Case Study

Environmental Impact Assessment for Transportation

Projects: Case Study Using Remote-Sensing Technology,
Geographic Information Systems, and Spatial Modeling
M. A. El-Gafy, Ph.D., P.E.1; Y. A. Abdelrazig, Ph.D.2; and T. S. Abdelhamid, Ph.D.3
Downloaded from ascelibrary.org by GADJAH MADA UNIVERSITY on 11/28/17. Copyright ASCE. For personal use only; all rights reserved.

Abstract: A major function of planners is to promote the best use of a community’s land and resources for different construction projects;
especially critical are infrastructure projects on which economic development relies. Transportation projects typically involve both environ-
mental and economic issues facing a community as it grows and changes. Considerations of sustainability, as well as the widespread use of
collaborative planning, design, and construction, require tools that facilitate long-term impact analysis and easy communication among built-
environment professionals. Environmental impact assessment (EIA) is instrumental for studying transportation project impacts, but it is a
time-consuming process because of the large number of dependent and independent variables involved. This paper presents a descriptive case
study for the analysis phase of an EIA for transportation projects on the basis of the integration of remote-sensing technology, geographic
information systems, and spatial modeling. Environmental vulnerability around the project areas is presented by exploiting the advantages of
the map overlay method and the matrix method. A vulnerability grade map and road distribution map were produced, providing an overall
vulnerability score for each of the three planning alternatives considered. A road alignment was indicated as the preferred corridor, which was
further recommended for a project development and environmental (PD&E) study. The proposed framework provides for a comprehensive
environmental assessment of transportation projects and improvement of the quality of the decision-making process in urban and interurban
projects. DOI: 10.1061/(ASCE)UP.1943-5444.0000050. © 2011 American Society of Civil Engineers.
CE Database subject headings: Environmental issues; Assessment; Spatial data; Spatial analysis; Geographic information systems;
Remote sensing.
Author keywords: Environmental impact assessment; Geospatial information; Environmental assessment; Environmental vulnerability;
Spatial modeling.

Introduction Given this situation, implementing a transportation project in-

volves dealing with a range of challenges to control undesired ef-
Developments in the transportation industry have had a consider- fects on the environment. Environmental impact assessment (EIA)
able impact on the natural environment because transportation is has been developed as a tool to address environmental issues in
the main source of greenhouse gas emissions in the United States. decision making. Over the past few decades, the EIA process
Roads and parking create impervious surfaces that increase water has evolved to meet concerns about applying the decision-making
runoff and produce urban heat island effects (DeCicco and Mark process to strategic environmental assessment (Kreske 1996).
1998; Greene 1996). Transportation facilities can also cause habitat Colorni et al. (2000) proposed a decision support system (SILVIA)
fragmentation, generate noise and vibration, and impact wetlands containing a methodological framework and a decision support sys-
and other natural ecosystems as well as affect historic resources, tem for EIA problems and illustrated the impact-analysis phase and
community cohesion, and other social and community characteris- decision phase of two case studies: the Pedemontana highway in
tics. Because of the wide range of potential impacts on the natural Lombardy and Mantua ring-road. Feng and Wang (2007) proposed
and human environment, the U.S. National Environmental Protec- a step-by-step process that integrates cost-benefit analysis (CBA)
tion Act (NEPA), the Clean Air Act (CAA), and other legislation with EIA to assess the overall effects (both environmental and eco-
require an environmental analysis of the potential impacts of trans- nomic) on a transportation project.
portation projects before they are implemented (Carson 1992). EIA is defined by Glasson et al. (1999) as a systematic process
that examines, in advance, the environmental consequences of a
1
Assistant Professor, School of Planning, Design and Construction, proposed development action. In EIA, the environment is viewed
Michigan State Univ., East Lansing, MI. E-mail: [email protected] as a system comprising human beings, fauna and flora, soil, water,
2
Associate Professor, Dept. of Civil Engineering, Florida State Univ., air, the climate, and the landscape; EIA is also concerned with
Tallahassee, FL. interactions among these components. EIA provides a unique
3
Associate Professor, School of Planning, Design and Construction, opportunity to predict ways in which the environment may be
Michigan State Univ., East Lansing, MI (corresponding author). E-mail: improved and predicts the conflicts and constraints between a
[email protected]
proposed project or program and its environment (Hensher and
Note. This manuscript was submitted on October 27, 2008; approved on
July 22, 2010; published online on July 24, 2010. Discussion period open
Button 2003).
until November 1, 2011; separate discussions must be submitted for indi- Although EIA is dependent on geospatial information to make
vidual papers. This paper is part of the Journal of Urban Planning and an assessment, there are no standard rules regarding how to conduct
Development, Vol. 137, No. 2, June 1, 2011. ©ASCE, ISSN 0733-9488/ an environmental assessment. Currently, the particular objective of
2011/2-153–158/$25.00. each assessment is driven by the project specifics and by what

JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / JUNE 2011 / 153

J. Urban Plann. Dev., 2011, 137(2): 153-158

 information is available (Booz et al. 2003). The Federal Highway
Administration (FHWA) issued its own “Guidance for Preparing
and Processing Environmental and Section 4(F) Documents”
(FHWA 1987). However, there are no requirements defining
how the evaluation must be conducted or what data sets and analy-
ses must be utilized for the assessment.
This paper presents a descriptive case study at the network level
for the analysis phase of the EIA for a transportation project.
This comprehensive methodology will allow transportation plan-
ners to make rational decisions through cost-effective coordination
of environmental information and data management. The results
assess the environmental vulnerability around the road and its
Downloaded from ascelibrary.org by GADJAH MADA UNIVERSITY on 11/28/17. Copyright ASCE. For personal use only; all rights reserved.

impact on the environment by integrating the merits of the map

overlay method and the matrix method—two common methods
in the GIS and remote-sensing fields.

Case Study Site

A new roadway (Crestview Bypass), located in northern Okaloosa

County, Florida, was used as a case study to illustrate the proposed
methodology. The 2025 cost feasible plan identifies this proposed
road segment as a four-lane divided facility. Based on a traffic
analysis in zones adjacent to the corridor, the population was
anticipated to grow by 7,115 (from 3,935 to 11,050) or 64.39%,
between 1997 and 2020. Employment was projected to increase
by 154 (from 224 to 378) or 40.74%. The number of dwelling units
was forecasted to rise by 2,728 (from 1,501 to 4,229) or 64.51%.
The planned roadway was to maintain an existing bicycle path and
Fig. 1. Crestview Bypass project location map
sidewalk in the corridor. Fig. 1 shows the three different road align-
ments proposed for this project.

Methodology

The proposed methodology can be represented by five main

processes, as illustrated in Fig. 2.

Problem Structuring
The Environmental Technical Advisory Team (ETAT), established
by the State of Florida Department of Transportation (FDOT)
for each FDOT district, defined the environmental assessment
problem as a decision tree, comprising environmental and mon-
etary issues. The objective was the protection of the environment
and minimization of negative impacts on the surrounding areas.
The system structure of the assessment factors is defined as illus-
trated in Fig. 3.
To determine the assessment factors, the matrix method was
used to relate various factors to the road’s behavior of construction
and operation, which revealed the road’s environmental impact in a
selection table. Table 1 illustrates the assessment factors for the
case study.
The degree of environmental protection of each option is ex- Fig. 2. Main processes of the proposed EIA methodology
pressed on an artificial scale of impact scores, which corresponds
to a verbal description as shown in Table 2. This artificial scale Data Analysis
ranges from 0 for no impact to 3 for strong negative impact. A score
of 100 on this artificial scale was used for reflecting regulatory The data analysis process was executed by four subprocesses,
conditions, such as one stating that the road cannot pass over a namely, data bank and system retrievals, geographic information
registered historical place. system (GIS) module, acquiring satellite images, and image
This scale was used for rating each criterion shown in Table 1, processing module. Two GIS software programs—ArcInfo and
and the three main alternative options. Then the importance of the ERDAS—were used extensively in the case study. The ArcInfo
criteria was specified on a scale by using, in this case, gradations software was used in the manipulation, analysis, and modeling
from 0 to 5 (Table 3). These values are used to generate the weights of spatial and nonspatial data. ERDAS was used primarily in
of the evaluation. the classification and processing of land use and land cover data.

