a tool for
use in the Arc View environment to produce landscape metrics. Tools also are available to design
individual-specific outputs. Western datasets from southeastern Arizona also are available on the Web
site. Additionally, there are demonstrations of ESD's previous (i.e., the last 5 to 10 years) work with
datasets. The Web site also contains summary date for various regions of the United States. The
Landscape Ecology Branch Web Site (http:/-w\vAv.epa,KOV'nerlesdl'land^xiLdefault.htm) also is very
useful. Dr. Goranson added that Rick Farnsworth a Midwest partnership that encouraged
participation to increase the value of the toolkit, including "off the shelf* data and user models. The Web
address for the partnership is http://www.epa.gov/waterspace. The partnership is working with ORD and
state programs in nonpoint source run-off, total maximum daily loads, and watershed analysis.
Partnerships with and data from other agencies, groups, and organizations would be helpful. Anyone
interested in collaborating can contact Rich Zdanowicz of EPA Region 5. Dr. Lopez reiterated that there
are many resources on the ESD Web Site, including a link to the Environmental Photographic
Interpretation Center (EPIC) group in Reston, Virginia, which is a field station of the Landscape Ecology
Branch of ESD and provides many beneficial data and tools for remote sensing. Dr. Young added that
there also is a phenomenal reconstruction of grounds from a historical perspective. Dr. Lopez added that
historic photography dating back to the 1930s also is available for EPIC, which could be accessed by
visiting the on the ESD Web Site.
Mr. Prescott asked if any of the Federal agencies were able to obtain the Sanborn Fire Insurance maps in
color, because they are invaluable for Superfimd sites Dr- T.npey <;nppe«
able to provide color maps or guide him to resources that could provide this information.
Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization 19
-------
Mr. Herter commented that, although datasharing in the past has not always been forthcoming, New York
State has created a datasharing cooperative that has moved beyond former impediments to datasharing.
At the New York State GIS Clearinghouse Web Site there is a description of the datasharing cooperative
and directions on how to join. Mr. Haney stated that datasharing has different applications and, therefore,
there are many different software programs necessary. He asked Mr. Herter who is responsible for
converting data into a uniformly readable format in the datasharing cooperative. Mr. Herter responded
that when the cooperative was formed, recommendations for the appropriate data format were made.
Generally, map information data are in the tab file format Those who send in data are responsible for
ensuring that the data are in the correct format. Bert Guindon of the CCRS added that there is a lot of
commercially available software for converting data to the proper format.
Mr. Esch returned the discussion to the National Environmental Information Exchange Network grant
program. He commented that the program may be good for certain types of projects but not all. The
Superfund Electronic Data Deliverable (EDD) process made it clear that different groups within EPA are
not communicating, and different Superfund regions have different EDDs. Dr. Young responded that this
was a fair criticism and that EPA is trying to improve communications. The Superfund programs have a
vast amount of money and tend to be somewhat autonomous as a result. Their budget, however, has
decreased, and many programs within EPA are increasing communication and collaboration. The
problem has not been solved completely, but the Office of Environmental Information was created to
work across programs and increase communication. Dr. Goranson added that research and development
architecture needed to be created in addition to program and administrative architecture. There are an
increasing number of Federal partnerships so that Federal partnering agencies can optimize efficiency. In
regard to the Superfund issue, Dave Wilson, the Superfund Division Remedial Project Manager, is
currently trying to convert the EDD data format to an XML format. The attempt is being made to tie
things together as much as possible, including advertising software (e.g., Geode) that is being developed
by the Regions.
Zenny Sadlon of EPA Region 5 Office of Information Services stated that when he first began work at
EPA 5 years ago, he spent a lot of time inputting data into databases because it was not known that the
information was already available. The climate has changed, and EPA is now trying to solve these
problems, including a variety of collaborations with outside organizations. Starting with the belief that
real problems for real people are being solved is a good beginning. The NEON may have the answers.
David Macarus adjourned the meeting at 11:58 a.m.
20 Office of Science Policy Regional Science Workshop on Remote Sensing and Landscape Characterization
-------
Attachment 1
EPA Office of Science Policy
Regional Science Workshop on Remote Sensing and
Landscape Characterization Agenda
-------
Regional Science Workshop on Remote Sensing and Landscape Characterization
U.S. EPA Region 5
Lake Michigan Room, 12th Floor
Valdas V. Adamkus Environmental Resource Center
77 W Jackson Boulevard
Chicago, IL 60604
November 1-3,2005
AGENDA
Pay 1: Tuesday,. November 1.2005
1:00 p.m. Welcome to the Workshop
Bharat Mathur, EPA Region 5 Deputy Regional Administrator
Introduction of the ORD/Regional Workshop Series
David Klauder, ORD/OSP, and Gary Foley, Director of NCER, by video.
