International Journal of Geosciences, 2011, 2, 195-203
doi:10.4236/ijg.2011.23021 Published Online August 2011 (http://www.SciRP.org/journal/ijg)
Web-Based Spatial Decision Support System and
Watershed Management with a Case Study
Yanli Zhang1, Ramanathan Sugumaran2, Matthew McBroom1, John DeGroote2,
Rebecca L. Kauten3 Paul K. Barten4
1Arthur Temple College of Forestry and Agriculture, Stephen F. Austin State University, Nacogdoches, USA
2Department of Ge ography, University of Northern Iowa, Cedar Falls, USA
3Iowa Dept. of Natural Resources, Des Moines, USA
4Department of Environmental Conservation, University of Massachusetts, Amherst, USA
Received May 7, 2011; revised June 12, 2011; accepted July 21, 2011
Abstract
In order to maintain a proper balance between development pressure and water resources protection, and also
to improve public participation, efficient tools and techniques for soil and water conservation projects are
needed. This paper describes the development and application of a web-based watershed management spatial
decision support system, WebWMPI. The WebWMPI uses the Watershed Management Priority Indices
(WMPI) approach which is a prioritizing method for watershed management planning and it integrates land
use/cover, hydrological data, soils, slope, roads, and other spatial data. The land is divided into three catego-
ries: Conservation Priority Index (CPI) land, Restoration Priority Index (RPI) land, and Stormwater Man-
agement Priority Index (SMPI) land. Within each category, spatial factors are rated based on their influence
on water resources and critical areas can be identified for soil conservation, water quality protection and im-
provement. The WebWMPI has user-friendly client side graphical interfaces which enable the public to in-
teractively run the server side Geographic Information System to evaluate different scenarios for watershed
planning and management. The system was applied for Dry Run Creek watershed (Cedar Falls, Iowa, US) as
a demonstration and it can be easily used in other watersheds to prioritize crucial areas and to increase public
participation for soil and water conservation projects.
Keywords: Web-Based SDSS, Watershed Management, GIS
1. Introduction
It is well understood that clean water is one of the essen-
tial elements of sustainable development. The watershed
is the natural water resources un it widely recognized as a
fundamental geographic unit for measuring and analyz-
ing the relative health or co ndition of th e landscap e [1 ,2].
However watershed degradation is a common phenome-
non around the world that directly results in poor water
quality. For example, the state of Iowa assessed streams
whose designated uses are water supply, recreation,
wildlife, and aquatic life harv esting in the years 2000 and
2001. It was reported that 34% of them were threatened
and 38% were impaired, which means any one of its as-
sessed uses was not met. For assessed ponds, 64% were
threatened and 33% were impaired [3]. At a national
level, in the Water Quality Inventory, USEPA [2] re-
ported that about 40% of assessed river and stream
lengths, 46% of assessed lake areas, 51% of assessed
estuarine areas, and 78% of assessed Great Lakes shore-
line lengths did not meet water quality standards.
Watersheds are also important from an ecological per-
spective. As water flows over the ground and along riv-
ers, it can pick up nutrients, sediment, and other pollut-
ants, which are transported to the outlet of the basin.
These pollutants can negatively impact ecological proc-
esses within the watershed, as well as the receiving water
body. Hence, scientific watershed management is critical
to protecting water resources and ecosystems.
Watersheds are characterized by meteorological, sur-
face water and groundwater, as well as physical and bio-
logical factors functioning within the context of natural
and human disturbance regimes. The flow, quality, or
timing of water within a watershed is influenced by these

Y. ZHANG ET AL. 196
factors [4]. Watershed management decision-making is
inherently complex because it integrates biophysical
sciences, socioeconomic information, simulation models,
and expert judgment. Decision makers can benefit from
scientific and user friendly decision support systems
(DSS) that allow them to better understand and evaluate
alternative management approaches.