154 / JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / JUNE 2011

J. Urban Plann. Dev., 2011, 137(2): 153-158

Downloaded from ascelibrary.org by GADJAH MADA UNIVERSITY on 11/28/17. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. System structure for the Crestview Bypass case

Table 1. Selected Assessment Factors for the Case Study Table 3. Importance of the Criteria
System criterion Assessment factor ID Verbal description Value
Economic impact Population density 1 Extreme negative impact 5
Historical and archeological sites Historical and archeological sites 2 Strong negative impact 4
Land use Land cover and use of the land 3 Large negative impact 3
Water quality and quantity DRASTIC index for the surface 4 Moderate negative impact 2
and the aquifer systems 5 Small negative impact 1
Environmental geology of Sediment type located within 10 ft
similarity of the land’s surface
Contaminated sites Green project’s ecological network Data Bank and System Retrievals
model results The Florida Geographic Data Library (FGDL) was selected as a
Wetland Wetland classification mechanism for distributing aerial photographs and spatial (GIS)
Flood plain Flood Insurance Rate Maps of data. The main reason for choosing it is that this library has been
FEMA designed carefully and is well managed, as well as providing the
fast track for the updating procedure. Table 4 shows a partial list of
Wildlife, threatened, and Wildlife zones
spatial (GIS) data layers used for the proposed project.
endangered species
Special outstanding water Special outstanding water GIS Module
Military land Military land ArcInfo software was used as a GIS platform to prepare the data
processing and preparation, including the following tasks:
1. File format conversion: the software was used to convert the
image data from the Geo tiff format to the img format, which is
the ERDAS IMAGINE format of the picture.
2. Reprojection: gathered spatial (GIS) data layers were collected
Table 2. Degree of Environmental Impact of Each Score from different sources and were projected with different
projections. All data layers were reprojected to the same geo-
Score Verbal description graphic reference to establish an implicit spatial correspon-
100 Extreme negative impact dence relationship between the layers, which facilitates
3 Strong negative impact subsequent data processing and analysis.
2 Moderate negative impact 3. Area of interest (AOI) clipping: the satellite image, aerial
photographs, and spatial (GIS) data layers covered different
1 Small negative impact
area extents. The ArcInfo submodule was used to clip the area
0 No impact
of the internet to be used for further analysis. The AOI was

JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / JUNE 2011 / 155

J. Urban Plann. Dev., 2011, 137(2): 153-158

 Table 4. Partial List of Data Layers Required
Description File name Type Source Update/new Extent
2000 Census blocks CENBLK Polygon USCB New County
Drastic coverage for the surficial aquifer system AQDRSU Polygon FDEP — County
FEMA Flood Insurance Rate Maps 1996 FEMA96 Polygon FEMA — County
Historic bridges SHPO_BRIDGES Line BAR New State
Strategic Habitat and Conservation Areas 2000 SHCA Polygon FFWCC New State

identified by the boundary limits of Okaloosa County. For the Experts determined the pairwise comparisons matrix as shown in
clipping of the raster data, the Spatial Analyst calculator Table 5.
Factors’ weights were calculated by the “BLZPACK” software
Downloaded from ascelibrary.org by GADJAH MADA UNIVERSITY on 11/28/17. Copyright ASCE. For personal use only; all rights reserved.

was used.
that is an implementation of the block Lanczos algorithm (Marques
Acquiring Satellite Images and Vasco 2000). The weights vectors are 0.125, 0.128, 0.069,
IKONOS imaginary scenes, licensed by Space Imagining, Inc., was 0.126, 0.034, 0.064, 0.168, 0.063, 0.099, 0.092, and 0.032; and
selected as the data source. The IKONOS data is relatively high
the consistency ratio is 0.038, less than 0.1, which satisfies the
and provides large coverage. The available revisit frequency for
consistency criterion.
collection areas near the equator for an image collected at a ground-
sample distance of 1 m in the panchromatic and 4 m in the multi- Environmental Vulnerability Grade Map
spectral bands, is every 3.9 days. The current output is either UTM
or state-plane projected, with WGS 84, NAD 83, or NAD 27 data, Developing the environmental vulnerability grade map can be rep-
so the image must be reprocessed to FGDL Albers equal-area resented by the following: (1) reclassifying the GIS layers and the
projection. classified maps according to the regional weights within the same
criteria to develop criteria vulnerability maps; (2) rasterizing all the
Image Processing Module maps and GIS layers using the Spatial Analyst module, which is an
To project the data onto a plane and make it conform to the base
extension module within ArcInfo (1 m was chosen as the output
map projection system, image-to-image rectification was used to
cell size for the layers); and (3) the ERDAS IMAGINE Model
rectify the IKONOS map using a georeferenced Digital Orthophoto
Quarter Quads (DOQQs)—1 m resolution of the same area. Maker is used to construct a script model to combine the criteria
Unsupervised and supervised classification approaches were used vulnerability maps and the criteria-related weights to develop the
to derive land use/land cover of the area of interest. Unsupervised environmental vulnerability map. Fig. 4 illustrates the concept of
classification is used to extract the superblock clusters based on combining the criteria vulnerability maps and the criteria-related
natural groupings of spectral data. When the signature is satisfac- weights to develop the environmental vulnerability map.
tory, all subclasses of each signature class are employed in the
supervised classification. Finally, an ERDAS Image Accuracy Distribution Maps of Road Impact Extent
Assessment submodule was used to generate random check points
Road impact extent should be considered to achieve a more
throughout the classified image.
scientific EIA and to select more reasonable optimal alignment.
Criteria Weighting Module Two studies in the Netherlands and Massachusetts evaluated sev-
eral “road-effect zones” over which significant ecological effects
For the measurement of criteria-related weights, pairwise compar- extend outward from a road (Forman and Alexander 1998; Forman
isons were utilized for practicality, operational efficiency, and et al. 1997). The effects of all factors extended > 100 m from the
reliability, as well as acceptance by decision makers. The ETAT
road, and moose corridors, road avoidance by grassland birds, and
experts were asked to compare all criteria, thus creating a matrix
perhaps road salt in a shallow reservoir extended outward > 1 km
of values for each comparison pair. The average of all the experts’
matrices was used to determine the main eigenvector of the com- (Forman and Deblinger 2000). The proposed methodology allowed
parison matrix. Then the weights of the vector were normalized to for performing the analysis on different buffer distances from the
apply Yager’s methods (1977) for calculating the required weights. centerline of the proposed alignment.