Introduction to this Workshop: David Macarus, Planning Committee Chair
1:20 p.m. Remote Sensing: How Do We Do It - What Does It Give Us - Resolution -
Opportunities for Spectral Analysis - Cost of the Various Options
Ross Lunetta, ORD/NERL/ESD, Research Triangle Park, NC
2:20 p.m. National Land Cover Database - Historical 1992-2001
James Wickham, ORD/NERL/ESD, Research Triangle Park, NC
2:50 p.m. Break
3:05 p.m. National Wetland Inventory and Remote Sensing Applications in Wetlands Evaluation
Brian Huberty, Regional NWI Coordinator, USFWS, Ft. Snelling, MN
3:35 p.m. CCRS Activities hi Great Lakes Landscape Characterization and Urban Energy
Use Sustainability
Bert Guindon, Canada Centre for Remote Sensing, Ottawa, ON
4:05 p.m. Discussion and Announcements
4:30 p.m. Adjournment
Pay 2; Wednesday. November 2.2005
8:30 a.m. Great Lakes Challenges
8:30 a.m. Great Lakes Geographic Analysis
Ric Lopez, ORD/NERL/ESD, Las Vegas, NV
9:00 a.m. Great Lakes Observation System "GLOS"
Roger Gauthier, Great Lakes Commission, Ann Arbor, MI
-------
9:20 a,m. The Midwest Watershed Decision Support Systems Partnership - Great Lakes
Update
Rick Farnsworth, Purdue University, West Lafayette, IN, and Jon Bartholic,
Michigan State University, East Lansing, MI
9:50 a.m. "Did the May 2004 Milwaukee Sewer Overflows Affect Chicago Beaches?"
David C. Rockwell, Great Lakes National Program Office, Chicago, IL
10:10a.m. Break
10:30 a.m. Remote Sensing for Day Versus Night Population Estimating/Uses in Emergency Response
and Homeland Security
Budhendra L. Bhaduri, Oak Ridge National Laboratory, Oak Ridge, TN
11:30 a.m. Discussion and Announcements
12:00 noon Lunch
1:00 p.m. Data Integration: Making Data Accessible and Useful for Regions, States, and Tribes
Steve Young, OEI, Washington, DC
1:40 p.m. GEOSS: My Life and Times With a Global Consortium
John Lyon, ORD/NERL/ESD, Las Vegas, NV
2:10 p.m. Landscape Changes - Opportunities for Real Time Monitoring and Enforcement Follow-Up
Ross Lunetta, ORD/NERL/ESD, Research Triangle Park, NC
2:50 p.m. Break
3:10 p.m. Forecasting Environmental Change
Bruce Jones, ORD/NERL/ESD, Las Vegas, NV
3:50 p.m. NOAA C-CAP Program
John McCombs, NOAA, Charleston, SC
4:20 p.m. Regional Vulnerability Assessment (ReVA)
Betsy Smith, ORD ReVA Program Director, Research Triangle Park, NC
4:30 p.m. Adjournment
|>ay 3; Thursday. November 3.2005
8:30 a.m. Crop Identification: Genetically Modified Crops/Identification of Crop Damage
John Glaser, ORD/NRMRL, Cincinnati, OH
9:00 a.m. LIDAR for Leak Detection
Barry Feldman, EPA Region 6, Dallas, TX
9:30 a.m. Quantifying Stream Attributes Using LIDAR and Hyperspectral Imagery
Robert K. Hall, U.S. EPA Region 9, San Francisco, CA
10:00 a.m. Break
-------
10:20 a.m. Regional, State, Tribal, and local Views - Round Table
"What's in It for Me?" Are These Tools Helpful Now or in the Future?
For: Air Quality/Urban and Rural Air Pollution
Water Quality Monitoring/Aquatic Ecosystems
Contamination Sources for Recreational Waters
Pollutant Source Identification Agriculture
Open discussion with audience regarding the application and value of remote sensing for
monitoring, landscape characterization, and enforcement.