2. Decision Support System and Application
Decision support systems (DSS) are a specific class of
computerized information systems that support deci-
sion-making activities. In general, DSS are interactive
computer-based systems and subsystems intended to
assist decision makers’ use of communication technolo-
gies, data, documents, knowledge and/or models to iden-
tify and solve problems and make decisions [5]. A DSS
should help with formulating alternatives, accessing data,
using models to evaluate alternatives, displaying, and
interpreting results [6]. Th ere is a wide range of spatially
distributed information on characteristics of watersheds,
such as soil, land use/cover, topogr aphy, strea ms, zoning,
etc. For this reason, DSS used in watershed management
are correspondingly classified as spatial decision support
systems (SDSS). The Geographic Information System
(GIS) is considered a generator for SDSS [7] because of
its power and efficient functions to store, retrieve, ana-
lyze, manipulate, and display large volumes of spatial
digital data and to create maps. In the last 20 years, gov-
ernment agencies, academic institutions, and consulting
firms have developed many SDSS for watershed man-
agement that utilize GIS for modeling watershed proc-
esses, data management, and other purposes. These
SDSS, such as WAMview [8], WARMF [9], WAWER
[10], and SWAT [11] are good references for future DSS
development.
Watershed management requires cooperation between
federal, state, and local stakeholders to integrate bio-
physical and socioeconomic processes [12]. In a recent
national random digit dial survey, the vast majority (83%)
of participants said that the public should play a more
prominent role in environmental management from
serving on advisory boards to actually making manage-
ment decisions [13]. Public participation in the forest
planning process, especially in small groups, can help
reduce the number of appeals and can help managers in
identifying the concerns of local residents early in the
forest planning process [14]. A bottom-up approach that
involves stakeholders at the beginning of a planning
process with a SDSS could be more efficient. Barten and
Ernst [15] pointed out that watershed management re-
quires the sustained involvement of a broad set of stake-
holders. People and organizations that are actively in-
volved in the process from the ou tset are more willing to
make a substantial commitment of time and resources to
ensure successful implementation of water quality im-
provement plans. However, many SDSS require expen-
sive and complex GIS software whose use requires spe-
cialized professional knowledge by well trained staff.
Thus, the sophisticated nature of GIS-based SDSS often
excludes many potential stakeholders or the public who
would otherwise benefit from them. Fortunately, due to
the rapid development of distributed computing tech-
nologies and high speed Internet, available GIS software
products already enable people to share not only spatial
datasets but also advanced geoprocessing functions
across the Internet. In other words, GIS software can
now be centralized in application servers and web serv-
ers to deliver GIS capabilities to many users over net-
works. Correspondingly, SDSS is moving to a new
web-based version era and there are many articles and
books relating the development of Web-based SDSS
since year 2000 [16-19]. When compared to traditional
computer based SDSS, web-based SDSS have many ad-
vantages. According to Power [5], Paz et al. [6], Miller
et al. [12], Peng and Tsou [16], Rinner [17], Sugumaran
et al. [18], Wang and Cheng [19], Dymond et al. [20],
and Choi et al. [21], these advantages include:
1) Centralized control over model and data, which
means lower costs for hardware, software, distribution,
maintenance, and training, as well as greater efficiency
in model improvement and data update, especially for
models using real time info rmation , such as water qua lity
monitoring;
2) Global and easy accessibility, which means users do
not need professional GIS knowledge, training, or ex-
pensive and complex hardware/software;
3) Platform independence.
These advantages allow and encourage stakeholders
and the public to access and participate in the planning
and decision-making process, which can impact the qual-
ity of their lives. Additionally, interactive, web-based
SDSS can:
1) Greatly improve communication and coordination
between and among decision makers, stakeholders, and
the public. Any potential problems or conflict can be
found at the earliest stage;
2) Be easily transferred to and implemented for other
watersheds;
3) Play a demonstration and educational role to the
public in environmental conservation, modeling science,
GIS, and other areas;
4) Improve public awareness of the existence of spa-
tial digital data and scien tific models.
Recognizing these advantages, a number of web-based
models have been developed to support watershed man-

Y. ZHANG ET AL.197
agement. These models vary in focus areas and complex-
ity. Sugumaran et al. [22] built a Web-based Floodplain
Advisory Tool (WFAT) to visualize and retrieve data to
support floodplain management in St. Charles County,
Missouri. Utilizing remote sensing and GIS data, a user
could query and display different flood plain related data
layers and determine the elevation of a land parcel and
its location with regards to the Federal Emergency Man-
agement Agency (FEMA) 100-year flood plain. Engel et
al. [23] used L-THIA (Long-term Hydrologic Impact
Assessment) web DSS to evaluate how land use changes
impact long term hydrology and nonpoint source (NPS)
pollution in a watershed. The L-THIA used the National
Resources Conservation Service (NRCS) curve number
technique. To assess the potential effectiveness of best
management practices (BMP), Miller et al. (2003) de-
veloped a prototype spatial web-DSS for rangeland wa-
tershed management which integrated water quality,
livestock management, and economic concerns. Sugu-
maran et al. [18] designed a web-based environmental
DSS (WEDSS) to prioritize local subwatersheds in terms
of environmental sensitivity using multiple criteria.
