Remote Sensing Applications

Remote Sensing Applications at the
State and Local Level

Report of a User Workshop
February 23, 2001
Lamont Doherty Earth Observatory
Palisades, NY
Sponsored by the Socioeconomic Data and Application Center (SEDAC)

Center for International Earth Science Information Network (CIESIN)
Columbia University
Report of the Workshop on Remote Sensing Applications at the State and Local Level
The User Workshop on Remote Sensing Applications at the State and Local Level was
sponsored by the Socioeconomic Data and Application Center (SEDAC) of the Center for
International Earth Science Information Network (CIESIN) at Columbia University.
The Workshop home page can be found at
http://sedac.ciesin.columbia.edu/remote_sens/
The Center for International Earth Science Information Network (CIESIN) was
established in 1989 as an independent non-governmental organization to provide
information that would help scientists, decision-makers, and the public better understand
their changing world. In 1998, CIESIN relocated from its original home in Michigan to
become part of Columbia University’s Earth Institute.
CIESIN focuses on applying state-of-the-art information technology (e.g., computer

science, data archiving, geographic information systems, and remote sensing) to
interdisciplinary problems at the intersection of human and environmental systems.
CIESIN at Columbia University

P.O. Box 1000
61 Route 9W
Palisades, NY 10964 USA
Tel. +1-845-365-8988
Fax: +1-845-365-8922
Web: http://www.ciesin.columbia.edu
Report edited by Francesca Pozzi, CIESIN. Work supported by the U.S. National
Aeronautics and Space Administration, Goddard Space Flight Center under contract
NAS5-98162. The opinions expressed here are those of the participants and editor and are
not necessarily those of CIESIN or NASA.
Copyright © 2001 The Trustees of Columbia University in the City of New York
ii
Executive Summary
The Workshop on Remote Sensing Applications at the State and Local Level was
organized by CIESIN’s Socioeconomic Data and Application Center (SEDAC) to address
issues of availability and the needs for remotely sensed data products by local users. The
workshop took place at Columbia’s Lamont-Doherty Earth Observatory, in Palisades, NY
and was attended by about twenty representatives from local and state governments and
non-governmental organizations in New York and New Jersey.
The objective was to solicit input from potential users of geo-referenced data products
derived from remotely sensed imagery, in terms of specific products needed for on-going
activities and the exploration of new applications.
The increasing availability and accessibility of new technology and data for local
governments and agencies are proving helpful for day-to-day decision-making processes.
In particular, remotely sensed data and Geographic Information Systems have been
increasingly used together for a vast range of applications, ranging from land use/land
cover mapping to emergency management to characterization and monitoring of
environmental and human health conditions.
SEDAC has been working with five sets of Landsat 7 ETM+ images covering parts of
New York, New Jersey and Connecticut, acquired during the fall 1999 and the winter
2000. Using the above images, SEDAC’s staff have explored ways to develop new and
innovative products that might be useful for a wide range of applications within local and
state governments. Some of these prototype products were presented at the workshop in
order to get feedback from potential users about general usefulness, specific applications
of remote sensing products for local governments, and specific issues users would like to
see addressed by these products.
The two prototype data products presented were (i) a tailored land cover/land use
classification and (ii) a greenness map. The discussion focused on usefulness of such
products for a variety of applications, such as landscape assessment, environmental and
health conditions monitoring, urban sprawl and urban/vegetation change studies. This led
to discussions of specific applications that local governments and agency are currently
involved with and interested in using, with particular emphasis on different sensors and
their resolutions.
Using an evaluation form distributed the same day, participants provided helpful
comments in terms of specialized classifications they would like to have, time frames for
image acquisition, and data processing and integration with other data sets. Although
there was an agenda item on mechanisms for data sharing and models for collaborative
work between governments and organizations, those topics were not discussed in depth,
given the differences in interests that emerged during the workshop and the possible
difficulties in coordinating such different activities. Such discussions seemed more likely
to develop in the future through individual collaborations rather than through a broader
group or ‘consortium’.
iii
TABLE OF CONTENTS
1 Introduction .................................................................................................................... 1
2 Prototype Data Products Discussion ............................................................................ 2

2.1 Tailored land cover/land use classification. ....................................................... 2
Background ................................................................................................................. 2
Comments.................................................................................................................... 7
2.2 Vegetation Fraction Index and Urban/Vegetation Change Detection................ 8
Background ................................................................................................................. 8
Comments.................................................................................................................. 12
3 Other Applications and Integration with Relevant Data sets .................................. 13

3.1 Applications....................................................................................................... 13
Landscape Assessment.............................................................................................. 13
Air Quality................................................................................................................. 14
Water quality............................................................................................................. 14
Floodplains mapping and Emergency Management ................................................. 14
Health Applications................................................................................................... 15
3.2 Integration with Relevant Data sets .................................................................. 16
Annex 1. Workshop Agenda........................................................................................... 17
Annex 2. List of Participants.......................................................................................... 18
Annex 3. Background Information................................................................................ 20
References ........................................................................................................................ 26
iv
1 Introduction
The Workshop on Remote Sensing Applications at the State and Local Level was
organized by CIESIN’s Socioeconomic Data and Application Center (SEDAC) to address
issues of availability and needs for remotely sensed data products among local users.
Specifically about twenty representatives from local and state governments and non-
governmental organizations in New York and New Jersey attended the workshop to
discuss the development of tailored applications using remotely sensed imagery. SEDAC
developed prototype data products using Landsat 7 ETM+ imagery, such as a tailored
land use classification and a greenness map (the latter being a simpler version of the
Vegetation Fraction Index calculated for New York City by Chris Small).
The objective was to solicit input from potential users of geo-referenced data products
derived from remotely sensed imagery, in terms of specific products needed for on-going
activities and exploration of new applications. The possibility of forming local and
regional collaborations to purchase and share imagery and other data resources was also
on the agenda (see Annex 1 for the Agenda and Annex 2 for the Participants list).
The increasing availability and accessibility of new technology and data for local
governments and agencies is proving helpful in day-to-day decision-making processes. In
particular remotely sensed data and Geographic Information Systems have increasingly
been used together for a vast range of applications, spanning from land use/land cover
mapping to emergency management to characterization and monitoring of environmental
and human health conditions.
The higher spatial and spectral resolutions, more frequent coverage and increased
availability of new sensors will bring remote sensing to a more accessible level within
local and state governments and help them deal with several issues in regional planning,
resource management, public health and environmental protection.
The workshop’s objectives were:
• To examine the prototype data products and discuss the possible uses for local
application in terms of Land Cover/Land Use and Vegetation change detection;
• To discuss other potential applications, such as emergency management, flood

mapping, air and water quality monitoring and related issues of scale and timing;
• To discuss other relevant data sets and their integration with remotely sensed data;
• To discuss other satellite data products and possible mechanisms for data sharing.
This reports provides a summary of the workshop presentations and discussions. Section
2 describes the discussion on prototype data products, including background material and
comments from workshop participants. Section 3 summarizes the discussion on specific
applications and integration with other data sets. Finally, Annex 3 provides background
information on Landsat 7 ETM+ and other remote sensing satellite and sensors.
1
2 Prototype Data Products Discussion