Table 5. Pairwise Comparisons Matrix

Population density Arch LULC DRASTIC Foundation Eco. model Wetland Flood Wildlife Spec. water Military
Population density 1 0.91 0.45 0.9 0.23 0.46 1.1 0.45 0.7 0.68 0.23
Arch 1.10 1 0.46 0.91 0.22 0.46 1.14 0.46 0.68 0.7 0.22
LULC 2.22 2.17 1 1.82 0.45 0.91 2.27 0.91 1.3 1.4 0.45
DRASTIC 1.11 1.10 0.55 1 0.22 0.45 1.1 0.45 0.6 0.7 0.23
Foundation 4.35 4.55 2.22 4.55 1 1.8 4.5 1.8 2.7 2.7 0.9
Eco. model 2.17 2.17 1.10 2.22 0.56 1 2.25 0.9 1.3 1.4 0.44
Wetland 0.91 0.88 0.44 0.91 0.22 0.44 1 0.4 0.5 0.6 0.17
Flood 2.22 2.17 1.10 2.22 0.56 1.11 2.50 1 1.3 1.4 0.44
Wildlife 1.43 1.47 0.77 1.67 0.37 0.77 2.00 0.77 1 0.9 0.29
Spec. water 1.47 1.43 0.71 1.43 0.37 0.71 1.67 0.71 1.11 1 0.29
Military 4.35 4.55 2.22 4.35 1.11 2.27 5.88 2.27 3.45 3.45 1

156 / JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / JUNE 2011

J. Urban Plann. Dev., 2011, 137(2): 153-158

Downloaded from ascelibrary.org by GADJAH MADA UNIVERSITY on 11/28/17. Copyright ASCE. For personal use only; all rights reserved.

Fig. 4. Combining criteria vulnerability maps and criteria-related weights toward developing an environmental vulnerability map