WebL2W is a model that predicts hydrologic and eco-
nomic effects and fish habitat quality based on user de-
fined land use development s cenarios [20].
Overall, the rapid development of internet and GIS
provide an opportunity to integrate state-of-the-art tech-
nology with new modeling systems to create online deci-
sion support tools for decision makers. This study de-
scribes the development of a Web-based SDSS, Web-
WMPI, which is based on the Watershed Management
Priority Indices (WMPI) approach [15]. The WebWMPI
provides tools for prioritizing and ranking critical areas
within a watershed that influence water resources, as-
sisting decision makers and other stakeholders to under-
stand their watershed, and for investigating measures to
protect water quality and quantity.
3. Watershed Management Priority Indices
Approach
Nonpoint source pollution from agriculture and urban
and suburban development accounts for more than 60%
of the impairment in U.S. waterways, including many
drinking water sources [24]. Nonpoint source pollution is
the cumulative effect of poor land use and natural re-
source management. Bhaduri et al. [25] found that an
18% increase in urban or impervious areas resulted in an
estimated 80% increase in annual av erage runoff volume
and estimated increases of more than 50% in annual av-
erage loads for lead, copper, and zinc in the Little Eagle
Creek watershed (Indianapolis, IN). At present, under
prevailing development pressure, more and more people
are aware of the necessity for water conservation. Also,
land conservation and pollution prevention have proven
to be cost effective strategies [26]. Nevertheless, water
conservation, or watershed protection can be a vague and
limitless task. Where and how to start watershed man-
agement in order to have the maximum environmental
benefits are common questions raised by foresters, plan-
ners, environmental protection organizations, and com-
munities. A practical and efficient watershed analysis
tool is needed to help people answer these questions.
There are many reasons for environmental degradation
of watersheds, but the most important reason is the im-
proper utilization of watershed resources, among which
land use allocation and practices are the key issues be-
cause water is naturally accumulated from the land sur-
face. For the purpose of conserving water resources, the
landscape features that significantly influence water re-
sources include forestland s, wetlands, natural grasslands,
steep slopes, riparian area, and land with erodable soils.
However, these landscape properties are spatially dis-
tributed and intermixed with each other. How to effec-
tively combine, analyze, and interpret information on
these landscape properties is a challenge and decisions
must be made about where to spend limited resources
and how to best prioritize land parcels or areas for con-
servation, restoration, or other treatments to improve
water quality. The identification of areas that have been
degraded or impaired by human activities as well as
those that favorably influence water quality is the pri-
mary objective of watershed an alysis.
Ian McHarg pioneered the basic concept of overlay
analysis of ecological, hydrological, and other environ-
mental data in 1969 [27-29]. At that time the overlay
work was done with transparent maps. This method has
been widely used in planning, natural resources man-
agement, and other fields. GIS makes it more rigorous
and objective and also makes it possible to undertake
large and more complex projects. From 2001 to 2004,
The United States Environmental Protection Agency
(USEPA) Office of Ground Water and Drinking Water,
the Trust for Public Land (TPL), University of Massa-
chusetts-Amherst, and the United States Department of
Agriculture (USDA) Forest Service cooperated in the
Source Water Stewardship Project. Our team developed
a new GIS analysis approach called WMPI (Watershed
Management Priority Indices), which is used to identify
and classify areas and activities that positively or nega-
tively influence source water quality in watersh eds based
on the overlay theory [15]. The procedure uses raster
overlay and creates three indices broadly representing the
principal uses or conditions of land: 1) forests and wet-
lands that positively influence water quality, 2) agricul-
ture and barren land that negatively in fluence water qual-

Y. ZHANG ET AL. 198
ity, and 3) residential, commercial, and industrial areas
that need specific management strategies. They are
named the Conservation, Restoration, and Stormwater
management priority indices (CPI, RPI, and SMPI) land,
respectively. Appropriate management activities can be
adopted for each category. For example, forest land with
a high CPI value would be a top candidate for protection
and enhancement measurements to prevent further deg-
radation, such as conservation easement if funds are
available; construction or restoration of riparian forest on
agricultural land and strict BMPs could be used for high
RPI score areas. Similarly for high SMPI value areas,
construction of infiltration systems, bio-cells, or storm
water ponds could be suggestions. Also, WMPI analysis
results can be used with water quality da ta. If heavy met-
als are the primary cause of water quality degradation,
then managers need to work on areas with high SMPI
scores. If nutrients are the main water quality problem,
working on some of the critical RPI areas would help to
reduce agricultural pollutants.
