Applying Geographic Information Systems: Siting of Coastal Hotels in Costa Rica

Applying Geographic Information Systems

Siting of Coastal Hotels

in Costa Rica

by Albert Joeyev, Stephen D. DeGloria, and

Malcolm A. Noden

Computer-based geographic information systems (GISs) can improve hotel-siting

decisions in environmentally sensitive locations, as shown by a study of the

northwestern coast of Costa Rica.

H otel developers have long realized that location is

the key ingredient to a hotel’s economic success. Determining the ideal location for a prospective hotel, however, involves a complicated set of criteria. In this article we describe a computer-based approach to systematically identie appropriate development sites based on a defined set of criteria.The tool we describe is geographic information systems (GISs), which can integrate economic, social, and environmental data into the hotel-siting decision.

We believe GIS-based siting decisions are particularly well-suited to determining the appropriate location of hotel properties in emerging destinations, particularly those with fragile ecosystems. To illustrate the use of a GIS, we dis- cuss one such decision process as it applies to Peninsula De Nicoya, on the northwestern, Pacific coast of Costa Rica. GIS analysis dovetails with traditional evaluation methods for siting, such as economic feasibility studies. We do not propose the use of GIS technology as a substi-

Albert Joerger, Ph.D., is a research associate at the Cornell University School of Hotel Administration c(ag’[email protected], where Malcolm A. Noden is a senior lecturer [email protected]. Stephen D. DeGloria, Ph.D., is an associate professor in the Department of Soil, Crop, and Atmospheric Sciences and director of the Cornell Institutefor Resource Information Systems at the College ofAgriculture and Lfe Sciences [email protected]~>. The authors are indebted to the organixa- tions that supported the research described here: at Cornell University, the School of Hotel Administration, the Department of Soil, Crop, and Atmospheric Sciences, the Program on Ecological and Social Science Challenges of Conservation, and the Institutefor Resource Information Systems; and at the University of Costa Rica, the Centro de Investigaciones Agronomicas. These individuals were instrumental in advancing this research: Aljiiedo Alvarado, Freddie Sancho, Stephen Smith, and Patrick vay1 Locke. 0 1999, Cornell University

48 IIIIKNFI I i I I HOTEL AND RESTAURANT ADMINISTRATION QUARTERLY

tute for a traditional feasibility analysis, but we suggest that a GIS study is a valuable, complementary analysis to typical feasibility studies.

In a region such as Costa Rica, which is still in the discovery stage of the tourism life cycle, GIS analysis allows hotel developers or planning bodies to consider the prospects for a large region or multiple regions and then narrow their development focus to a relatively small number of sites. Especially when a hotel developer is looking at large tracts of land or is evaluating a nation or region for multiple development sites, a GIS analysis can quickly produce a list of sites to consider based on such key criteria as, for instance, distances from existing roads, distances from beaches and other natural attractions, suitability of the site for construction, and the risk of natural disasters.

As new hotels are being developed in environmentally sensitive zones, siting issues become increasingly complex. GIS analysis helps to identify both opportunities for and threats to development.The advantage of a GIS is that it allows for an efficient and effective analysis based on a wide set of criteria, the com- plexity of which may not lend themselves to traditional evaluation. The technology can evaluate complex relationships that are otherwise not readily apparent.

Because GISs are not widely applied in the tourism industry,’ this article offers a definition of GISs, explains their applications, advantages, and disadvantages, and then illustrates the applicability of using GISs to site resorts in a case study from Costa Rica. We applied GIS siting analysis to a coastal area comprising nearly 40,000 hectares.

’ For another application of GIS, see: Michael Schriber, Christopher Muller, and Crist Inman, “Population Changes and Restaurant Success,” Cornell Hotel and Restarrrant Administration Quarterly,Vol. 36, No. 3 (June 1995), pp. 43-49.

This study examines the i:i

We analyzed this area’s development prospects based on a number of siting criteria and found a total area ofjust over 4,000 hectares that is potentially suitable for coastal hotel development.

What a GIS Is A GIS is essentially a spatial database that features geographic and at- tribute data.The geographic data define the location of attributes, such as hotels, roads, and wetlands.2

’ David J. Maguire, “An Overview and Defini- tion of GIS,” in Geographical Information Systems: Principles andApplications, Vol. 1, ed. David J. Maguire, Michael E Goodchild, and David W. Rhind (Harlow, UK: Longman Scientific and Technical, 1991), pp. 9-20.

Peninsula de Nicoya, on

the northwest coast of

Costa Rica.

On a GIS map these attributes are shown as points,

lines, and two-dimensional shapes. As Stan Aronoff points out, the full GIS is a system that includes the input, management, manipula- tion and analysis, and output of geo-referenced data.3 Through geo-referencing a GIS allows the analyst to overlay (that is, to make unions or intersections o!tJ

multiple attributes on a map and thus examine the spatial relationship of, say, a possible hotel location to a road or wetland.