12:00 noon Adjournment
-------
Attachment 2
EPA Office of Science Policy
Regional Science Workshop on Remote Sensing and
Landscape Characterization Participants List
-------
Regional Science Workshop on Remote Sensing and Landscape Characterization
U.S. EPA Region 5
Lake Michigan Room, 12th Floor
Valdas V. Adamkus Environmental Resource Center
77 W Jackson Boulevard
Chicago, IL 60604
November 1-3,2005
PARTICIPANTS LIST
Al Alwan
U.S. Environmental Protection Agency
Region 5
Water Division
Water Quality Branch
77 W Jackson Boulevard (WQ-16J)
Chicago, IL 60604-3507
Telephone: (312)353-2004
E-mail: alwan.al@epa.gov
Thomas Baugh
U.S. Environmental Protection Agency
Region 4
Office of the Regional Administrator
61 Forsyth Street, SW, 13th Floor
Atlanta, GA 30303
Telephone: (404)562-8275
E-mail: baugh.thomasl@epa.gov
Robert Beltran
U.S. Environmental Protection Agency
Region 5
77 W Jackson Boulevard (P-19J)
Chicago, IL 60604
Telephone: (312)353-0826
E-mail: beltran.robert@epa.gov
Paul Bertram
U.S. Environmental Protection Agency
Region 5
Great Lakes National Program Office
77 W Jackson Boulevard (G-17J)
Chicago, IL 60604
Telephone: (312)353-0153
E-mail: bertram.paul@epa.gov
Budhendra Bhaduri
Oak Ridge National Laboratory
Geographic Information Science and Technology
PO Box 2008 (MS 6017)
Oak Ridge, TN 37831-6017
Telephone: (865) 241-9272
E-mail: bhaduribl@oml.gov
Mike Bland
U.S. Environmental Protection Agency
Region 5
77 W Jackson Boulevard (P-19J)
Chicago, IL 60604
Telephone: (312)353-9196
E-mail: bland.michael@epa.gov
Tom Brody
U.S. Environmental Protection Agency
Region 5
Office of Information Services
Resource Management Division
77 W Jackson Boulevard (MG-9J)
Chicago, IL 60604
Telephone: (312)353-8340
E-mail: brody.tom@epa,gov
Pamela Brooks
Illinois Environmental Protection Agency
Bureau of Air
1021 N Grand Avenue, East
Springfield, IL 62794
Telephone: (217)524-4776
E-mail: pamela.brooks@epa.state.il.us
-------
Jonathan Burian
U.S. Environmental Protection Agency
Region 5
Water Division
Water Quality Branch
77 W Jackson Boulevard (WQ-16J)
Chicago, IL 60604
Telephone: (312)886-2916
E-mail: burian.jonathan@epa.gov
Larry Cailant
U.S. Environmental Protection Agency
Region 5
Superfund Division
77 W Jackson Boulevard
Chicago, IL 60604
Telephone: (312) 886-3545
E-mail: callant.larry@epa.gov
Jennifer Clarke
Illinois Environmental Protection Agency
Bureau of Water
1021 N Grand Avenue, East
Springfield, IL 62794
Telephone: (217) 524-3040
E-mail: jennifer.clarke@epa.state.il.us
Ken Copenhaver
Institute for Technology Development
National Risk Management Research Laboratory
2401 S Neil Street, Suite 2
Champaign, IL 61820
Telephone: (217)239-7099
E-mail: kcopenhaver@iftd.org
John Esch
Michigan Department of Environmental Quality
Remediation and Redevelopment Division
PO Box 30426
Constitution Hall, 3rd Floor, SW
Lansing, MI 48909
Telephone: (517) 241 -7603
E-mail: eschj@michigan.gov
Richard Farnsworth
Purdue University
Department of Forestry and Natural Resources
195 Marsteller Street
West Lafayette, IN 47906
Telephone: (765)496-3245
E-mail: rlfarnsw@purdue.edu
Barry Feldman
U.S. Environmental Protection Agency
Region 6
Multimedia Planning and Permitting Division
1445 Ross Avenue (6PD-R)
Dallas, TX 75252
Telephone: (214) 665-7439
E-mail: feldman.barry@epa.gov
Roger Gauthier
Great Lakes Commission
Data and Information Management
2805 S Industrial
Ann Arbor, MI 48104
Telephone: (734)971-9135
E-mail: gauuier@glc.org
Michael Gentleman
U.S. Environmental Protection Agency
Region 5
Water Division
Ground Water and Drinking Water Branch
77 W Jackson Boulevard (WU-16J)
Chicago, IL 60604
Telephone: (312) 886-1508
E-mail: gentieman.michael@epa.gov
John Glaser
U.S. Environmental Protection Agency
Sustainable Environments Branch
26 W Martin Luther King Drive (498)
Cincinnati, OH 45268
Telephone: (513)569-7568
E-mail: glaser.john@epa.gov
ToddGoeks
National Oceanic and Atmospheric
Administration—Trustee Representative
U.S. Environmental Protection Agency
Region 5
Coastal Protection and Restoration Division
77 W Jackson Boulevard (SR-6J)
Chicago, IL 60604
Telephone: (312)886-7527
E-mail: todd.goeks@noaa.gov
Stephen Goranson
U.S. Environmental Protection Agency
Region 5
Resource Management Division
Office of Information Services
77 W Jackson Boulevard (MG-9J)