Spatial data needed for the overlay process are avail-
able from the U.S. Geological Survey (USGS), USDA,
state GIS data clearinghouses, and other agencies. Within
WMPI, the potential influence of each land cell on water
resources is represented by the total score generated by
ranking and combining all the input layers. The detailed
WMPI calculation process has four steps. First, land
use/cover is classified into three categories: CPI, RPI,
and SMPI. Second, each cell of ev ery input GIS layer, or
land property, is assigned a high (3), intermediate (2),
low (1), or not applicable (0) value, based on its influ-
ence on water quality. Using a soil input layer as an ex-
ample, a cell with sandy soils would be assigned a value
of low (1) while another cell with silty soils should have
a value of high (3). Third, each input layer will be multi-
plied by its assigned weight. This allows the user to ad-
just the relative importance of each layer. Finally, all of
the input layers will be spatially overlain or added to
calculate the total score for each cell. Cell values in the
resulting layer reflect all of the input bio-physical prop-
erties. For example, a site which has forest cover, steep
slopes, silty soils, and is located adj acent to a water body,
would receive high scores for conservation in relation to
other sites. By contrast, a level, sandy site, far away from
the stream network would have low scores. This ap-
proach accounts for all of the possible combinations of
soil, land use/cover, and location characteristics and am-
plifies the difference among diverse areas. The default
cell value assignments were based on literature review.
For example, soil data were used to develop a permeabil-
ity profile and depth to seasonal high water table layers
as surrogates for the likelihood of overland flow and
NPS pollutant loading. In the case of riparian areas, the
30-meter proximity to water body corresponds to the
100-foot buffer zone recommended by USDA or man-
dated in some state regulations for riparian management
[30].
WMPI approach has been used successfully to evalu-
ate the four watersheds in the Source Water Stewardship
Project (Barten and Ernst 2004). The relatively complex
calculation procedure makes it difficult to be widely used.
To automate the four calculation steps, WMPI was de-
veloped as one part of the Watershed Forest Manage-
ment Information System [31] which is a SDSS in the
form of an ArcGIS (version 9 and above) extension. It
has friendly graphical user interfaces and allows the user
to make decisions based on his/her knowledge and un-
derstanding of the model. Extra factors, such as imper-
vious area, flood plains, aquifer protection areas, buffers
of contaminated sites, biodiversity, and channel migra-
tion zones, can be added easily into the analysis. The
WMPI approach has several advantages when compared
with other models. Required data are easy to access and
process. It is not site specific. It is raster cell based and
has more detailed results than subwatershed based mod-
els. Also it is based on the state-of-the-art software, Ar-
cGIS 9.
4. WebWMPI Design and Development
To exploit the advantages of web-based SDSS as dis-
cussed in a previous section, WMPI has been moved to
an on line version. WebWMPI is based on the cli-
ent/server model in which clients send requests to ser-
vices running on a server and receive appropriate infor-
mation in response (Sugumaran 2004). In a very similar
interface mode to the local version, WebWMPI was de-
veloped with Arc GI S Se rver, Active Server Pages (ASP),
Java Script, and Visual Basic .NET. It is designed as a
server-side SDSS or a thin client architecture and its
structure is shown with Figure 1. A Client-side SDSS
approach was not adopted because it would require the
installation of ArcGIS software to the users’ computer. A
thin-client SDSS can be thought of as moving the user
interfaces from a local computer based DSS to the users’
web browser. The client/server used here has a three
tiered configuration consisting of: Tier 1: web browser;
Tier 2: IIS (Internet Information Services) web server
and GIS server; and Tier 3: data. The information flow is
as follows: 1) the user initiates a request by manipulating
tools, textboxes, and buttons on the web browser, 2) the
IIS web server passes the requests to the GIS server to do
the processing such as spatial data access, vector to raster
conversion, reclassification, and overlay. 3) the GIS
server creates map images based on the geographical
data and passes to the IIS web server, 4) IIS formats the

Y. ZHANG ET AL. 199
Figure 1. WebWMPI Architecture and transaction.
output into HTML pages and serves the content to the
client’s web browser, and 5) the web browser displays
the results and supports further user interaction.