SEDAC has been working with five sets of images covering parts of New York, New
Jersey, and Connecticut, acquired during the fall of 1999 and the winter of 2000. The
images were collected by Landsat 7 using the Enhanced Thematic Mapper Plus (ETM+)
sensor (see Annex 3 for more information on Landsat and other satellites and sensors).
The spatial resolution (ground cell or pixel size) varies between 15 and 60 meters and
allows detection and mapping of quantities such as vegetation abundance and health,
presence of standing water, forest clearing, agricultural usage and other land cover
changes.
Using the above images we explored several ways of making use of the high-resolution
panchromatic band and the improved thermal band, and of developing new products.
Some of these prototype products were presented at the workshop, in order to get
feedback from potential users about general usefulness, specific applications of remote
sensing products for local governments, and specific issues users would like to see
addressed by these products.
The advantage of having such user-specific products is that they will be developed
according to local government needs, they will be compatible with higher resolution data
sets already collected and available at the local level, and they will allow a better
understanding of dynamics and features that can be integrated in decision-making
processes.
A classic example of such integration would be a land cover/land use change study: the
resolution of the Landsat 7 data does not allow for detailed discrimination between
features in the way that aerial photography or field surveys do. However, the availability
of multispectral data can help better identify certain features and their condition (i.e.,
vegetation health monitoring). The availability of repeat coverage over a certain area will
then permit the production of land cover maps for different periods and thus aid change
detection studies. Changes that might be of interest can range from short-term
phenomena like flooding or snow cover to long-term phenomena like urban sprawl or
deforestation. Vegetation health monitoring can also be performed using images from
different seasons. This could be useful in urban vegetation studies or to relate air
pollution and vegetation health.
At the workshop we mainly discussed two prototype products potentially useful for local
governments: (i) a tailored land cover/land use classification and (ii) a greenness map.
2.1 Tailored land cover/land use classification.
Background
Using the Landsat 7 ETM+ scene from September 1999 (Figure 1, showing Paterson, NJ
and surrounding areas in visible/infrared false color), a maximum-likelihood
classification was performed to produce the classified image in Figure 2.
2
The maximum-likelihood approach is one of the methods used in supervised

classification. This type of classification is used to cluster pixels into classes
corresponding to user-defined “training areas” or “region of interest” (groups of pixels
that represent areas or materials that the user wants to have mapped in the final product).
The first step in the supervised classification is to identify representative training areas
and to develop a numerical description of the spectral attribute for each land cover type
of interest. This is followed by the classification process in which each pixel is assigned
to the land cover class it is most likely to belong to. The maximum likelihood method
requires a large number of computations to classify each pixel, making it time and
memory consuming, but it is considered one of the most accurate classification
algorithms.
The classification of the Landsat scene was performed using only 6 classes of land
use/land cover to show the possibility of using a very simple classification scheme that
can be used for many different applications, in particular in the urban/suburban context.
Such applications might include mapping and monitoring of new developments (housing,
roads, commercial development, etc.), golf courses or changes in density of vegetation
and urban areas.
3
Report of Workshop on Remote Sensing Applications at the State and Local Level
Figure 1. Landsat 7 ETM+ (9/23/99). The image is shown in visible/infrared false color (Red= Band 7,
Green= Band 4, Blue= Band 2).
4
Figure 2. Maximum Likelihood Classification. Legend:

Red: Infrastructure (high intensity residential, commercial, transportation)
Yellow: Low Intensity Residential
Light Green: Urban and Recreational Grasses
Green: Forests
Blue: Water
Coral: Bare Soil, Rock
The correspondence between the unclassified image and the classified one is relatively
good. Given the exploratory nature of this product, no ground truthing was performed
and, consequently, no statistical accuracy tests were conducted based on control areas on
the ground. This was meant to be a first attempt at establishing whether such ‘tailored’
products can be developed and to show an example of simplified classifications.
Looking at the “Regions of Interest” statistical distribution, it is possible to get a sense of

their separability (i.e., how well the classes can be distinguished from each other). The
classes that are best distinguished are Water and Grass, Water and Soil, Water and Low
Intensity Residential, and Grass and Soil. The two most poorly separated are:
Infrastructure and Low Intensity Residential, as expected.
5
It is important to remember that this simply gives an idea of which classes are easier to
separate based only on their spectral responses in the different bands and does not ensure
a good result during the classification process.
The advantage of a user-specific classification, compared with other available products,