Chicago, IL 60604-3507
Telephone: (312)886-3445
E-niail: goranson, sicphcn@epa.gov
-------
George Graettinger
National Oceanic and Atmospheric
Administration Ocean Service
Office of Response and Restoration
Coastal Protection and Restoration Division
Building Four, Room 2117A
7600 Sand Point Way, NE
Seattle, WA 98115
Telephone: (206)5264660
E-mail: george.graettinger@noaa.gov
Bert Guindon
Natural Resources Canada
Canada Centre for Remote Sensing
588 Booth Street
Ottawa, ON K1AOY7
CANADA
Telephone: (613)947-1228
E-mail: bert.guindon@ccrs.nrcan.gc.ca
Robert Hall
U.S. Environmental Protection Agency
Region 9
Water Division
75 Hawthorne Street (WTR-2)
San Francisco, CA 94105
Telephone: (415) 972-3430
E-mail: hall.robertk@epa.gov
Mark Hamilton
U.S. Environmental Protection Agency
Office of Water
East Building, Room 3349E
1200 Pennsylvania Avenue, NW (MC4101M)
Washington, DC 20460-0001
Telephone: (202)564-0383
E-mail: hamilton.mark@epa.gov
Ronell Haney
U.S. Environmental Protection Agency
Region 5
Resources Management Division
Information Management Branch
77 W Jackson Boulevard (MI-9J)
Chicago, IL 60604
Telephone: (312)353-8975
E-mail: haney.ronell@epa.gov
Jeffrey Herter
New York State Department of State
Division of Coastal Resources
41 State Street
Albany, NY 12182
Telephone: (518)486-7942
E-mail: jherter@dos.state.ny.us
Elizabeth Hinchey Malloy
U.S. Environmental Protection Agency
Region 5
Forestry and Natural Resources Division
Great Lakes National Program Office
77 W Jackson Boulevard (G-17J)
Chicago, IL 60604
Telephone: (312)886-3451
E-mail: hinchey.elizabeth@epa.gov
Brian Huberty
U.S. Fish and Wildlife Service
National Wetland Inventory
One Federal Drive (MS 4056)
Fort Snelling, MN 55111-4056
Telephone: (612) 713-5332
h-mail: brian_huberty@fws.gov
Megan Jamnok
U.S. Environmental Protection Agency
Region 5
Waste Pesticides and Toxics Division
77 W Jackson Boulevard
Chicago, IL 60604
Telephone: (312)886-6716
E-mail: jamnok.megan@epa.gov
Bill Jenkins
U.S. Environmental Protection Agency
Region 3
Environmental Information and
Analysis Branch
1650 Arch Street (3EA10)
Philadelphia, PA 19103
Telephone: (215)814-2911
E-mail: jenkins.bill@epa.gov
Bruce Jones
U.S. Environmental Protection Agency
Office of Research and Development
944 E Harmon Road (6RC-M)
Las Vegas, NV 89119
Telephone: (702)798-2671
E-mail: jones.bruce@epa.gov
-------
David Klauder
U.S. Environmental Protection Agency
Office of Research and Development
Office of Science Policy
109 TW Alexander Drive (D305-1)
Research Triangle Park, NC 27711
Telephone: (919) 541-2232
E-mail: klauder.david@epa.gov
Matthew Koch
U.S. Environmental Protection Agency
Region 5
Superfund Division
77 W Jackson Boulevard
Chicago, IL 60604
Telephone: (312)313-0324
E-mail: koch.matthew@epa.gov
Noel Kohl
U.S. Environmental Protection Agency
Region 5
Office of Information Services
77 W Jackson Boulevard (MI-9J)
Chicago, IL 60604
Telephone: (312)886-6224
E-mail: kohl.noel@epa.gov
Alfred Krause
U.S. Environmental Protection Agency
Region 5
77 W Jackson Boulevard (P-19J)
Chicago, IL 60604
Telephone: (312)353-9196
E-mail: krause.alrred@epa.gov
Michael LaBatte
Lake Traverse Reservation Tribe
PO Box 509
Agency Village, SD 57262
Telephone: (605) 698-4998
E-mail: oepgis@sbtc.net
Ric Lopez
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Environmental Sciences Division
944 E Harmon Avenue
Las Vegas, NV 89119
Telephone: (702) 798-2394
E-mail: lopez.ricardo@epa.gov
Ross Lunetta
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Environmental Sciences Division
Landscape Characterization Branch
109 TW Alexander Drive (E243-05)
Research Triangle Park, NC 27711
Telephone: (919)541-4256
E-mail: lunetta.ross@epa.gov
John Lyon
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Environmental Sciences Division
944 E Harmon Avenue
Las Vegas, NV 89119
Telephone: (702) 798-2535
E-mail: lyon.johng@epa.gov
David Macarus
U.S. Environmental Protection Agency
Region 5
Office of Science, Ecosystems, and Communities
77 W Jackson Boulevard (B-19J)
Chicago, IL 60604
Telephone: (312)353-5814
E-mail: macarus.david@epa.gov
Cynthia Martin
Indiana Department of Environmental
Management
Office of Water Quality
100 N Senate Avenue (65-40)
Indianapolis, IN 46204-2251
Telephone: (317)308-3081
E-mail: cmartin@idem.in.gov
Carmen Maso