WebWMPI consists of a data system, a model, and
user interfaces. The system provides dynamic web forms
allowing interaction b etween th e user and the server. It is
envisioned that local, state, and federal agencies along
with non-profit watershed associations could be the po-
tential users of this system. The users could then demon-
strate and pass on information derived from the system to
a wider audience through meetings and public participa-
tion. The advantages of the system are that it is simple
and straightforward, and does not require expertise in
GIS applications, but rather just the use of a web bro wser
and some knowledge of watershed biophysical processes.
5. WebWMPI Application
Dry Run Creek watershed (Cedar Falls, IA) was selected
as the demonstration area for WebWMPI. It has an area
of 61.5 km2 (15,197 acres) and 47 km (29.2 miles) of
streams. According to the Iowa DNR land cover classi-
fication of 2002, 37 .7 km2 (9,3 16 acr es) or 61 .3 % of lan d
area are in agriculture, 13.3 km2 (3287 acres) or 21.6%
are in developed area, such as residential, commercial,
industrial, and roads, and only 10.5 km2 (2595 acres) or
17.0% are considered as natural areas, such as water,
wetland, forest, and unmanaged grasslands. Dry Run
Creek watershed has been subject to urban development
over time and its water resources are facing serious
problems. In 2002 and 2004, Dry Run was listed in
Iowa’s Section 303(d) list as category 5b waters, which
means the watershed is impaired by unknown reasons
and is in need of the establishment of a Total Maximum
Daily Loading (TMDL)[32]. In order to prevent further
water degradation and improve water quality within the
watershed, Dry Run Creek watershed needs a scientific
analysis of its water related bio-physical resources. The
successful application of WebWMPI on Dry Run Creek
watershed could provide a “proof of concept” to other
impaired water systems.
All of the original spatial data were collected from
Iowa DNR. Topographical data in the form of a 30-meter
resolution DEM was used to delineate the watershed
boundary and to derive slope data. Land cover is for the
year 2002. Other data included were a road network, the
Soil Survey Geographic (SSURGO) database, rivers,
National Wetlands Inventory data, and water bodies.
Users can access WebWMPI at
http://geotree2.geog.uni.edu/webwmpi (verified on May
1, 2011) and Figure 2 shows the main interface. There
are general GIS functions like zoom in, zoom out, layer
turn on/off, and identify on the map interface to let users
check available spatial informatio n of the watershed. The
two buttons with watershed symbols on the toolbar are
WebWMPI specific tools. One is for watershed delinea-
tion. When the user selects this button and clicks a point
on the map, a watershed will be delineated assuming that
point is the outlet of a watershed. The WMPI button in-
vokes the WMPI analysis interfaces (see Figure 3).
The user needs to go through all interfaces to set up
WMPI parameters. The first interface is used for input
layer selection and to define the analysis boundary. Users
just need to check those layers that would be used in the
analysis. The weight is a number that is used to multiply
by the corresponding layer rankings. In contrast to pair-
wise comparison or rating, the weight is used to adjust
the relative importance among input layers and allows
the user to make exp licit trade-off decisions. The second
interface is used to classify land use/cover types into CPI,
RPI, and SMPI categories and to assign a value based on
their potential impact on water quality. Default values

Y. ZHANG ET AL. 200
Figure 2. WebWMPI home page.
Figure 3. WebWMPI user interfaces.

Y. ZHANG ET AL. 201
are assigned as a reference but can be changed. For ex-
ample, forest land can be set to CPI with a score of three
because in general it positively influences water quality.
The server uses the information entered in the second
interface to classify land use/cover data into three rasters
based on this setting. Users can exclude some land cate-
gories by not assigning them to one of the three indices.
The third interface is for parameter setting, such as
buffer width and slope classification intervals. In the last
interface the output format is selected. The entire range
of CPI, RPI, and SMPI can be displayed, or the 70th, 80th,
and 90th percentile of corresponding PI categories can be
calculated automatically. An optional output is a chess-
board which can be used to divide the watershed into
small areas for management purpose. Last, users just
need to click the analysis button and wait until the web
browser displays the results of the analysis.