such as the recently released USGS National Land Cover Data set (see Figure 3), lies
within the word “tailored”: such classification will be produced based on inputs from
local governments, according to their needs and potential uses of the product. The classes
can be identified and selected based on specific applications and by integrating relatively
coarse-resolution data, such as Landsat data, with high-resolution data, such as aerial
photography—data already available in many counties. The major advantage would be
the possibility of having a regular updating of the product thanks to the availability of
repeat coverage, leading to appropriate land cover and land use change products.
Figure 3. USGS National Land Cover Data Set. (USGS, National Land Cover Characterization
Program, National Land Cover Data (NLCD) Set, 1999).
6
Comments
Workshop attendees expressed considerable interest in the ideas of developing tailored
classification schemes and updating land use/land cover maps.
The need for user-tailored classifications was emphasized both by local governments
(county level) and by regional and state groups. In the case of county governments, where
detailed land use maps may already be available, we discussed the idea of integrating
their high-resolution data with satellite-derived products in order to refine existing maps
and to get more diverse information. Such information might be related to vegetation or
the possibility of aggregating multiple classes (based on the classification results) into
fewer classes according to specific uses and applications. Producing an accurate, useful
classification scheme for local governments using Landsat 7 data would undoubtedly
require a large amount of work, but once the methodology and the final products are in
place it could represent a significant improvement for data acquisition and integration in
decision-making processes.
As for the regional and state level, given the difference in scale, the development of a
tailored classification scheme will require less ancillary data and could be more easily
automated. In this case, participants expressed the need for having a comparable set of
land use/land cover classes for assessing landscape change, in particular between forest,
agriculture, and development.
These ideas led the discussion to ‘change detection’ studies and to the possibility of
having regular updates of the classified images. Although limitations in the spatial
resolution of the classified image were recognized, the possibility of having regular
updates of such products was particularly appealing. Even county governments already
provided with detailed maps liked the idea of finding a way to get their data and maps
updated more regularly, such as through aggregation of data from different sources
(satellite, aerial photography, ground survey, etc.). An example of a way to improve the
resolution and make use of the repeat coverage is to calibrate the classification derived
from satellite imagery with information from aerial photography and ground data, and
then use that calibrated image to perform the classification on other images, looking both
back and forward in time. This procedure minimizes the costs of collecting ancillary data
to validate the classification in the future and therefore reduces both the costs and the
time for the classification process itself.
Other applications that emerged during the discussion were related to integration of land
use classification maps with other socioeconomic and demographic data, as further
discussed in Section 3.
7
2.2 Vegetation Fraction Index and Urban/Vegetation Change Detection
Background
Most major metropolitan areas and neighboring suburban settlements are facing problems
of urban sprawl, loss of natural vegetation and open space, and drastic change in land
cover and land use. For years, land mapping at the county level has relied on aerial
photography and field surveys because of the high spatial resolution required and the
high level of details that can be obtained. However, satellite imagery may become
increasingly important in identifying and monitoring soil conditions and various related
land characteristics, such as vegetation and crops. Monitoring is one of the key uses of
remote sensing, due to the availability of repeat coverage and associated high spatial and
spectral resolution. Remote sensing is extensively used in monitoring changes in natural
and semi-natural areas and, to a lesser extent, the effects of pollution on trees, soil, and
water.
In areas such as the New York City metropolitan area, where the effects of urban sprawl
are of particular concern, remote sensing can be used to monitor urban/vegetation
changes and be integrated with higher resolution data available at the county level. Also,
new products and methodologies, such as the ones presented at the workshop, can be
developed for a vast range of applications and could potentially be updated regularly
(annually or seasonally) according to user needs.
SEDAC has been exploring possible products that might help local governments display
and monitor such changes and that can be used as basic tools in day-to-day decision-
making processes. These products were presented in the morning session.
One such product, potentially very useful in studying urban sprawl and urban and
suburban dynamics, is the Vegetation Fraction Index.
During his overview, Chris Small briefly described the results of his study on New York
City vegetation (shown in Figure 4). Estimates of vegetation fractions have been obtained
through a linear spectral mixing model. The model is based on the assumption that the
reflectance of each pixel of the image is a linear combination of the reflectance of each
‘material’ present within the pixel. This allow one to determine the relative abundance of
the ‘materials’ based on their spectral characteristics. In the case of New York City, three
main components have been recognized to describe the spectral variance in the scene:
low albedo (water, shadow, and roofing), high albedo (cloud and roofing), and
vegetation. The model has proven to be very accurate: a quantitative validation of
Landsat-derived vegetation estimates with vegetation measurements from aerial
photography showed an agreement to within 10% for vegetation fractions greater than
20%, across the full range of vegetation abundance (Small, 2001).
Such a product allows one to identify the green areas and quantify the proportions of
vegetation versus urban land in an immediate and easily interpretable way.
8
Figure 4. New York Urban Vegetation, estimated from Landsat 7 imagery, 9/23/99. Vegetation
fraction showed above range from sparse (blue <10%) to dense (red > 90%). Black areas contain
negligible vegetation or are in shadow (Small, 2001).
9
An even simpler model is a Greenness map, obtained by performing a Tasseled Cap

Transformation on the original Landsat data.
The Tasseled Cap Transformation on Landsat TM data performs an orthogonal

transformation of the original data into a new three-dimensional space, consisting of a
“Brightness” index, a “Greenness” index, and a third component related to soil features.
In the case of Landsat MSS data, the Tasseled Cap Transformation generates four
indices: a Soil Brightness Index, a Green Vegetation Index, a Yellow Stuff Index, and a
Non-such Index, related to atmospheric effects (ENVI, 2000).
In order to visualize the differences between vegetation and non-vegetated areas, such as
urban and water, only the Greenness Index has been displayed using a logarithmic green
and white scale, where the deeper green corresponds to more vegetation. Figure 5 and
Figure 6 show the Greenness Index for a portion of southern New York, obtained using
Landsat MSS data from 1987 and 1991, respectively.
Performing the transformation on data obtained in two different periods permits showing
changes in the proportion of urban and forested areas. Some of the most visible
differences are indicated by the red arrows in the 1991 image. Note that the spatial
resolution for these images (Landsat MSS) is 80 x 80 meters; therefore, many minor
changes cannot be detected.
Figure 5. Greenness map obtained from Landsat MSS imagery, 06/10/87.
10
Figure 6. Greenness map obtained from Landsat MSS imagery, 06/21/91.
The most striking change is the road in the lower left corner, not visible in the 1987
scene, but clearly identifiable in the 1991 scene. That road is the last section of Interstate
I-287 to have been built, between I-80 and I-87. The whole Interstate was completed in
1993. The area in the top left corner (Sugar Loaf, NY) shows a slight reduction in the
amount of vegetation, which cannot be uniquely attributed to anything without further
study. Another area showing change is the one in the bottom right corner (in Westchester
County), where a minor urban enlargement is clearly recognizable.
It must be said that the previous two greenness maps show vegetation and urban areas
mainly in a qualitative way. Looking at the greenness histograms reported in Figure 7 can
give an idea of the distribution of pixels according to the greenness values, although
properly quantifying such changes requires a quantitative method, like the Vegetation
Fraction model above described.
11
Figure 7. Greenness histograms for the 1987 and 1991 scenes, respectively.
The peak in the negative values corresponds to water, while the peak in the positive
values (around 60) corresponds to vegetation. The broad range of numbers in between
corresponds to a variety of combinations of vegetation and urban land cover. The shape
of the curve is not dramatically different for the two scenes; only the vegetation peak
appears to be lower (fewer pixels for ‘pure’ vegetation) in 1991, and there seems to be a
larger ‘mixed’ area for values between 30 and 50. However, as already mentioned, this
should not be interpreted as a quantitative representation of vegetation change.
Comments
Greenness maps and the consequent vegetation/urban change detection studies are
straightforward to understand and interpret. For this reason, these products did not
generate the same level of comment and feedback as the land use/land cover maps,
although they might find applications in a variety of fields, such as the study of urban
sprawl.
The resolution of the Landsat data limits their use to tracking vegetation changes at a
broader scale (regional, rather than county level) and therefore did not elicit too much
interest among the local government representatives. Participants from county
governments seemed more interested in changes between different land cover classes
than just vegetation vs. urban land cover.
Nevertheless, the potential importance of such tools and products in monitoring

vegetation health should be noted, given the availability of repeat coverage and of data
collected at different wavelengths. This type of information could be integrated with
higher resolution data, such as from IKONOS (increased spatial resolution: 4 m in the
multispectral, 1 m in the panchromatic band) or AVIRIS (images collected in 224
contiguous bands, with ground resolution of 20 meters (high altitude)or 4 meters (low
altitude)) in order to be more useful for local and county level applications (see the
Satellites/Sensors section in Annex 3).
12
3 Other Applications and Integration with Relevant Data Sets