U.S. Environmental Protection Agency
Region 5
Office of Sustainable Ecosystems and
Communities
77 W Jackson Boulevard (B-19J)
Chicago, IL 60604
Telephone: (312)886-1070
E-mail: maso.carmen@epa.gov
-------
Charles Maurice
U.S. Environmental Protection Agency
Region 5
Office of Research and Development
Superrund Division
77 W Jackson Boulevard (SR-4J)
Chicago, IL 60604
Telephone: (312)886-6635
E-mail: maurice.charles@epa.gov
James Mayka
U.S. Environmental Protection Agency
Region 5
Office of Research and Development
Superrund Division
77 W Jackson Boulevard (SR-6J)
Chicago, IL 60604
Telephone: (312)353-9229
E-mail: mayka.james@epa.gov
Barbara Mazur
U.S. Environmental Protection Agency
Region 5
Office of Science, Ecosystems, and Communities
77 W Jackson Boulevard (B-19J)
Chicago, IL 60604
Telephone: (312)886-1491
E-mail: mazur.barbara@epa.gov
Brent McCloskey
Chesapeake Research Consortium
Chesapeake Bay Program
410 Severn Avenue, Suite 109
Annapolis, MD 21403
Telephone: (410) 267-9830
E-mail: mccloskey.brent@epa.gov
John McCombs
National Oceanic and Atmospheric
Administration
Coastal Services Center
2234 S Hobson Avenue
Charleston, SC 29405
Telephone: (843)740-1164
E-mail: john.mccombs@noaa.gov
Cary McElhinney
U.S. Environmental Protection Agency
Region 5
Water Division
Ground Water and Drinking Water
77 W Jackson Boulevard
Chicago, IL 60604
Telephone: (312)886-4313
F-mail:
E J. McNaughton
Indiana Department of Environmental
Management
Geographic Information Systems
100 N Senate Avenue
Indianapolis, IN 46206
Telephone: (317)232-8197
E-mail: emcnaugh@idem.in.gov
Matthew Nicholson
U.S. Environmental Protection Agency
Region 3
1650 Arch Street
Philadelphia, PA 19103
Telephone: (401) 782-9655
E-mail: nicholson.matt@epa.gov
Marie Oliver
U.S. Environmental Protection Agency
Region 5
Waste Pesticides and Toxics Division
77 W Jackson Boulevard (DM-7J)
Chicago, IL 60604
Telephone: (312)886-6339
E-mail: oliver.marie@epa.gov
John Perrecone
U.S. Environmental Protection Agency
Region 5
Office of Science, Ecosystems, and Communities
77 W Jackson Boulevard (B-19J)
Chicago, IL 60604
Telephone: (312)353-1149
E-mail: perrecone.john@epa.gov
Abigail Popp
U.S. Environmental Protection Agency
Region 5
Great Lakes National Program Office
77 W Jackson Boulevard
Chicago, IL 60604
Telephone: (312)886-8022
E-mail: popp.abigail@epa.gov
Pranas Pranckeviclus
U.S. Environmental Protection Agency
Great Lakes National Program Office
77 W Jackson Boulevard (G-17J)
Chicago, IL 60604
Telephone: (312) 353-3437
E-mail: pranckevicius.pranas@epa.gov
-------
Ted Prescott
Illinois Environmental Protection Agency
Bureau of Land
1021 N Grand Avenue, East
Springfield, IL 62794
Telephone: (217) 524-3511
E-mail: ted.prescott@epa.state.il.us
Jane Ratcliffe
U.S. Environmental Protection Agency
Region 5
Waste Pesticides and Toxics Division
77 W Jackson Boulevard (DM-7J)
Chicago, IL 60604
Telephone: (312)886-7449
E-mail: ratclirTe.jane@epa.gov
James Robb
Indiana Department of Environmental
Management
Office of Water Quality
100 N Senate Avenue (65-42)
Indianapolis, IN 46204-2251
Telephone: (317)233-8802
E-mail: jrobb@idem.5n.gov
David Rockwell
U.S. Environmental Protection Agency
Region 5
Great Lakes National Program Office
Monitoring Indicators and Reporting Branch
77 W Jackson Boulevard (G-17J)
Chicago, IL 60604
Telephone: (312)353-1373
E-mail: rockwell.david@epa.gov
Zenny Sadlon
U.S. Environmental Protection Agency
Region 5
Office of Information Services
77 W Jackson Boulevard (MI-9J)
Chicago, IL 60604
Telephone: (312)886-6682
E-mail: sadlon.zenny@epa.gov
Jonathan Schweizer
U.S. Environmental Protection Agency
Region 5
Water Division
National Pollution Discharge Elimination System
Programs Branch
77 W Jackson Boulevard (WN-16J)
Chicago, IL 60604
Telephone: (312)886-0211
E-mail: schweizef.jonathaTi@epa.gov
YiShi
Michigan State University
Institute of Water Research
Manly Miles Building, Room 115
1405 S Harrison Road
East Lansing, MI 48823-5243
Telephone: (517)353-3742
E-mail: shiyil@msu.edu
Jatlnder Singh
U.S. Environmental Protection Agency
Region 5
Superfund Division
77 W Jackson Boulevard (SR-6J)
Chicago, IL 60604
Telephone: (312)353-6756
E-mail: singh.jatinder@epa.gov
Betsy Smith
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
109 TW Alexander Drive (MD 243-05)
Research Triangle Park, NC 27711