Application and uses. The ability to store and analyze spatial data affords many uses for a GIS. It can be used in diverse applications, such as land-use planning and management, infrastructure siting, site management, market analysis,” auto-

‘Stan Aronoff, Geographic Information Systems: A Management Perspective (Ottawa, Ontario: WDL Publications, 1991).

4 John R. Beaumont, “GIS and Market Analy- sis,” in Geographical Iflformation Systems: Principler andApplications, Vol. 2, ed. David J. Maguire, Michael F. Goodchild, and David W. Rhind (Harlow, UK: Longman Scientific andTechnical, 1991),pp. 139-151.

August1999 l 49

mobile navigation systems,s and emergency planning. Indeed, any information that can be referenced geographically can be analyzed in a GIS system.Thus, one could examine on a single map natural-resource data, infrastructure data, demographic data, and political boundaries, for instance.

We found few studies that apply GIS to tourism.The work that has been done includes that of Sussman and Rashad, who explored the uses of GIS in tourism marketing research;’ Boyd and Butler, who used GIS to identify potential ecotourism sites;’ and Berry, who developed a GIS-based map analysis seeking to optimize an island’s resource use among the conflicting land uses of conservation and recreation, ecological research, and residential development8

GIS Advantages and Disadvantages

Because the digital data in a GIS are particularly efficient for storage, transfer, and retrieval, a GIS allows the user to see and analyze complex spatial relationships that may not be readily apparent. For example, an

’ Marvin White, “Car Navigation Systems,” in Geqmphical I$wmarion Systems: Primiplcs and Application, Vol. 2, ed. David J. Maguire, Michael E Goodchild, and David W. Rhind (Ha&w, UK: Longman Scientific andTechnical, 1991), pp. 115-125.

’ Silvia Sussman and Tamer Rashad, “Geo- graphic Information Systems in Tourism Market- ing,” in Prqrcss in Tourism, Rematiun, and Hospi- t&y A4anapwtcrzt, Vol. 6, ed. Chris Cooper and Andrew Lockwood (London: Belhaven Press, 1994), pp. 250-258.

‘Stephen W. Boyd and Richard W. Butler, “Seeing the Forest through the Trees: Usmg Geographical Information Systems to Identify Potential Ecotourim Sites in Northern Ontario, Canada,” in Practiciq Respmsible Totrrism: Irrterna- tional Case Studies in Towism Plarrwing, Policy, arld Development, ed. Lynn Harrison and Winston Husbands (NewYork: John Wiley & Sons, Inc., 1996), pp. 380-403.

‘Joseph K. Berry, “GIS in Island Resource Planning: A Case Study in Map Analysis,” in Geqraphical Information Sysfems: Principles and Applications, Vol. 2, ed. David J. Maguiie, Michael E Goodchild, and David W. Rhind (Harlow. UK: Longman Scientific andTechnical, i991), pp. 285-295.

Exhibit 1 Spatial requirements for a coastal hotel

1. Coastal proximity l Walking distance or adjacent to the beach.

2. Suitable building sites 9 Sufficient space having no need for special foundations (which would increase

construction costs); l Site free from flooding; and l Suitable for on-site wastewater disposal.

3. infrastructure l Access to roads, as it is economically unfeasible for the hotel developer also

to build roads; and l Access to site for deliveries, employees, and guests.

4. Environmental amenities l Areas of natural character in close proximity; l Proximity to national parks or other natural land uses; and l Not located in an environmentally sensitive area.

Note: Above criteria apply to a coastal hotel sited in northwestern Costa Rica based on surveys of developers in the region and government officials.

operator who is considering ex- panding a ski resort might want to map all locations in a given area that have households earning more than $55,000 per year living within 150 miles of the resort. Using the GIS data, the would-be developer could then overlay that map with one that shows environmentally appropriate areas for development of additional ski lifts and condominiums. Further- more, the developer can narrow (or widen) the scope of the search as appropriate, because digital data offer the flexibility of allowing the user to work at multiple scales. This means that data that are ini- tially represented at a scale of, say, 1:50,000 can be used in the same analysis with data collected at a scale of 1:20,000 (as long as the user rec- ognizes the limitations of data reso- lution when moving from broad to narrow scales). Additionally, a GIS offers the opportunity for iterative analysis including sensitivity analysis to changes in the output area and variations in any individual criteria. That approach is similar to the sensitivity analysis that might be performed in a real-estate-valuation model, but the GIS allows multiple

iterations given changes in base assumptions. For instance, one might look at potential hotel sites within 1,000 meters of the ocean and then check on sites within 2,000 meters.