During the workshop, only the above prototype data products were presented. However, a
number of other possible applications that might benefit from the use of remotely sensed
data were discussed.
Specifically, we reviewed the feasibility of using remote sensing for a variety of

applications (like air and water quality issues, floodplains mapping, etc.) and the
possibility of integrating satellite data with other relevant data sets, in particular
socioeconomic and population data.
3.1 Applications
Workshop attendees were interested in re-examining projects in which they have been
involved to see whether the use of remote sensing data could have helped to improve data
acquisition and facilitate the process of building a powerful information system for their
organizations.
As mentioned in the Introduction, the purpose of getting more and better data is to allow
a better understanding of environmental and human conditions that will, in turn, lead to
better management of natural resources and improve not only quality of life for local
communities, but also ease the entire decision-making process at the government level.
By looking at specific applications and interests, the needs and use of remotely sensed
data can be more easily outlined.
Presented here are some of the key areas discussed, in addition to the obvious benefit that
state and local governments may gain from having regular updates on urbanized areas,
infrastructure, and new developments. Urban and suburban applications will be further
discussed in the “Integration with Relevant Data” discussion.
Landscape Assessment
One of the key applications that can greatly benefit from satellite data, including medium
resolution sensors such as Landsat ETM+, is regional landscape assessment and
landscape change. Such studies can be pursued using the tailored land cover map or the
vegetation maps presented during the workshop (see Section 2).
The basic concept is that vegetation features can be distinguished from other features
because of their different spectral responses in the visible and infrared regions. In the
near infrared bands, vegetation has a peak of reflectance that water and urban do not have
because of reduced absorption by chlorophyll and other pigments. Also, as previously
described, the scale at which landscape assessment is done is usually regional. This
allows the use of Landsat data even for the 1970s and 1980s, when the TM sensor was
not yet available. The coarser resolution of MSS data can still be useful for regional-scale
applications.
The priority for landscape studies will then be to determine which land cover classes are
most important to identify based on specific needs. These needs vary from watershed
13
management to forest cover or agricultural mapping to linking forest extent and closeness
to residential areas for disease mapping and monitoring (e.g., Lyme disease).
Air Quality
Air quality applications basically deal with use of remote sensing to determine the effects
of air pollution on vegetation and would be particularly useful in urban and suburban
areas.
The change in spectral signature of a damaged plant gives an indication of the nature and
the level of air pollution damage. Such changes can be detected by looking at the
corresponding change in foliage colors and differences in structure and texture of the
canopy. For small areas or within cities, this will require high-resolution sensors such as
IKONOS. Assessing the effects of air pollution and separating them from those of
drought or diseases are not easy tasks. Diseases can equally affect trees and make them
change color and give a different spectral signature, which can be misinterpreted as an
effect caused by pollution. Remote sensing data in this case must be integrated with field
observations and measures of air pollution. The increasing number of studies of trees
affected by specific diseases will allow mapping of their spectral signature and
comparison with trees suspected of being affected by air pollution.
Water quality
Pure water reflects some of the incident radiation in the visible bands of the
electromagnetic spectrum and absorbs almost all of it in the near- and middle-infrared
bands. Therefore, in the infrared, water appears dark and is easily distinguishable from
other land features. The spectral response of water may vary with the presence of
suspended sediments, which increase the amount of radiation reflected. Some of these
sediments, such as suspended solids from soil erosion, can significantly impact the
spectral reflectance and be identified easily in the visible bands. Phytoplankton cannot be
easily distinguished from inorganic materials by common satellites (Landsat). Even other
more specific sensors (e.g., SeaWiFS, MODIS) require complex and sophisticated
calibrations.
Nevertheless, for the kind of applications discussed at the workshop, such as water
quality modeling and watershed management, traditional remotely sensed data (e.g.,
slope, soil, and land cover characteristics) can be used as input to hydrological models.
Such models can then be used to estimate non-point source water pollution (e.g., derived
from urban runoff, construction, agriculture, irrigation, and soil erosion). Non-point
source water pollution is very common but also difficult to detect given the limitation of
traditional in situ measurements techniques in identifying and modeling such diffuse
sources of pollution. The integration of in situ measurements with remotely sensed data
in GIS modeling can provide useful information on water quality applicable to many
planning and management issues.
Floodplain Mapping and Emergency Management

Monitoring sensitive areas such as wetlands, parks, or land in urbanized watershed for
emergency management usually requires high spatial resolution data in a timely fashion.
14
Relatively stable areas can be monitored every one or two years, whereas more critical
areas need to be monitored more often (at least seasonally (Lillesand, 2000)). In the case
of disasters such as floods, storms, tornadoes, earthquakes, or fires, it would be ideal to
have images both pre- and post-disaster. Even though the pre-disaster data would need to
be updated less frequently (every one to five years), it is important that they are at a
comparable high resolution (1-5 m) with the post-disaster data (less than 2 m). In disaster
management, satellites with medium ground resolution, such as Landsat or SPOT, are not
very useful, even though the temporal resolution is relatively high (16 and 26 days,
respectively). Normally, panchromatic and near-infrared aerial photography or IKONOS
panchromatic data acquired immediately after a disaster are the best ways to map the
extent of the disaster and to estimate its effects. If clouds are present, imaging radar can
provide the most useful information.
Health Applications
The contribution of environmental factors to human health and diseases has been
recognized and discussed since the time of Hippocrates. Well before the formulation of
scientific explanations for illnesses, there have been references to areas with “bad air”
and “bad water.” Responsibility for controlling such environmental hazards in order to
safeguard public health has increasingly fallen on the shoulders of local and regional
governments. In turn, officials attempt to educate and solicit input and assistance from
local residents.
Key environmental health issues include human exposure to contaminated air, water, and
soil and the spread of infectious diseases. Some of these health concerns are related
directly to human activities, such as toxic spills and emissions, pesticide runoff, and
radiation releases. Others can be naturally occurring, though are often indirectly affected
by human-induced changes, such as arsenic in groundwater or vector-borne diseases. In
the case of West Nile Virus and Lyme Disease, human infections depend upon the
proximity of arthropod vectors, animal hosts, and human populations.
Land cover, land use, and environmental conditions are useful predictors of potential risk.
In order to monitor and control the spread of West Nile Virus, local governments have
attempted to map the locations of dead birds and mosquito breeding sites (areas of
standing water). To track high-risk areas for Lyme Disease, they focus on the suburban
extensions of new developments into heavily vegetated areas. Identifying areas of high
risk using traditional field surveys can be costly and time-consuming. Local governments
could utilize remotely sensed imagery to more effectively select areas for surveillance
and intervention.
Unfortunately, the resolution of satellites such as Landsat is not sufficient to identify