Telephone; (919) 541-0620
E-mail: smidi.betsy@epa.gov
Grace Smith
U.S. Environmental Protection Agency
Region 2
290 Broadway
New York, NY 10007
Telephone: (212)637-3589
E-mail: smith.grace@epa.gov
Karen Stainbrook
Illinois Natural History Survey
Center for Aquatic Ecology
Illinois Department of Natural Resources
9511 Harrison Street
Des Plaines, IL 60016
Telephone: (847)294-4134
E-mail: kms@uiuc.edu
Michael Tenenbaum
Gun Lake Tribe
1743 142nd Avenue, Suite 7
PO Box 218
Dorr, MI 49323
Telephone: (616) 681-8830, ext 305
E-mail: mbtenenbaum@mbpi.org
-------
Lee Walston
U.S. Environmental Protection Agency
Region 5
Superfund Division
Field Environmental Decision Support Team
77 W Jackson Boulevard (SRF-5J)
Chicago, IL 60604
Telephone: (312)886-7196
E-mail: walston.leroy@epa.gov
Claudia Walters
U.S. Environmental Protection Agency
Office of Research and Development
Office of Science Policy
Ariel Rios Building (8104R)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
Telephone: (202)564-6762
E-mail: walters.claudia@epa.gov
Bonnie Weinbach
U.S. Environmental Protection Agency
Region 5
Air and Radiation Division
77 W Jackson Boulevard (AR-18J)
Chicago, IL 60604
Telephone: (312)886-0258
E-mail: weinbach.bonnie@epa.gov
James Wickham
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
109 TW Alexander Drive (E243-05)
Research Triangle Park, NC 27711
Telephone: (919) 541-3077
E-mail: wickham.james@epa.gov
Dorsey Worthy
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Environmental Sciences Division
Landscape Characterization Branch
109 TW Alexander Drive
Research Triangle Park, NC 27711
Telephone: (919) 541-3075
E-mail: worthy.dorsey@epa.gov
Steve Young
U.S. Environmental Protection Agency
Office of Environmental Information
Office of Information Analysis and Access
Ariel Rios Building (2841T)
1200 Pennsylvania Avenue, NW
Washington, DC 20460
Telephone: (202)566-0608
E-mail: young.steve@epa.gov
Richard Zdanowicz
U.S. Environmental Protection Agency
Region 5
Water Division
Ground Water and Drinking Water Branch
77 W Jackson Boulevard (WQ-16J)
Chicago, IL 60604
Telephone: (312)886-1502
E-mail: zdanowicz.richard.@epa.gov
Glynis Zywicki
U.S. Environmental Protection Agency
Region 5
77 W Jackson Boulevard (P-19J)
Chicago, IL 60604
Telephone: (312)886-4571
E-mail: zywicki.glynis@epa.gov
Contractor Support
Angela Hays
The Scientific Consulting Group, Inc.
656 Quince Orchard Road, Suite 210
Gaithersburg, MD 20878
Telephone: (301)670-4990
E-mail: ahays@scgcorp.com
Kristen LeBaron
The Scientific Consulting Group, Inc.
656 Quince Orchard Road, Suite 210
Gaithersburg, MD 20878
Telephone: (301)670-4990
E-mail: klebaron@scgcorp.com
-------
Attachment 3
EPA Office of Science Policy
Regional Science Workshop on Remote Sensing and
Landscape Characterization Evaluation Summary
-------
Regional Science Workshop on Remote Sensing and Landscape Characterization
U.S. EPA Region 5
Lake Michigan Room, 12th Floor
Valdas V. Adamkus Environmental Resource Center
77 W Jackson Boulevard
Chicago, EL 60604
November 1-3,2005
EVALUATION SUMMARY
EPA is one of the U.S. Federal Agencies that is promoting the multi-national Global Earth Observation
System of Systems (GEOSS). When developed, this system is expected to deliver timely information on
conditions of the Earth based on databases shared globally by more than 30 countries. It is expected that
the availability of this global information will be valuable to assist in planning renewable and non-
renewable resources and large-scale environmental changes. Regional staff remain skeptical that GEOSS
can provide assistance in enforcement, pollution prevention, and sustainability initiatives. Some of the
data to be used in GEOSS, however, will be derived from remote sensing technologies, many of which
have application to local conditions. This workshop addressed the tools that may assist the Regions,
states, and tribes in learning how to access and use remote sensing tools in landscape characterization in
ways that can assist in Regional work.
The purpose of the workshop was to bring available remote sensing technology to Regional staff and
illustrate how adding remote sensing to ground-based data enhances landscape characterization that
guides Regional enforcement, pollution prevention, and long-term sustainability efforts.