Cost and expertise. The primary disadvantages of GISs are the associated costs and technological expertise required.The cost for a complete GIS is currently approximately $5,000 (not including data, but including the necessary hardware and software). Fortunately, the cost has decreased substantially in the past five years, while computers have improved to the point that GIS technology operates on a desktop computer. Although the software has become increasingly accessible and user friendly, the technical expertise and training required to run a GIS is still a drawback.

GIS Input Data

One can examine the following types of data using GIS technology:

l Environment (e.g., geological, marine, ecological, hydrological, and atmospheric);

l Census (e.g., demographic, socioeconomic);

50 IIIK~ El I , , HOTELAND RESTAURANTADMINISTRATION QUARTERLY

l Infrastructure (e.g., location of roads, power, bridges, airports, and communications);

l Political and physical boundaries; l Landforms (e.g., topography and

terrain); and l Land use.

The fact that a GIS can analyze such a wide variety of data is one aspect of the technology that makes it so versatile and useful.The user is limited only by access to the necessary data, and those are becoming increasingly available.

Data collection. The most time-consuming and expensive portion of GIS data collection em- ploys primary methods, such as surveys, fieldwork, remote sensing, and image processing. However, an increasing amount of data is already collected and available.

Such secondary data include maps and government census information.These data often have application to more than one type of analysis and are usually available through government and nongovernment agencies and academic institutions. Reference resources are available that identify the location of existing data sets and their accessibility.’ While the necessary data are often already digitized, one still must plan to input data.This is commonly done by tracing attributes on an electronic tablet to encode map attributes and their georeference position for computer analysis. Needless to say, in terms of time and cost, using existing digital data is typically more efficient and effective than developing new data.

‘See, for instance: Peter F. Fisher,“Spatial Data Sources and Data Problems,” in Geographical Infornzation Systems: Principles and Applications, Vol. 1, ed. David J. Maguire, Michael F. Goodchild, and David Khind (Harlow, UK: Longman Scientific andTechnical, 1991), pp. 175-189. Map sources include: USGS-topo- graphic, land cover, and other thematic maps; U.S. Bureau of the Census-TIGER data; USDA Soil Conservation Servicesoil maps; and U.S. Fish and Wildlife Service--national wetland inventory.

Exhibit 2 Information used in G/S analysis

Spatial requirement Source

Study area Political-boundary maps at a scale of 1:500,000 were copied from a 1995 Official Political Division by the University of Costa Rica (UCR) at the National Geographic Institute (IGN) and digitized using ARCINFO following USA standards for digital cartography.

Coast Coastal zone maps at a scale of 1500,000 delineation were copied from a 1995 Official Political Division by UCR at the IGN and digitized using ARCINFO following USA standards for digital cartography.

Suitable A soils map at a scale of 1:200,000 prepared by ACON in coopera- building site tion with Ministry of Agriculture and Grazing (MAG). This map uses

the Costa Rican soil capacity classification system (MAG 1991). Digitizing was performed on ARCINFO according to USA standards for cartographic data.

Infrastructure A road map at a scale 1:200,000.

Environmental A 1992 Land Use Map at a scale of 1:200,000 prepared by Clemson

attraction University and the National University of Costa Rica. The map was prepared using a Landsat TM composite image that was interpreted manually then digitized using ARCINFO GIS software. The data were visually field checked in 1995; the quality of the data is unknown.

National conservation areas were digitized from “Los Parques Nacionales y Otras Areas Protegidas de Costa Rica” (the national parks and protected areas of Costa Rica), at a scale of 1:500,000, using ARCINFO.

Indigenous reserves were digitized from “Los Parques Nacionales y Otras Areas Protegidas de Costa Rica,” at a scale of 1:500,000, using ARCINFO.

Notes: All maps were geo-referenced using the Costa Rica Lambert North projection.

Unless otherwise noted, the maps were distributed in 1996 by the United Nations FAO. The soil map was produced by ACON and Ministerio de Agricultura y Ganaderia, and the land-use map was prepared by Clemson University and the National University of Costa Rica. The producers of the other maps are not known.

Data analysis. The process of Defining the spatial requirements. using a GIS in a resort-siting problem consists of the following six steps: defining the problem, selecting the study area, defining the spatial requirements of the problem, assessing spatial data needs, defining the diagnostic land characteristics, and creating the land evaluation.

Defining the problem. One first formulates the spatial question to be answered (e.g., For what type of tourism land use are we seeking a site?).

Selecting the study area. Next, one defines the area that is to be considered in the analysis.This area can be quite broad to begin with and can subsequently be narrowed.