Figure 4 shows an example analysis result for Dry
Run Creek watershed based on our default parameter
settings. Within the map, the symbology for CPI, RPI,
and SMPI are green, orange, and red, respectively. The
darker the color, the higher the score is. Management
priorities could be given to those areas with the highest
scores after field assessment. Until now, the coordinator
of Dry Run Creek watershed has used the WebWMPI to
identify hot spots to bu ild stormwater retention pond and
to restore stream bank. The tool has been used to demon-
strate those hot spots in local watershed management
meetings involving the public. BMP suggestions were
provided to land owners having lands with critical areas.
With cadastral data, a zonal analysis could be used to
identify critical parcels.
Figure 4. WebWMPI analysis result for Dry Run Creek
Watershed (Cedar Falls, IA, USA).
The example application of WebWMPI used a small
local watershed. However, WebWMPI has a flexible
analysis scale as the analysis boundary is selected by the
user and there is a watershed delineation function. With
the proper data and hardware preparation, it could be
expanded to county, state, or regional level depending on
the analysis boundary selection. In other words, WebWMPI
can be considered as a hierarchical SDSS. For example,
if the server is populated with state wide data, users
could identify critical areas at the state level or at the city
level based on the analysi s bo un dary setting.
There are potential disadvantages to the application of
Web-based SDSS. These systems provide greater poten-
tial for wide use among not only academic and regula-
tory organizations but other stakeholders such as public
interest groups. This introduces the possibility of misun-
derstanding or mis-application of the modeling system.
Questions like how to prevent the model from being
misused or misinterpreted and others like security prob-
lems should be kept in mind for all web-based SDSS
designers. The WebWMPI system has been used so far
for academic demonstration purposes and for use with
local Dry Run Creek watershed stakeholders. When the
model is released for wider use stricter access controls
would be put in place. These would include the require-
ment of a login id that would be provided by the
WebWMPI application manager upon registration by the
user. The use of the system would be monitored by the
WebWMPI managers. In addition, extensive help sys-
tems and tutorials are p rovided to guide the user through
application of WebWMPI. Finally, in the future, when
the outputs are prov ided as downloadab le datasets, all of
the input parameters will also be provided with any
analysis results.
6. Conclusions
Using the dynamic information delivery capabilities of
web technology, the WebWMPI, a web-based watershed
management SDSS, has been designed to support water-
shed decision-makers and to provide information about
the critical areas within a watershed that influence water
resources. With friendly user interfaces, it can achieve
the purpose of making watershed conservation knowl-
edge accessible to stakeholders and the public who may
have limited GIS or watershed science knowledge and
providing them a tool to evaluate different management
scenarios.
At present, WebWMPI does not incorporate social or
economic factors, such as land ownership. However, this
does not preclude the application of WebWMPI in pre-
dominately private owned watersheds. For example,
WebWMPI analysis result could provide recommenda-

Y. ZHANG ET AL. 202
tions for conservation easement purchasing of parcels or
for zoning regulations. Nonetheless, adding social and
economic factors to WebWMPI would be one potential
future development direction. Other possible develop-
ment of the WebWMPI model may include, but is not
limited to, archiving analyses for users to let them re-
trieve previous results, data download, adding hydro-
logical analysis, cost-benefit analysis, improving visu-
alization through three dimensional techniques, soil ero-
sion calculation, linking to real-time water quality moni-
toring system (pollution indication), and providing BMP
suggestions for the management of critical area.
The work done so far for the WebWMPI has illus-
trated the great advantages of a web-based SDSS in soil
and water conservation project. Compared with the ap-
plication of desktop WMPI in the four watersheds of the
Source Water Stewardship Project (Barten and Ernst
2004), WebWMPI attracted and allowed more stake-
holders to investigate watershed conditions. Also Dry
Run Creek watershed coordinators have used it as a re-
mote assessment tool without concern for GIS software
and locally stored datasets. At the same time, its devel-
opment highlights potential future directions of web
SDSS, online geoprocessing, or web-based GIS service.
These directions could include:
1) Adoption of new technologies, such as Ajax which
allows partially refreshing web pages and makes web
applications smaller, faster and more user friendly.