An evaluation of the workshop was conducted to elicit information from attendees regarding the
workshop organization and logistics, the information presented, and potential improvements in future
workshops. A total of nine elements were developed for the evaluation form (see Appendix A). All nine
elements allowed attendees to rate the sessions and elements of the meeting on a scale of 1 (poor) to 4
(excellent). A fill-in-the-blank section allowed the attendees to state the most and least informative
sessions. Three open-ended questions allowed attendees to provide any other comments or suggestions
for future workshops. Attendees also could provide additional comments regarding each of these
questions. A summary of the evaluation findings is provided below. Results received for each evaluation
question, including ratings and comments, follow the summary of findings.
Part I. Summary of Findings: Sessions
• Of the 64 meeting participants, 21 completed the evaluation questionnaire for an overall response rate
of 33%
• The attendees rated their overall impression of the meeting. Of 21 respondents, 11 (52.4%) provided
a rating of 3, and 10 (47.6%) provided a rating of 4, for an average rating of 3.48.
• The attendees rated the presentations. Of 21 respondents, 8 (38%) provided a rating of 3, and 13
(62%) provided a rating of 4, for an average rating of 3.62.
• The attendees rated the panel discussions. Of 9 respondents, 1 (11.1%) provided a rating of 1,2
(22.2%) provided a rating of 3, and 6 (66.7%) provided a rating of 4, for an average rating of 3.56.
-------
The attendees rated the round-table discussions. Of 11 respondents, 1 (9.1%) provided a rating of 2,3
(27.3%) provided a rating of 3, and 7 (63.6%) provided a rating of 4, for an average rating of 3.55.
Responses to Evaluation Questions
Part I. Summary of Findings: Sessions
Question 1: Overall Impression of Workshop
Rating: Number of Responses: 21
Highest Rating: 4
Lowest Rating: 3
Average Rating: 3.48
Question 2: Presentations
Rating: Number of Responses: 21
Highest Rating: 4
Lowest Rating: 3
Average Rating: 3.62
Questions: Panel Discussions
Rating: Number of Responses: 9
Highest Rating: 4
Lowest Rating: 1
Average Rating: 3.56
Question 4: Round Table Discussion
Rating: Number of Responses: 11
Highest Rating: 4
Lowest Rating: 2
Average Rating: 3.55
Part II. Summary of Findings: Meeting Elements
• The attendees rated the workshop materials. Of 17 respondents, 1 (5.9%) provided a rating of 1, 5
(29.4%) provided a rating of 2, 9 (52.9%) provided a rating of 3, and 2 (11.8%) provided a rating of
4, for an average rating of 2.70.
• The attendees rated the registration process. Of 17 respondents, 4 (23.5%) provided a rating of 3, and
13 (76.5%) provided a rating of 4, for an average rating of 3.76.
• The attendees rated the hotel accommodations. Of 10 respondents, 10 (100%) provided a rating of 4,
for an average rating of 4.00.
-------
The attendees rated the helpfulness of onsite support staff. Of 15 respondents, 6 (40%) provided a
rating of 3, and 9 (60%) provided a rating of 4, for an average rating of 3.60.
The attendees rated the meeting room. Of 17 respondents, 6 (35.3%) provided a rating of 3, and 11
(64.7%) provided a rating of 4, for an average rating of 3.65.
Part II. Summary of Findings: Meeting Elements
Question 1: Workshop Materials
Rating: Number of Responses: 17
Highest Rating: 4
Lowest Rating: 1
Average Rating: 2.70
Question 2: Registration Process
Rating: Number of Responses: 17
Highest Rating: 4
Lowest Raring: 3
Average Rating: 3.76
Question 3: Hotel Accommodations
Rating: Number of Responses: 10
Highest Rating: 4
Lowest Rating: N/A
Average Rating: 4.00
Question 4: Helpfulness of Onsite Support Staff
Rating: Number of Responses: 15
Highest Rating: 4
Lowest Rating: 3
Average Rating: 3.60
Question S: Meeting Room (sound, space, lighting)
Rating; Number of Responses: 17
Highest Rating: 4
Lowest Rating: 3
Average Rating: 3.65
The most informative session was:
• Hard to say.
• Great Lakes Challenge.
• The overview by Ross Lunetta was very helpful to understanding subsequent presentations.
• All.
-------
Population Estimating. It helps us understand potentially effected populations over time at various
sites. Also, Risk Detection—has real-time potential for inspectors. 2. Multi-spectral onto LIDAR.
Roundtable discussion. Identified needs and potential solutions.
1. Ross Lunetta on QA. Eye opening on QA of Remote Sensing—I learned a lot. 2. Roger Gauthier
on Great Lakes Observation System (GLOS). Content rich on GLOS—great possibilities for
partnership.
Steam Characterization. Opened wide range of opportunities and possible applications.
All very helpful.
Ric Lopez, Budhendra Bhaduri, and Robert Hall. [Their] presentations gave me great ideas!
Budhendra Bhaduri; Barry Feldman because it was the most "nuts and bolts" talk; and Robert Hall's
talk.
All very good. Practical applications good.
Dr. Bhaduri's. It illuminated the high value of the synergy of remote sensing data and census data.
Days 1 and 3.
Transgenic Crop Characterizations.