Once one has determined the type of land use in question and the area to,be evaluated, one next must ex- plore the prospective spatial requirements. The spatial requirements of the land use refer to the data necessary for the analysis process, including the environmental, economic, and social conditions necessary for the successful development of the tourism activity. Those spatial requirements can be determined sim- ply through past experience or be set according to statute, industry standards, or surveys of existing operations. In defining the spatial aspects of the siting problem, one determines the necessary resource

August 1999 l 51

Exhibit 3 Decision tree for coastal-hotel diagnostic characteristics

s

Environmental, legal

A stepwise diagnostic approach narrows the ‘\ c

En virontr-+- J r w set of potential development sites, according \

! to criteria established for coastal hotels. If a 5 given diagnostic characteristic is acceptable ‘\

for a site (shown in color), the analyst can \ I test that site against the next criterion. An e

unacceptable outcome on any criterion ends \

the consideration of a site. t

a Environmental.

\ ec

a b

G

\@ ‘0

mental, ,̂ -̂ I A

Must reject

site

G- Economic

relationships and the limits, boundaries, and categories of spatial attributes that determine the suitability of land for a specific use.

Assessment of spatial data needs. After defining the spatial requirements of the problem the spatial data needs are assessed. In this step the analyst determines what data are needed to make the siting decision. One must determine what data are

available or must be developed regarding the study area, such as information on building sites, infrastructure, and local population size and constitution, for instance.

Dejning diagnostic land characteristics. The diagnostic characteristics in question are those measurable components that allow the analysts to make an evaluation of the land for possible development sites. When

conducting a GIS analysis it is help- ful to create a decision tree that illustrates the critical elements of the analysis process based on the resource needs of a particular tourism activity. GIS users should begin the decision tree with the most limiting factor of the analysis. To give a simple example, the building site for a coastal hotel must be near the ocean.

52 COKNEH i HOTELANDRESTAURANTADMINISTRATION QUARTERLY

The decision tree drives the organization and examination of diagnostic land characteristics, which are the measurable (or estimated) attributes of the land.“’ Examples of land characteristics include availability of water, distance to services or infrastructure, rainfall, slope gradient, and climate. Land characteristics are often determined from maps of resource inventories, surveys, and census data.

Lund evaluation. With all the available data in hand, one finally com- pares the required characteristics of the prospective tourism land use with the conditions present in the study area. The analysis categorizes the region under study according to each section’s suitability for development, as determined by the decision tree that shows the diagnostic land characteristics.The result of the analysis is a set of sites that are likely to meet the decision criteria (e.g., a site near the coast that also has a road nearby and is not on a cliff). One advantage of GIS methodology is that once the initial tourism activity is defined, the spatial requirements and diagnostic characteristics needed to support that activity can be applied to any given study area or to several areas at a time.

A Case Study of Norfhwesfern Costa Rica We demonstrate the six-step GIS process described above in an examination of possible development sites on the northwestern coast of Costa Rica. We chose Costa Rica for this case study because the country to date represents one of the great success stories in conservation and tourism. While no country is perfect, Costa Rica’s conservation efforts and its national focus on sustainable development represent the centerpiece of its tourism indus-

“‘Food and Agriculture Organization (FAO), Soils B&tin 32:A Frameworkfor Land Evalualion (Rome: Food and Agriculture Organization, 1976).

try.” That industry has expanded steadily over the last decade and has become one of the nation’s most important means of earning foreign exchange. Because of Costa Rica’s commitment to sustainable tourism development, the country is seeking an integrated approach to evaluating siting and planning problems as increasing tourism development puts pressure on environmentally fragile areas. GIS analysis offers the quantitative and systematic approach to assessing land for tourism uses that makes sustainability possible.

Our Costa Rican case study is based on expert knowledge that we developed from surveys of hotel owner-operators, existing geographic databases, field observations, and the opinions of industry profes- sionals.‘2 We conducted primary research in the form of a survey to determine the economic, social, and environmental considerations of hotel developers and operators in the northwestern coastal zone of Costa Rica. We collected further environmental data from the Costa Rican national government, and the United Nations Food and Agricul- ture Organization (FAO) . The box on the next page describes in more detail the specific research methodology and some of the findings.

Step-by-step Analysis Costa Rica has extensive Pacific and Caribbean coastal zones. We chose to analyze the possible siting of a coastal hotel, because such coastal hotels are the most common type of development in the region and they have some of the greatest environmental ramifications.

The study area. The northwestern coastal zone is the fastest-

“Mario A. Boza,“Conservation in Action: Past, Present, and Future of the National Park System of Costa Rica,” Conservation Biolqy,Vol. 7, No. 2 (1993), pp. 239-247.

“Author Joerger has spent considerable time in the field researching Costa Rica’s land-use situation.

Exhibit 4 Land-use classifications in Costa Rica

Land- Area Natural

Uee (square bnd

code Land uee or cover km) use?