many of the parameters needed to monitor vector-borne diseases. However, it would be
possible to use the medium-resolution satellite data as a baseline for major features
(vegetated areas and large wetlands). Those images could then be integrated with either
high-resolution satellite (IKONOS) or aerial photography data and supplemented with
data collected on the ground. Together, remote sensing and GIS could provide local and
regional governments with useful tools and information to help address these types of
environmental health issues. Although not discussed during the workshop, a similar
15
approach might be used to determine routes of exposure from hazardous waste sites and
spills. In addition, integrated remote sensing/GIS could be used to evaluate alternative
sites for hazardous facilities such as incinerators, in order to minimize the impact on local
populations and ecosystems.
3.2 Integration with Relevant Data Sets

Products derived from satellite imagery can be easily integrated into a GIS with other
data layers relevant to specific applications.
One of the most useful applications that emerged during the workshop was the
integration of land cover and land use change maps with population data derived from the
U.S. Census. The main interest is to determine the land cover and land use classes in
which population changes are occurring, in terms of both total population density and
population composition (by age, sex, and race). Another interesting application is to
determine whether and to what extent population changes are occurring in the same areas
where land use changes are occurring or have occurred. With the release of data from the
2000 Census, such analyses will be possible for 1990 and 2000.
The main issue in this case would be to find the proper scale to combine satellite-derived
data (particularly if Landsat data are used) and population data, collected at the tract or
block level.
The integration of remotely sensed data with population data relates also to the broader
topic of urban and suburban remote sensing and the resolution needed for many urban
applications. The spatial resolution at which urban and suburban features and changes
can be detected is usually more important than the spectral resolution. Therefore, if data
are needed more frequently than every five or ten years (usually the case), IKONOS is
probably the best source, although such data are currently quite expensive. Again,
calibration of coarser resolution data (panchromatic Landsat, for instance) with higher
resolution (panchromatic IKONOS) data could solve some of the resolution-related
issues, reducing costs and potentially providing more frequent coverage.
16
Annex 1. Workshop Agenda

February 23rd, 2001
Lamont Hall, at Columbia University’s Lamont-Doherty Earth Observatory
9:15 Coffee and Refreshments
9:30 – 9:45 Introduction, Robert Chen, Deputy Director, CIESIN, and Manager,
SEDAC
9:45 – 10:00 Workshop Goals, Francesca Pozzi, Research Associate, CIESIN
10:00 – 10:30 Remote Sensing Overview, Christopher Small, Lamont-Doherty Earth

Observatory, co-Project Scientist, SEDAC
10:30 – 12:30 Demonstration and discussion of prototype derived products:

1. (10:30-11:30) Land cover classification: how can this be tailored to users’
needs, which classes are useful to identify, potential applications
2. (11:30-12:30) Vegetation Fraction Index and Vegetation Change
Detection: uses and applications
12:30 – 1:15 Lunch (catered)
1:15 – 2:00 Discussions on users’ needs:

- Other possible applications (emergency management, flood mapping
etc)
- Scale
- Timing: what is required to make imagery and associated data valuable
to users?
- Other relevant datasets and their integration with remotely sensed data
- Issues of privacy and sensitive information
2:00 – 2:30 Discussion: Mechanisms for data sharing.
2:30 – 2:45 Break and change location (to Geoscience)
2:45 – 4:30 Hands On: Visit to Remote Sensing and Visual Lab and discussion of the
images and the prototype derived products in more details.
- Discussion of other satellite products (IKONOS).
- Break up into smaller working groups
17
Annex 2. List of Participants
Name Organization Address City/State Email

Artigas, Francisco Rutgers University 180 University Avenue Newark, NJ artigas@cimic.rutgers.edu
Brower, Robert Regional Application 197 Franklin Street Auburn, NY bbrower@co.cayuga.ny.us

Center for the NorthEast
(RACNE)
Chen, Robert S. CIESIN, Columbia 61 Route 9W Palisades, NY bchen@ciesin.columbia.edu

University
Constantine, Emily Regional Application 197 Franklin Street Auburn, NY ekc@racne.org

Center for the NorthEast
(RACNE)
Fedyniak, Myra Ulster County PO Box 557, Ulster Stone Ridge, NY myra_fedyniak@yahoo.com
Environmental Community College,
Management Council 12484
Fritscheler, Rick Ulster County PO Box 557, Ulster Stone Ridge, NY rfrits@aol.com
Environmental Community College,
Management Council 12484
Giri, Chandra CIESIN, Columbia 61 Route 9W Palisades, NY cgiri@ciesin.columbia.edu

University
Golden, Meredith CIESIN, Columbia 61 Route 9W Palisades, NY mgolden@ciesin.columbia.e

University du
Katzer, Paul NYC Parks & Recreation/ Arsenal North, 1234 New York, NY paul.katzer@parks.nyc.gov
Natural Resources Group Fifth Avenue, Room 240
Keneally, Carrie Westchester County GIS 148 Martine Avenue White Plains, NY cek1@westechestergov.com
Kyle, James Hunterdon County Administration Building- Flemington, NJ jkyle@co.hunterdon.nj.us

Route 12 County
Complex, PO Box 2900
Lopez-Torrijo, NYS Department of 50 Wolf Road, room Albany, NY rxlopez@dec.state.ny.us

Ricardo Environmental 302
Conservation
McGinty, Laura Westchester County GIS 148 Martine Avenue, White Plains, NY lam7@westechestergov.co
Room 305 m
Mickelson, John CIESIN, Columbia 61 Route 9W Palisades, NY jmickels@ciesin.columbia.ed

University u
O'Brien, Daniel NYS Emergency Building 22, Suite 101, Albany, NY daniel.obrien@semo.state.n
Management Office 1220 Washington Ave. y.us
Olney, Chris The Catskill Center for Rt. 28 Arkville, NY chriso@catskill.net

Conservation and
Development
Phillips, Steven Center for Biodiversity Central Park West at New York, NY phillips@research.att.com
and Conservation, 79th St.
American Museum of
Natural History
18
Name Organization Address City/State Email

Pozzi, Francesca CIESIN, Columbia 61 Route 9W Palisades, NY fpozzi@ciesin.columbia.edu
University
Puchalik, Joseph Rockland County Health County Health Pomona, NY puchaliJ@co.rockland.ny.us