Remote Sensing for Day Versus Night Population Estimating, Budhendra Bhaduri, and Quantifying
Stream Attributes Using LIDAR and Hyperspectral Imagery, Robert Hall. Clear, cutting-edge
technology.
Budhendra Bhaduri—he was an excellent speaker and the information was well presented, even for
someone not in the field.
Wednesday a.m.—Great Lakes analysis, including Milwaukee. Most relevant to my work (Region 5
Water/NPDES Permits).
Ross Lunetta's presentations.
All sessions were good.
The least informative session was:
None.
Panel Discussion.
None.
They were all good.
None.
John Glaser—the agricultural component is not of use to my particular needs.
John Glaser—Needs of Large Area Monitoring for Pesticides/Transgenic Crops. Too much jargon
and technical terms, does not relate to my work.
Some of the calks were so regional and national in scope.
Day 2.
Remote Sensing Overview.
First session, Day 3—Crop Identification. All were informative in some way. However, Day 3 Crop
Identification session was somewhat hard to follow—confusing graphics.
• All good.
General Comments:
What did you learn that you are most likely to take back and share with staff?
• Connections.
• I will be connecting scientists in the Office of Research and Development's Research Triangle Park
Laboratory with some of the speakers.
-------
• Great workshop—all useful information.
• The ability to track contaminants (Milwaukee-Chicago) in Lake Michigan.
• Contacts; networking.
• The diversity of remote sensing capabilities, problems with the loss of useful platforms (i.e., Landsat),
census applications, and the need for a portal/access to information that many cannot afford.
• Hearing what the states and tribes need.
• That Ross Lunette's process to derive land cover change from MODIS may serve us and that he is
looking for collaboration opportunities.
• That C-CAP data was available for Great Lakes. The Barry Feldman talk we can likely use right
now.
• The need to share data over a broader audience.
• The variety and depth of tools and services available, much of it free, and the quality and passion of
the practitioners in this field.
• Some of the accuracy measures and other analytical and statistical data treatments. The evolving use
ofLFDAR.
• Real-time monitoring.
• That remote sensing can provide maps that can clearly show where problems are, and possibly with
enough data, the nature of problems and changes over time. Information on EPA and other Federal
Web sites can be pointed out to others.
• Datasets available for use in applications.
• I learned a lot about the various remote sensing tools.
What follow-up activities from this workshop would you recommend or like to participate in?
• Applications specific to Region 5 Tribal Environmental Staff. Making available FUNDING
SOURCES for remote sensing applications and data acquisition. Suggest RTOC/GAP meetings as
good starting points. See John Haughland to implement.
• Lots of ideas—who and how will the issues/needs be followed up by the appropriate people/groups?
• A series of workshops broken down by satellite, airborne, and ground-based data collection systems.
What's new, what works.
• Development of interested party discussion group for the use of remote sensing approaches to EPA-
related issues.
• I just want to learn and understand more about remote sensing.
• Possibly more detailed (maybe "hands on" in computer room) on specific types of remote sensing.
Landscape characterization (e.g., water pollution-related for those in water programs).
What could be done to make future workshops on this topic more informative and/or productive?
• Broader audiences.
• Ask speakers to show how their "tools" address specific Region/state/local health and/or
environmental problems.
• Note attendees from states, tribes, and local communities.
• Will depend upon knowledge and skills of participants. Assume little to no tribal
capacity/knowledge. Offer beginning, followed by intermediate, followed by more advanced. Move
toward creation of tribal users group that can share information, sources, data, cooperative ventures,
etc.
• Use the needs of states/tribes to formulate next workshop.
• More talks on the local scare and more practical application. It would have helped to have the slide
handouts before each talk to be able to take notes on each slide handout, rather than trying to quickly
write notes as fast as possible.
• Maybe breakout sessions by media.
-------
• It seems like a lot of the information was jargon-filled, and several people commented that it went
over their heads. More basic information and/or a clearer mission for die workshop would have
helped.
• Include all handouts, including hardcopies of slides in notebook and identified by speaker. Also
provide list of related Web sites and/or other references.
• Give specifics on how data can be distributed to states.
Additional Comments:
• Request information from states and Regions to formulate meeting agenda.
• Good feedback on the forum. Have handouts available prior to talks.
• Some handouts hard to read—size and quality.
• Came into this with a desire to learn more about how to provide land cover/land use data for the tribe
I work for. Found there are MANY ways this can be done; techniques continue to morph and
advance; various types of information are already available; there is much more for us to do!!
• Great workshop—great start.
• Enjoyed it.
• It would have been nice if speakers would not race though Web sites. It was hard to get them all
down. Side note: not everyone drinks coffee. Hot tea or bottled water would have been nice.
• Many slides too "busy" and with type in captions and labels hard to read. Perhaps in general, fewer
and more readable slides would help.
• This workshop was a good opportunity to meet the scientists who are leading the research on remote
sensing.
-------
Page Intentionally Blank
-------
Page Intentionally Blank
-------
Page Intentionally Blank
------- |