8 Pasture 3,287 No 9 Permanent crops 939 No 10 Rocky terrain 51 No 11 Seasonal crops 530 No 12 SecIondafy fore& l,@f.Yes 13 Tacotal 1.093 No

ii?9 No I I 85 Yes

growing tourist destination in Costa Rica. This region is a logical place to develop a tourism industry due to its natural beauty, its beaches, and its sport fishing-not to mention available land. Since environmental amenities are the big attraction to northwestern Costa Rica, the sen- sible course is to pursue development plans that protect the resources supporting that tourism.

During our fieldwork we ob- served (and discussed with other researchers) critical land-use conflicts within the rapidly growing tourist economy. For example, a banana plantation in the Atlantic lowlands was causing siltation and nutrient loading in bordering reefs, thereby damaging conservation efforts and spoiling the area’s tourism potential. In another instance, hotels themselves were degrading water quality due to poorly planned on-site wastewater disposal. Further- more, one hotel removed a man- grove forest to accommodate a ma- rina. Not only does that reduce the stability of the coastline, but it re- moves some of the area’s tropical charm, of which mangroves are a part.We also heard reports that the

August 1999 l 53

Field Research to Gather GIS Data The projects considered here were hotels located in the coastal zone of northwestern Costa Rica. The formal survey underlying the accompanying article went to hotel owner-operators and government officials, but we did not formally survey local residents or visitors. Our first-hand observation was that hotel developers were not taking a proactive role in the evaluation of the environmental resources that supported their projects. For example, it was clear on inspection that operators were not considering water quality, and insufficient wastewater treatment was degrading water supplies. That is why we directed our survey at hotel owners and managers. We wanted to determine the type of environmental, economic, and social criteria used to site a hotel in the research area.

Twenty hotel owner-operators participated in the survey. The hotels, located on the Peninsula de Nicoya (from Playa Tambor on the southern tip of the peninsula to Playa Panama on the northern extreme of the peninsula), ranged in size from six rooms to over 135 rooms. The operators included local people, Costa Rican nationals, foreign investors, and on-site representatives from international

We administered the survey orally in person either in English or Spanish, at the respondent’s option. Each survey took approximately one hour to complete. Covering seven thematic areas, the survey comprised both multiple choice and open-ended questions.

Our questions related to information needed on environmental, social, and economic resources. We also collected information on occupancy, seasonality, ADR, number of rooms, construction costs, siting decisions, community impact, and guest characteristics, as well as the previous land use at the hotel site. Additionally, we asked respondents to consider the relationship of the environment to social and economic factors used by operators and owners when siting hotels.

Social resources. Those individuals constructing and operating hotels in the Nicoya Peninsula in northwestern Costa Rica did not consider social or human resources. That is, hotel owners and developers generally disregarded the local community when choosing their hotel site. Few respondents chose their site based on perceived labor availability (1 respondent), friendliness of the host community (3), local culture (0), or low population density (1). None of the respondents reported considering the character of the local community when siting their hotel property.

Economic resources. Not surprisingly, economic success was typically the focus of feasibility studies conducted by or for hotel developers. The majority of respondents used resources from their country of origin to finance their hotel project. They reported that debt leverage was unimportant, but the perceived life-style and a life change appeared to be strong motivators for developing a coastal hotel. Many owner-operators wanted a financially successful activity to fill their day while they lived in what they thought was one of the most beautiful places in the region. Some of the respondents located in specific areas because of the existing infrastructure.

The survey questions included the following. Words in parenthesis

are examples of prompts we used to elicit specific information on each question.

1. Why did you locate in this particular area? (Attraction, infrastructure, already owned the land, labor availability.)

2. Did you perform a feasibility study before you built or purchased this property? (What information was in the study?)

3. What are the most important economic reasons for locating in this location? (Cost, occupancy, seasonality, debt.)

4. What do you perceive as the impact of this hotel on the community?

5. Did you choose this location because of the community? (Labor availability, friendliness of people, lack of people, local culture.)

6. What are the most important environmental reasons for building on this site?

7. What is the nature of your clientele? (Country of origin, age of guests.)

In the national capital, San Jose, we interviewed many government officials and tourism experts. Government and nongovernment agencies that we surveyed included Emergency Planning, the Cadastral Survey Office, and the Ministry of Agriculture and its partner, the FAO. The national government of Costa Rica reported collecting and using environmental data for a variety of tourism- planning activities, including: (1) determining tourism-project locations, (2) monitoring and evaluating tourism impact, (3) establish- ing tourism policy, and (4) regulating the tourism industry. As we noted in the accompanying article, considerable thought is given at a national level to tourism planning, and the government is quite progressive about tourism with regard to the environment. Neverthe- less, our interviews with the hotel owner-operators made it apparent that the local projects use little of the information collected for national tourism planning.-A.J., S.D.G., and MAN

We gained a loo-percent response rate in part because we conducted the research in person during the low season and respondents had time to participate. We also found favorable name

recognition for Cornell University by hotel operators. One respondent noted, for instance, that she had attended the Professional Develop- ment Program at the School of Hotel Administration.*

The survey revealed that proximity to the coast and a generally positive attitude toward the environment were the hotel owners’ main environmental concerns. Hotel developers also wanted access to the coast and pleasant scenery or views.