Department Complex, Building D
Resnick, Susan NYC Department of 126 Worth Street New York, NY sresnick@health.nyc.gov
Health
Ridgway, Charlie Federal Emergency 26 Federal Plaza, Room New York, NY charlie.ridgway@fema.gov
Management Agency 1307
(FEMA)
Small, Christopher Lamont-Doherty Earth 61 Route 9W Palisades, NY small@ldeo.columbia.edu

Observatory, Columbia
University
Sterling, Eleanor Center for Biodiversity Central Park West at New York, NY sterling@amnh.org
and Conservation, 79th St.
American Museum of
Natural History
Tang, Dong Ming Division of Planning & Sullivan County Monticello, NY tdm@health.co.sullivan.ny.u
Community Government Center, s
Development, County of 100 North Street
Sullivan
Teichmann, Sonja Division of Planning & Sullivan County Monticello, NY stb@health.co.sullivan.ny.us

Community Government Center,
Development, County of 100 North Street
Sullivan
Wear, Sam Westchester County GIS 148 Martine Avenue White Plains, NY stw1@westchestergov.com
Williams, Christine CIESIN, Columbia 61 Route 9W Palisades, NY cwilliams@ciesin.columbia.e

University du
19
Annex 3. Background Information

Landsat 7
Landsat 7 was launched on April 15, 1999 from the Western Test Range at Vandenburg
Air Force Base on a Delta-II launch vehicle. It consists of a spacecraft bus, built by
Lockheed Martin Missiles and Space in Valley Forge, Pennsylvania and the Enhanced
Thematic Mapper Plus (ETM+) instrument, developed by Raytheon Santa Barbara
Remote Sensing in Santa Barbara, California. The satellite orbits the Earth at an altitude
of approximately 438 miles (705 kilometers) with a sun-synchronous 98-degree
inclination and a descending equatorial crossing time of 10 a.m. It crosses over any given
area every 16 days.
The ETM+ instrument is an eight-band multispectral scanning radiometer capable of

providing high-resolution image information of the Earth’s surface. It detects spectrally-
filtered radiation at visible, near-infrared, short-wave, and thermal infrared frequency
bands from the sun-lit Earth.
Nominal ground sample distances or “pixel” sizes are 49 feet (15 meters) in the
panchromatic band; 98 feet (30 meters) in the 6 visible, near-, and short-wave infrared
bands; and 197 feet (60 meters) in the thermal infrared band.
The ETM+ will produce approximately 3.8 gigabits of data for each scene.
The spectral and spatial resolutions for the ETM+ are reported in Table 1.
Band Number Spectral range Ground Resolution

(microns) (meters)
1 .45 to .515 (blue) 30
2 .525 to .605 (green) 30
3 .63 to .690 (red) 30
4 .75 to .90 (near-infrared) 30
5 1.55 to 1.75 (mid- 30
infrared)
6 (Thermal) 10.40 to 12.5 (thermal 60
infrared)
7 2.09 to 2.35 (mid- 30
infrared)
8 (Panchromatic) .52 to .90 (red and near- 15
infrared)
Table 1. Landsat 7 ETM+ Spectral and spatial resolutions.
20
Satellites/Sensors
The following table lists the most known and used satellites and their sensors, with
specifications about spectral, spatial, and temporal resolutions, what they can detect, and
applications they can be used for.
Satellite Sensor Spectral Spatial Temporal What Can Be Detected?

Resolution Resolution Resolution Spatial Temporal
(Wavelength
in µm)
LANDSAT MMS 1: 0.5-0.6 80 m; 16 days Mapping coastal Deforestation
(Multispectra (G) 185 Km features in sediment- Urban and
4, 5 l scanner 2: 0.6-0.7 swath width laden water suburban
system) (R) Mapping roads and development
URL: 3: 0.7-0.8 urban areas
http://geo.arc.n
(VNIR) Vegetation studies
asa.gov/sge/lan
dsat/landsat.ht 4: 0.8-1.1 and mapping
ml (NIR) land/water
boundaries
TM 1: 0.45- 30 m 16 days Soil/vegetation Changes in
Thematic 0.515 (B) (visible, near differentiation & heat islands
Mapper 2: 0.52-0.60 and mid-IR); coastal water Vegetation/la
(G) 120 m mapping nd use
3: 0.63-0.69 (thermal IR); Vegetation mapping patterns
(R) 185 Km Plant species
4: 0.75-0.90 swath width differentiation
(NIR) Biomass survey
5: 1.55-1.75 Snow/cloud
(Mid-IR) differentiation
6 (thermal): Thermal mapping
10.40-12.5 Geological mapping
7: 2.09-2.35
(Mid-IR)
LANDSAT ETM + 1: 0.45- 30 m 16 days Major Thoroughfares Changes in
(Enhanced 0.515 (B) (visible, near Large Buildings human
7 (1, 2, 3, 6 Thematic 2: 0.52-0.60 and mid-IR), Forest Stands infrastructure
are inactive) Mapper) (G) 15 m Agricultural Plots Development
3: 0.63-0.69 (panchromat Coastline patterns
URL: (R) ic), 60 m Advance/Retreat Migration
http://landsat7.u
4: 0.75-0.90 (Thermal Rugged Topography patterns
sgs.gov/
(NIR) Infrared); Sea Ice Coverage Agricultural
5: 1.55-1.75 185 Km variations
(Mid-IR) swath width Urban/Rural
6 (thermal): interchange
10.40-12.5
7: 2.09-2.35
(Mid-IR)
8 (pan):
0.52-0.90
21

(Wavelength
in µm)
IKONOS MMS 1: 0.45-0.53 4 m (visible), 26 days (680 Roads, vehicles, Changes in
(Multispectra (B) 1m km sun- buildings, human
1, 2 l) and PAN 2: 0.52-0.61 (panchromat synchronous infrastructure infrastructure
Launched in (Panchromat (G) ic); orbit) (panchromatic) Development
1999 by the ic) 3: 0.64-0.72 11 Km swath patterns
United States (R) width Land use, agricultural Migration
(IKONOS 2 4: 0.76-0.88 uses, vegetation patterns
failed) (VNIR) (color imager) Agricultural
Pan: 0.45 – variations
URL: 0.90 Urban/Rural
http://www.tbs-
interchange
satellite.com/tse
/online/sat_ikon
os_2.html
SPOT 1, Two HRVIR 1: 0.50-0.59 20 m 26 days Agriculture Deforestation
(High (G) (Visible, (Resource mapping, Suburban/Urb
2, and 4 (3 Resolution 2: 0.61-0.68 Near production an land use
is inactive) Visible, (R) Infrared), 10 management, crop changes
Launched by Infrared) 3: 0.79-0.89 m classification) Residential
France from push-broom (NIR) (panchromat Land Use (Urban and Development
1986-1998 sensors. 4: 1.58-1.73 ic); suburban land use, Coastal
(SWIR) – 60 Km swath land mapping, Pollution
Engineering added on width energy, human Water
work has Provides SPOT 4 infrastructure) resource
begun on coverage Pan: 0.51- Oceanography (water pollution
SPOT 5 with between 87 0.73 quality management) monitoring
a proposed degrees Water resources Snow and Ice
launch date in north and 87 (Surface water, soil mapping
late 2001 degrees moisture and Harvest
south evapotranspiration, forecasting
URL: lakes and rivers Conservation
www.spot.com/
studies, wetlands and monitoring
habitat mapping, Hazard
VEGETATION 1: 0.43-0.47 1 Km; 2200 Daily resource prediction
instrument (B) Km swath assessment) Landslide
(on SPOT 2: 0.61-0.68 width Geological hazards
4). (R) applications Forest
3: 0.78-0.89 (mapping, economic damage
(NIR) geology, engineering assessment
4: 1.58-1.75 geology, hazards and
(SWIR) land morphology. oil
and gas exploration)
Engineering
applications (terrain
analysis, site
investigation, water
resources
engineering, transport
studies.
Forest monitoring
(inventory, forest
management) and
vegetation cover
study (especially the
VEGETATION
sensor)
22