Timely response is the key factor to attract and maintain
web users. However, geoprocessing with large datasets
normally takes time. Presently, it takes 1.5 minutes to
finish the analysis with the most complex parameter set-
tings for WebWMPI with Dry Run Creek watershed
data;
2) Allowing users to upload data for an analysis. At
present, most GIS services, including WebWMPI, can
only provide functions with server-side data. In the fu-
ture, accepting users’ data for analysis will be one poten-
tial direction if multi- spatial data validation can be done
quickly with artificial intelligence; and
Another direction is the potential to combine other
online tools developed by separate organizations. One
successful example is Shi et al.’s model [33], which in-
tegrates Michigan State University’s Digital Watershed
and Understanding Your Watershed Systems and Purdue
University’s Online Watershed Delineation Tool and
L-THIA across the web. This kind of integration can
avoid duplicated efforts and will promote the develop-
ment and application of SDSS at all levels.
6. References
[1] J. E. Williams, C. A. Wood and M. P. Dombeck, “Wa-
tershed Restoration: Principles and Practices,” American
Fisheries Society, Philadelphia, 1997.
[2] USEPA, “2000 National Water Quality Inventory Re-
port,” 2011.
http://www.epa.gov/305b/2000report.
[3] USEPA, “2002 National Assessment Database,” 2011.
http://www.epa.gov/waters/305b/index.html.
[4] B. McCammon, J. Rector and K. Gebhardt, “A Frame-
work for Analyzing the Hydrologic Condition of Water-
sheds,” USDA Forest Service and USDI Bureau of Land
Management, BLM Technical Note 405, Washington,
1998.
[5] J. D. Power, “Building Web-Based Decision Support
Systems,” Studies in Informatics and Control, Vol. 11,
No. 4, 2002, pp. 291-302.
[6] J. O. Paz, W. D. Batchelor and P. Pedersen, “WebGro: A
Web-Based Soybean Management Decision Support Sys-
tem,” Agronomy Journal, Vol. 96, 2004, pp. 1771-1779.
doi:10.2134/agronj2004.1771
[7] P. Keenan, “Using a GIS as a DSS Generator,” DSSRe-
sources.com, 2004.
http://dssresources.com/papers/features/keenan/keenan12
172004.html.
[8] D. Bottcher and J. G. Hiscock, “WAMview-a GIS/Land
Source Based Approach to Watershed Assessment
Model,” Proceedings of 2001 Water Environment Fed-
eration (WEF) TMDL Science Issues Conference, St.
Louis, 5-7 March 2001, pp. 112-123.
[9] L. H. Weintraub, C. W. Chen and J. Herr, “Demonstra-
tion of WARMF: A Decision Support Toll for TMDL
Development,” Proceedings Water Environment Federa-
tion (WEF) TMDL Science Issues Conference, St. Louis,
4-7 March 2001.
[10] E. Girvetz and F. Shilling, “Decision Support for Road
System Analysis and Modification on the Tahoe National
Forest,” Environmental Management, Vol. 32, No. 2,
2003, pp. 218-233. doi:10.1007/s00267-003-2970-1
[11] K. S. Bracmort, M. Arabi, J. R. Frankenberger, B. A.
Engel and J. G. Arnold, “Modeling Long-Term Water
Quality Impact of Structural BMPs,” Transactions of the
American Society of Agricultural and Biological Engi-
neers, Vol. 49, No. 2, 2006, pp. 367-374.
[12] R. Miller, D. P. Guertin and P. Heilman, “An Inter-
net-Based Spatial Decision Support System for Range-
land Watershed Management,” Proceedings of the 1st In-
teragency Conference on Research in the Watersheds,
Benson, 27-30 October 2003, pp. 725-730.
[13] B. S. Steel and E. Weber, “Ecosystem Management, De-
centralization, and Public Opinion,” Global Environ-
mental Change, Vol. 11, No. 2, 2001, pp. 119-131.
doi:10.1016/S0959-3780(00)00062-5
[14] R. L. Ryan, “Social Science to Improve Fuels Manage-
ment: A Synthesis of Research on Aesthetics and Fuels
Management,” USDA Forest Service, North Central Re-
search Station, General Technical Report. NC-261,
USDA, 2005.
[15] P. K. Barten and C. E. Ernst, “Land Conservation and
Watershed Management for Source Protection,” Journal

Y. ZHANG ET AL. 203
of the American Water Works Assoc iation, Vol. 96, No. 4,
2004, pp. 121-136.