Table 6. Criteria Vulnerability Score and the anthropogenic environment from the construction and op-
eration of built-environment projects. Image processing techniques
Assessment Alternative Alternative Alternative
value (E) at no. 1 no. 2 no. 3 of satellite images of the area of interest are used to acquire
historical environmental data and to reconstruct missing data. Next,
100 ft 467,064 384,858 98,856 a standard analysis approach is formulated based on the integration
200 ft 929,007 784,971 203,571 between GIS data and the remote sensing data.
500 ft 2,095,686 1,879,560 582,147 Using inputs by experts and stakeholders, evaluation criteria and
1 mi 26,291,331 13,484,547 13,216,431 the relative importance of these criteria is determined. Finally, the
integrated data and the relative weights and criteria are used to
arrive at a measure for the environment impact of the project. This
Selection of the Best Alternative analysis can be graphically and numerically presented during pub-
By buffering along the road-alignment alternatives at distances of lic hearings and/or community meetings in an accessible and easy
30.48 m (100 ft), 60.96 m (200 ft), 152.4 m (500 ft), and 1 mi manner. In particular, with a focus on sustainable planning, urban
(5,280 ft), respectively, the polygons of road-impact extent and regional planners can better explore long-term impacts caused
will be created. With PC ArcInfo, we overlay the vulnerability by growth and revitalization of urban, suburban, and rural com-
grade map and the road distribution maps. The calculated criteria munities and the regions in which they are located. The forecasting
vulnerability scores of the three alternatives are shown in Table 6. ability provided by the developed framework will assist urban plan-
ners, during their work with local officials, in understanding and
addressing environmental problems associated with locations for
Exploring the Results roads and other infrastructure. In general, this methodological
framework is expected to improve the quality of the decision-
As a result of the previous calculations, the proposed framework making process and planning for built-environment projects regard-
recommends road alignment 3 as the preferred corridor. This result ing the concerns for environmental impacts.
was compatible with the one endorsed by the Crestview Bypass
Eastern Corridor (E-one) committee. The corridor was also selected
for further project development and environmental study. Upon re- References
viewing the results, ETAT members agreed that the application
demonstrated its advantages by following a qualitative analysis ap- Booz Allen Hamilton and Michael Baker Jr., Inc. (2003). Environmental
proach rather than a conventional quantitative approach. In addi- information management and decision support system: Implementation
tion, the proposed framework acts as a comprehensive method handbook, National Academy Press, Washington, DC.
for environmental assessment of transportation projects. Carson, J. (1992). “On the preparation of environmental impact statements
in the United States of America.” Atmos. Environ. Part A, 26(15),
2759–2768.
Colorni, A., Laniado, E., and Muratori, S. (2000). “Application of decision
Summary and Conclusions support systems to environmental impact evaluation of road transport
infrastructures.” Proc., ICTTS 2000 (Traffic and Transportation
This paper presented a hybrid decision support framework for Studies), Beijing, China.
conducting EIA for urban and interurban projects, especially trans- DeCicco, J., and Mark, J. (1998). “Meeting the energy and climate
portation and infrastructure projects. The basic concept of the challenge for transportation in the United States.” Energ. Pol., 26(5),
developed framework is to develop a hierarchal decision-tree struc- 395–412.
ture, according to user objectives, to include impacts on the natural Federal Highway Administration (FHWA). (1987). “Guidance for prepar-

JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / JUNE 2011 / 157

J. Urban Plann. Dev., 2011, 137(2): 153-158

 ing and processing environmental and section 4(F) documents.” Glasson, J., Therivel, R., and Chadwick, A. (1999). Introduction to
6640.8A, U.S. Dept. of Transportation, Washington, DC. environmental impact assessment, University College London Press,
Feng, C., and Wang, S. (2007). “Integrated cost–benefit analysis with London.
environmental factors for a transportation project: Case of Pinglin Greene, D. L. (1996). Transportation and energy, Eno Transportation
interchange in Taiwan.” J. Urban Plann. Dev., 133(3), 172–178. Foundation, Lansdowne, VA.
Forman, R., and Alexander, L. (1998). “Roads and their major ecological Hensher, D., and Button, K. (2003). Handbook of transport and the envi-
effects.” Annu. Rev. Ecol. Systemat., 29, 207–231. ronment, Elsevier Science, Amsterdam.
Forman, R., and Deblinger, R. (2000). “The ecological road-effect zone of Kreske, D. L. (1996). Environmental impact statements: A practical guide
a Massachusetts (USA) suburban highway.” Conservat. Biol., 14(1), for agencies, citizens, and consultants, Wiley, New York, 480.
36–46.
Marques, O., and Vasco, D. (2000). “Solving large linear inverse problems
Forman, R., Friedman, D., Fitzhenry, D., Martin, J., Cheb, A., and
in geophysics by means of eigenvalue calculations.” Technical Rep.
Alexander, L. (1997). “Ecological effects of roads: Toward three
LBNL-45324, Berkeley, CA.
summary indices and overview for North America.” Habitat fragmen-
Yager, R. R. (1977). “Multiple objective decision-making using fuzzy sets.”
Downloaded from ascelibrary.org by GADJAH MADA UNIVERSITY on 11/28/17. Copyright ASCE. For personal use only; all rights reserved.

tation and infrastructure, K. Canters, ed., Ministry of Transport, Public

Works and Water Management, Delft, Netherlands. Int. J. Man-Machine Studies, 9, 375–382.

158 / JOURNAL OF URBAN PLANNING AND DEVELOPMENT © ASCE / JUNE 2011

J. Urban Plann. Dev., 2011, 137(2): 153-158