(Wavelength
in µm)
NOAA - 7 AVHRR 1: 0.58-0.68 4.4 Km 2 times per Day and night cloud Changes in
Launched in (Advanced (G and R) (Global Area day; 8-day top and sea surface climate and
1981 and Very High 2: 0.72-1.10 Coverage), and monthly temperatures global land
deactivated Resolution (NIR) 1.1 Km averaged Ice and snow and sea
1986 due to Radiometer) 3: 3.53-3.93 (Local Area data conditions temperatures
an power (Mid-IR) Coverage); available Changes in
failure 4: 10.3-11.3 2800 Km snow and ice
(Thermal IR) swath width coverages
URL: 5: 11.5-12.5
http://podaac.jpl (Thermal IR)
.nasa.gov/sst/
AVIRIS Hyperspect Contains 20 m (high Only Ecology (chlorophyll, Snow and Ice
Airborne ral airborne 224 different altitude), 4 m scheduled leaf water, lignin, Hydrology
Visible sensor detectors (low flights cellulose, pigments, (melting,
Infrared each with a altitude); structure, non- snow cover
Spectrometer Uses a wavelength 11 Km swath photosynthetic fraction)
(instrument on scanning sensitive width constituents) Commercial
board of mirror in a range of 10 Geology (mineralogy, (agricultural
planes) “wisk broom” nm, allowing soil type) correction)
URL: manner it to cover Cloud and Ecology
http://makalu.jpl the entire Atmospheric studies (changes in
.nasa.gov/aviris range (water vapor, clouds vegetation
.html between 0.4 properties, aerosols, and
and 25 µm. absorbing gases) community
Oceanography/Coast maps)
al and Inland Waters Oceanograph
(chlorophyll, y (changes in
dissolved organics, plankton
sediments, bottom coverage and
composition, chlorophyll)
bathymetry) Forest Fires
Snow and Ice
Hydrology (grainsize,
impurities)
Biomass burning
(smoke, combustion
products)
Environmental
Hazards
Commercial
ERS2 AMI (Active 5.3 GHz (C- 30 m (SAR) 3 day, 35 All-weather Alterations
Microwave Band) 50 Km day or 168 instrument and
(Active) Instrumenta- 13.5 GHz for (Scatter- day cycles Ocean wave observations
tion) with the Radar ometer); 80- height/lengths, wind in ocean,
SAR-Image Altimeter 100 Km speed/direction, ice land, ice,
Mode, SAR- swath width parameters, sea atmosphere,
Wave Mode, (SAR-Image surface & cloud top and climate
Scatter- mode); 5 Km temperatures, cloud Flood activity
ometer swath width cover and Changes in
Mode and (SAR-Wave atmospheric water ocean
Radar mode), 500 vapor. activity,
Altimeter Km swath coastal
width regions and
(Scatteromet ice caps
er mode)
23

(Wavelength
in µm)
ERS2 ATSR-M 1.6, 3.7, 11, 1 Km (IR), 3 day, 35 All-weather Alterations
(Along Track 12 (IR), 23.5 22 Km day or 168 instrument and
(Cont’d) Scanning and 36.5 (Microwave); day cycles Ocean wave observations
Radiometer GHZ 500 Km height/lengths, wind in ocean,
with (Microwave) swath width speed/direction, ice land, ice,
Microwave parameters, sea atmosphere,
Sounder) surface & cloud top and climate
GOME 1: 0.24- 40 x 2Km temperatures, cloud Flood activity
(Global 0.295 40 x 320 cover and Changes in
Ozone 2: 0.29- Km; 960 Km atmospheric water ocean
Monitoring 0.405 swath width vapor. activity,
Experiment). 3: 0.40- coastal
Sensor is a 0.605 regions and
double 4: 0.59-0.79 ice caps
spectromete
r
AATSR 0.65, 0.85, 0.5 Km; 500
(Advanced 1.27, 1.6 KM swath
Along Track width
Scanning
Radiometer)
SEASTAR SeaWiFS 1: 0.402- 1.1 Km 1 day Ocean color and Changes in
(Sea-viewing 0.422 (local area chlorophyll phytoplankton
URL: Wide Field 2: 0.433- coverage) Subsurface scattering Designed to
http://seawifs.gs Sensor) 0.453 4.5 Km Atmospheric provide global
fc.nasa.gov/SE 3: 0.480-0.5 (global area correction coverage of
AWIFS.html 4: 0.5-0.520 coverage); Atmospheric the oceans on
5: 0.545- 285 Km correction a regular
0.565 swath width Sea-surface basis
6: 0.66-0.68 temperature
7: 0.745-
0.785
8: 0.845-
0.885
TERRA ASTER 14 bands, 15 m 4-16 days Major Thoroughfares Infrastructure
(Advanced with (VNIR), 30 By request Large Buildings Changes
URL: Spaceborne wavelengths m (SWIR), Forest Stands Residential
http://terra.nasa Thermal ranging from 90 m (TIR); Agricultural Plots Development
.gov/About/ Emission 0.52 to 60 Km swath Coastline s
and 11.65 width Advance/Retreat Deforestation/
Reflection Rugged Topography Reforestation
Radiometer) Sea Ice Coverage Harvest
Flood Area
Landslides &
Mass
Movements
24