[16] Z. R. Peng and M. H. Tsou, “Internet GIS: Distributed
Geographic Information Services for the Internet and
Wireless Networks,” John Wiley & Sons, Hoboken,
2003.
[17] C. Rinner, “Web-Based Spatial Decision Support: Status
and Research Directions,” Journal of Geographic Infor-
mation and Decision Analysis, Vol. 7, No. 1, 2003, pp.
14-31.
[18] R. Sugumaran, J. Meyer, and J. Davis, “A Web-Based
Environmental Decision Support System (WEDSS) for
Environmental Planning and Watershed Management,”
Journal of Geographical Systems, Vol. 6, 2004, pp. 1-16.
doi:10.1007/s10109-004-0137-0
[19] L. Wang and Q. Cheng, “Web-Based Collaborative Deci-
sion Support Services: Concept, Challenges and Applica-
tion,” Proceedings of the ISPRS Vienna 2006 Symposium,
Vienna, 12-14 July 2006.
[20] R. L. Dymond, B. Regmi, V.K. Lohani and R. Dietz,
“Interdisciplinary Web-Enabled Spatial Decision Support
System for Watershed Management,” Journal of Water
Resources Planning and Management, Vol. 130, No. 4,
2004, pp. 290-300.
doi:10.1061/(ASCE)0733-9496(2004)130:4(290)
[21] J. B. Choi, A. Engel and R.L. Farnsworth, “Web-Based
GIS and Spatial Decision Support System for Watershed
Management,” Journal of Hydroinformatics, Vol. 7, No.
3, 2005, pp. 165-174.
[22] R. Sugumaran, C. Davis, J. Meyer, T. Prato and C. Ful-
cher, “Web-Based Decision Support Tool for Floodplain
Management Using High-Resolution DEM,” Journal of
Photogrammetric Engineering and Remote Sensing, Vol.
66, No. 10, 2000, pp. 1261-1265.
[23] B. A. Engel, J. Choi, J. Harbor and S. Pandey, “Web-
Based DSS for Hydrologic Impact Evaluation of Small
Watershed Land Use Changes,” Computers and Elec-
tronics in Agriculture, Vol. 39, No. 3, 2003, pp. 241-249.
doi:10.1016/S0168-1699(03)00078-4
[24] USEPA, “Nonpoint Source Pollution: The Nations’
Largest Water Quality Problem, EPA841-F-96-004A,”
USEPA, 1996.
[25] B. Bhaduri, J. Harbor, B. Engel, and M. Grove, “Assess-
ing Watershed-Scale, Long-Term Hydrologic Impacts of
Land-Use Change Using a GIS-NPS Model,” Environ-
mental Management, Vol. 26, No. 6, 2000, pp. 643-658.
doi:10.1007/s002670010122
[26] NRC (National Research Council), “Watershed Man-
agement for Potable Water Supply: Assessing New York
City’s Approach,” National Academies Press, Washing-
ton, 2000.
[27] L. I. Mcharg, “Design with Nature,” John Wiley & Sons,
Hoboken, 1969.
[28] I. Heywood, S. Cornelius and S. Carver, “An Introduction
to geographical Information Systems,” Prentice Hall,
Upper Saddle River, 1998.
[29] P. A. Longley, M. F. Goodchild, D. J. Maguire and D. W.
Rhind, “Geographic Information Systems and Science,”
Wiley, 2001.
[30] NRC (National Research Council), “Riparian Areas:
Functions and Strategies for Management,” National
Academies Press, Washington, 2002.
[31] Y. Zhang and P. K. Barten, “Watershed Forest Manage-
ment Information System (WFMIS),” Environmental
Modeling and Software, Vol. 24, No. 4, 2009, pp.
569-575. doi:10.1016/j.envsoft.2008.10.006
[32] Iowa DNR (The Iowa Department of Natural Resources).
“Iowa Section (303d) Impaired Waters Listings,” 2011.
http://wqm.igsb.uiowa.edu/wqa/303d.html
[33] Y. Shi, J. Asher, J. Bartholic, J. Choi, B. Engel and R.
Farnsworth, “An online webGIS-Based Hierarchical Wa-
tershed Decision Support System for United States,” En-
vironmental Informatics Archives, Vol. 2, 2004, pp.
838-845.