Environmental resources. While coastal access was a primary focus, with 11 of the 20 respondents citing the coast as a top consideration, the respondents also considered other environmental resources. Fourteen respondents cited one or more environmental attractions, namely, proximity to nature (6 responses), water features (5) forest (4) wildlife (3) and national parks (2).

*See: Glenn Withiam, “Cornell Executive Education: Student Number 20,000,” Cornell Hotel and Restaurant Administration Quarter/x Vol. 40, No. 3 (June 1999), p. IO.

region’s water table is declining due Northwestern Costa Rica is typi- Defining Spatial Requirements to excess consumption. Given those cal of areas that would benefit from anecdotes, one can see the potential siting properties by a GIS analysis

First, one determines the types of

negative impact of further develop- that integrates many environmental land attributes that will permit the

ment to the resource base that sup- concerns to mitigate resource-use successful development and opera-

ports tourism. conflicts. tion of a coastal hotel property. One must collect information regarding

54 CllKizlEll HOTELANDRESTAURANTADMlNlSTRATlONQUARTERLY

the germane attributes of the spatial requirements for a coastal resort, as well as relevant land-based opportunities and threats.We did this via interviews and surveys of stakeholders, owners, operators, and government and nongovernment officials.

We conducted a formal survey of hotel owner-operators and government officials, as described in the box on the previous page. The survey sought to determine the type of economic and environmental criteria used to site a hotel in the research area by focusing on the attributes of hotels already located in the region. Responses indicated that hotel developers were not taking a proactive role in evaluating the very environmental resources that supported their projects, as in the case of deteriorating water quality.

Hotel operators’ primary focus was coastal access-cited by ten of the 20 respondents as a consideration when siting their hotel. Other important environmental factors were proximity to nature, water features, forests, and national parks.

We then interviewed government officials and tourism experts in San Jose, meeting with representatives of Emergency Planning, the Cadastral Survey Of&e, and the Ministry of Agriculture and its partner, the FAO.We found that the government at a national level gives considerable thought to and is quite progressive about tourism planning with regard to the environment. Nevertheless, local developers seem to use little of the information collected for national tourism planning, as we discovered from our interviews with the hotel owner-operators.

Analyzing the data from the surveys and interviews and drawing on expert knowledge, we developed the list of spatial requirements for a successful siting of a coastal hotel in Costa Rica, as shown in Exhibit 1. Those requirements include: coastal proximity, environmental amenities,

Exhibit 5 GIS map of Costa Rica’s Peninsula de Nicoya

Each patch on the above map indicates a geographical feature to be considered in the GIS analysis, such as roads, existing hotels and wetlands. In this case, the geographic feature must occupy at least six square kilometers to register.

periodic flooding, on-site waste- a suitable building site free from

water management, and access to infrastructure.

Assessing Data Needs Starting with the spatial requirements for a coastal hotel shown in Exhibit 1, we further refined several data categories. For each requirement the analyst must determine what type of data would provide the information necessary to conduct an accurate evaluation.

To locate suitable potential building zone of northwestern Costa Rica.

sites we turned to regional soil maps that indicate slope gradient, suitability for construction, and suitability for on-site wastewater management, as well as risk of flo0ding.A road map showed which sites had a highway nearby. Finally, a land-use map and a map of national conservation areas showed where the environmental attractions might be. Exhibit 2 specifies the maps we used, which are in digital form for this analysis.13

Starting at the beginning, for instance, we used a map of political boundaries to define the coastal

l3 Digital data were provided by Cornell Uni- versity and the FAO.

August 1999 l 55

Exhibit 6 GIS spatial-analysis process

Each factor in the above set of boxes is assessed according to the criteria established for the analysis...

. ..to develop a list of limitations on potential sites.

These limiting criteria are then overlaid on the GIS map to create a potential set of development sites.

Defining Diagnostic Characteristics After determining spatial requirements, assessing the data needs, and collecting the digital data, we next defined the diagnostic land characteristics, or the limiting factors for development. These key attributes are drawn from the interviews with developers, owners, stakeholders, and experts. We evaluated each land characteristic for appropriate limits and thresholds. A component of this stage is a decision tree that illustrates the decision process, as shown in Exhibit 3.