(Wavelength
in µm)
TERRA MODIS 36 bands, 250 m 1-2 days Ideal for large scale Forest Fires
(Moderate with (bands 1-2), changes in the Regional
(cont’d) Resolution wavelengths 500 m biosphere, measures Harvest/
Imaging ranging from (bands 3-7), photosynthetic activity Cycles
Spectro- 0.405 to 1000 m of land and marine Plankton
Radiometer) 14.38 (bands 8- plants Blooms
36); 2330 x Surface temperature Sediment
10 Km swath measurements, Plumes
width Deforestation Maps extent
Forests, Open of snow and
Canopy Vegetation, ice brought by
Large Scale winter storms
Agriculture and frigid
Water Clarity, conditions
Atmospheric
Aerosols, Smoke
Plumes, Snow Cover,
Ocean Temperature
MISR (Multi- 4 bands, 275 m; 360 9 days The amount of Smoke
angle with Km swath sunlight scattered in Plumes
Imaging wavelengths width the atmosphere under Regional Air
Spectro- ranging from natural conditions, Quality
Radiometer) 0.44 to 0.86 Atmospheric aerosol Climate
particles (formed by Regional
both natural and Forest
human activities) Canopy
Cloud Cover/Type, Structure
Vegetation Type
CERES Shortwave: 20 km Daily Cloud/radiation flux Contributes to
(Clouds and 0.3-5 measurements for wider range
Earth’s Longwave: models of oceanic weather
Radiant 8-12 and atmospheric forecasting
Energy Total: 0.3- energetics
System) >200 The cross track mode
continues
measurements of
Earth Radiation
Budget Experiment
and Tropical Rainfall
Measuring Mission
MOPITT 2.3 (CH4) 22 Km 3 – 4 days Measurements of
(Measure- 2.4 and 4.7 horizontally pollution in the
ment of (CO) and 3 Km troposphere
Pollution in vertically; Used to determine
the 640 Km the amount of Carbon
Troposphere) swath width dioxide and methane
in the atmosphere
25
References
Buiten, H. J. and Clevers, J., Land Observation by Remote Sensing, Gordon and Breach
Science Publishers, 1993.
Lillesand, T.M. and Kiefe, R.W., Remote Sensing and Image Interpretation, John Wiley
& Sons, Inc, 2000.
Research Systems, ENVI User’s Guide, Research Systems, 2000.
Richards, J.A., Remote Sensing Digital Image Analysis, Springer-Verlag, 1994.
Small, C., Estimation of urban vegetation abundance by spectral mixture analysis,

International Journal of Remote Sensing, v.21, n.7, 2001.
26

Remote Sensing Applications

Uploaded by

Copyright:

Available Formats

Remote Sensing Applications

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Remote Sensing Applications

Uploaded by

Copyright:

Available Formats

Remote Sensing Applications at the

State and Local Level

Sponsored by the Socioeconomic Data and Application Center (SEDAC)

The Workshop home page can be found at

CIESIN focuses on applying state-of-the-art information technology (e.g., computer

CIESIN at Columbia University

2 Prototype Data Products Discussion ............................................................................ 2

3 Other Applications and Integration with Relevant Data sets .................................. 13

Annex 1. Workshop Agenda........................................................................................... 17

Annex 2. List of Participants.......................................................................................... 18

Annex 3. Background Information................................................................................ 20

The workshop’s objectives were:

• To discuss other potential applications, such as emergency management, flood

2 Prototype Data Products Discussion

2.1 Tailored land cover/land use classification.

The maximum-likelihood approach is one of the methods used in supervised

Figure 2. Maximum Likelihood Classification. Legend:

Looking at the “Regions of Interest” statistical distribution, it is possible to get a sense of

The advantage of a user-specific classification, compared with other available products,

2.2 Vegetation Fraction Index and Urban/Vegetation Change Detection

An even simpler model is a Greenness map, obtained by performing a Tasseled Cap

The Tasseled Cap Transformation on Landsat TM data performs an orthogonal

Figure 5. Greenness map obtained from Landsat MSS imagery, 06/10/87.

Figure 6. Greenness map obtained from Landsat MSS imagery, 06/21/91.

Nevertheless, the potential importance of such tools and products in monitoring

3 Other Applications and Integration with Relevant Data Sets

Specifically, we reviewed the feasibility of using remote sensing for a variety of

Floodplain Mapping and Emergency Management

Unfortunately, the resolution of satellites such as Landsat is not sufficient to identify

3.2 Integration with Relevant Data Sets

Annex 1. Workshop Agenda

Lamont Hall, at Columbia University’s Lamont-Doherty Earth Observatory

9:15 Coffee and Refreshments

9:45 – 10:00 Workshop Goals, Francesca Pozzi, Research Associate, CIESIN

10:00 – 10:30 Remote Sensing Overview, Christopher Small, Lamont-Doherty Earth

10:30 – 12:30 Demonstration and discussion of prototype derived products:

12:30 – 1:15 Lunch (catered)

1:15 – 2:00 Discussions on users’ needs:

2:00 – 2:30 Discussion: Mechanisms for data sharing.

2:30 – 2:45 Break and change location (to Geoscience)

Annex 2. List of Participants

Name Organization Address City/State Email

Brower, Robert Regional Application 197 Franklin Street Auburn, NY bbrower@co.cayuga.ny.us

Chen, Robert S. CIESIN, Columbia 61 Route 9W Palisades, NY bchen@ciesin.columbia.edu

Constantine, Emily Regional Application 197 Franklin Street Auburn, NY ekc@racne.org

Giri, Chandra CIESIN, Columbia 61 Route 9W Palisades, NY cgiri@ciesin.columbia.edu

Golden, Meredith CIESIN, Columbia 61 Route 9W Palisades, NY mgolden@ciesin.columbia.e

Kyle, James Hunterdon County Administration Building- Flemington, NJ jkyle@co.hunterdon.nj.us

Lopez-Torrijo, NYS Department of 50 Wolf Road, room Albany, NY rxlopez@dec.state.ny.us

Mickelson, John CIESIN, Columbia 61 Route 9W Palisades, NY jmickels@ciesin.columbia.ed

Olney, Chris The Catskill Center for Rt. 28 Arkville, NY chriso@catskill.net

Name Organization Address City/State Email

Puchalik, Joseph Rockland County Health County Health Pomona, NY puchaliJ@co.rockland.ny.us

Small, Christopher Lamont-Doherty Earth 61 Route 9W Palisades, NY small@ldeo.columbia.edu

Teichmann, Sonja Division of Planning & Sullivan County Monticello, NY stb@health.co.sullivan.ny.us

Williams, Christine CIESIN, Columbia 61 Route 9W Palisades, NY cwilliams@ciesin.columbia.e

Annex 3. Background Information

The ETM+ instrument is an eight-band multispectral scanning radiometer capable of

Band Number Spectral range Ground Resolution

Satellite Sensor Spectral Spatial Temporal What Can Be Detected?