Highways. The first node of the decision tree (after eliminating sites that are not adjacent to the coast) considers the distance from existing roads. This is primarily an economic

factor, although environmental con- Water. The second decision, siderations do enter in. The scale of distance from the water, is influ- a coastal hotel precludes costly in- enced by economic, environmental, vestment in extensive access roads and legal constraints. A viable site is and related infrastructure. Given the located more than 50 meters but distance of existing hotels from the less than 1,000 meters from the roads that serve them, we deter- coast. The economic aspect of this mined that an appropriate distance constraint addresses guests’ prefer- for a development site is no more ences for direct access to the coast. than 1,000 meters from an existing Based on our surveys of existing road. That distance allows for buffers properties, we believe that a guest from existing development (and the can and will walk 1,000 meters to highway itself) without incurring get to the ocean, but an operator disproportionate costs for new infra- would have trouble claiming prox- structure. We also considered the imity to water at a resort that is tendency of new-road construction more than 1,000 meters from the to cause environmental degradation, ocean. The environmental compo- since the land must be cleared and nent of this decision point again the construction itself and subse- suggests that viable sites would be quent runoff can damage the land. within 1,000 meters of the coast,

56 CIIKNFII i 1 I HOTELAND RESTAURANTADMINISTRATIONQUARTERLY

Overlay

Overlay

I Map I

The sites meeting those criteria are in turn further overlaid to fine-tune the list, or to fine-tune the criteria...

. ..until the final criteria can be applied.

The final set of criteria overlaid on the map yields the most likely development sites.

since maritime views, flora, fauna, and sounds diminish beyond that distance. The legal constraint on distance from the water is Costa Rica’s ban on development within 50 meters of the mean high tide line.

Soils. The third decision factor, selecting suitable soils, restricts development to areas that are appropriate under Costa Rica’s soil- classification system.This classification integrates the need for afford- able and effective on-site wastewater disposal, lack of inundation, and minimal erosion risk and resulting sedimentation. Suitability for on-site wastewater disposal is determined by analyzing soil texture, drainage, slope gradient, and soil depth. Sites

minimize run-off (although a 15-

that are likely to be flooded are

percent slope is relatively steep).14

eliminated, and the slope gradient should be less than 15 percent to

Soils in Costa Rica’s classes I, II, or III have the appropriate soil qualities for a coastal hotel.” The environmental and economic components of this decision point are best illustrated with regard to on- site wastewater disposal. Proper soil conditions make on-site waste dis-

I4 United States Department ofAgriculture, National Soil Survey Interpretation Handbook Draft (Washington, D.C.: GPO, 1992).

‘s Minister% de Agricultura y Ganaderia, Manual Descriptive de la Leyenda del Mapa de Asociacioner de Subgwpos de Suelor de Costa Rica

&ala 1:200.000 (San Jose: Ministerio de Agricultura y Ganaderia, 1991).

wastewater and it must be treated in

posal (with septic tanks and leach

sewage plants--or, worse, isn’t treated at all. Extremely saturated soils and

fields) less costly than locations

soils with free water at a shallow depth, for instance, limit the use of

where the soil cannot handle the

septic fields.16 The alternative to proper treatment is not acceptable, because discharges of untreated waste risks degrading coastal water quality (which is why guests come to the resort in the first place).Thus, the soil analysis primarily involves an environmental analysis, though soil properties have an economic component.

I6 United States Department of Agriculture, Soil Survey Manual (Washington, D.C.: GPO, 1993).

August 1999 l 57

Exhibit 7 Potential development sifes in Peninsula de Nicoya

Coastal areas that meet soil, road, and environmental criteria are shown in color. Areas that meet only soil or soil and road criteria are shown in heavy black. The web of gray lines within the peninsula represent highways, including the Pan American Highway. The map in Exhibit 8 focuses on the section of the coast within the circle.

soil for construction, our analysis With regara to tne suitability of

considered only the possibility of building relatively small properties (i.e., up to three stories in scatter- site style). Construction-related factors include slope, inundation, mass movement, depth to bedrock, shrinking and swelling, presence of rock fragments greater than 75 mm in diameter, erosion potential, sub- sidence, and soil strength.

only in the coast.The table in Ex- spots than people who are interested

hibit 4 gives the suitability classification of land uses, while Exhibit 5 shows the GIS map of land uses.

Sensitive areas. The final decision point integrates both environmental and legal components. To avoid environmental destruction, hotels should not be located in sensitive environmental areas. More- over, Costa Rica bans development within national conservation areas. Attractions. The final two deci-

sion points address tourists’ desire to see and enjoy natural sites besides the ocean. Suitable tourist sites are within 2,000 meters of environmental attractions, which include forests, lakes, wetlands, and mangroves. We chose 2,000 meters as the limiting distance because we believe that tourists interested in environmental attractions would be willing to travel farther to see those

GIS Land Evaluation

With the completion of step five, the preparation for a land evaluation is complete. The GIS analysis classi- fies the spatial data according to the decision tree in Exhibit 3.The most

limiting diagnostic characteristic determines the suitability of a given piece of land (map unit). Each piece of land is either suitable or not suit-

able, depending on whether it meets all the criteria in the diagnostic land characteristics.

The following categories of data were considered against the criteria in the decision tree.

(a) Twenty soil classes, shown on the Costa Rican soil-capacity map, were classified as suitable or not suitable based on expert analysis;

(b) Twelve land use types (e.g., wetlands, urban, brush, secondary forest, disturbed natural forests) were classified suitable or not suitable based on survey results;

(c) Seven national conservation areas were classified as not suitable based on Costa Rica law;

(d) Areas less than 1,000 meters from roads were classified as suitable based on expert analysis; and

(e) Areas more than 50 meters but less than 1,000 meters from the coast were classified as suitable based on survey results.

Spatial analysis. Here’s how the spatial-data analysis proceeds, as illustrated in Exhibit 6.” The political-boundaries map (at a scale of 1:500,000) is re-selected for Costa Rica’s northwestern coastal regions and given the coverage name Coust (top line of Exhibit 6). Coast coverage is buffered from 50 to 1,000 meters, which becomes the extent of the study area, termed Coast Blrffer.The soil map at a scale of 1:200,000 (next line) is reclassi- fied into suitable and unsuitable soil based on the factors of wastewater disposal and a maximum 15-percent slope. Suitable Soils and Coust Bufir are overlaid and the areas common to both coverages give the Coast- Soil Composite, comprising sites that meet these two sets of criteria.

The road map at a scale of 1:200,000 (third line) is buffered to 1,000 meters (that is, to select sites within 1,000 meters of a road) and

“PC AKCXNFO (Computer proyram), Environmental Systems Research Institute, Redlands, CA, 1994.

50 CURNEIL i HOTELANDRESTAURANTADMlNlSTRATlONQUARTERLY

is labeled Road Butr. Road Bug@ is overlaid with Coast-Soil Composite to form yet another composite of areas that fit three sets of criteria, Coast-Soil-Road Composite. This defines the coastal areas with suitable soils for development that are accessible via preexisting roads.

The 1992 Land Use Map is reselected and called Natural Land Uses (bottom boxes, Exhibit 6). Conservation areas, which comprise national parks, protected areas, and indigenous reserves, are overlaid with Natural Land Uses and defined as Environmental Attraction. Environ- mental Attraction is, in turn, buffered to 2,000 meters (as discussed above) and called Environmental Attraction Bu$er. Environmental Attraction Bufir is reselected for suitable environmental attractions to exclude conservation areas, wetlands, mangroves, and lakes, yielding Suitable Environ- mental Attraction.

Suitable Environmental Attraction is overlaid with Coast-Soil-Road Com- posite.The areas common to both coverages result in Coast-Soil- Road-Environmental Composite, a set of sites that fit all of the criteria we established.The resulting set of sites comprises areas suitable for coastal hotel development due to their proximity to the coast and their fulfilling the requirements of suitable soils and accessibility. Addition- ally, the selected areas are proximate to environmental attractions without being located within an environmentally sensitive area.

The map in Exhibit 7 shows the potential development sites in the entire study area of Costa Rica’s northern Pacific coast. Those areas that are shown in color meet all our criteria.

The map in Exhibit 8 shows the spatial analysis for a small section of the Pacific coast, showing existing hotels as dots.The areas that meet all our criteria are shown in color, while the areas that meet only one

Exhibit 8 Potential development sites near Jabilla, Costa Rica (see circled coastline in Exhibit 7)

Meets soil, road, and environmental criteria

Meets only soil criteria

Meets soil and road criteria

0 Existing hotel

or two criteria are shown in hatch- ing or crosshatching.

The output of the physical evaluation for the entire site was as follows:

l The total study area is 39,600 hectares;

l The area of suitable soils is 14,100 hectares;

l 7,300 hectares of suitable soils are accessible by existing roads;

l 4,400 hectares of suitable soils are accessible by existing roads and are close to natural areas but are not located within environmentally sensitive areas.

Worth the Trouble The map in Exhibit 8 shows four of the five hotels that are located on sites deemed suitable by the GIS analysis. Of the 20 hotel owner- operators we surveyed for this research, we found that the five ap- propriately located hotels have a higher RevPAR than the other 15 properties. This finding suggests that the hotels located according to the GIS criteria are currently economically more viable. We believe, there- fore, that basing siting decisions on spatial criteria is not only more effective and efficient than other approaches, but such an approach also offers a positive economic outcome.

The hotels that we studied have similar construction styles, management teams, and target markets. Because the main differentiating characteristic of these hotels is their location, we suggest (based on this study) a relationship between appropriate siting and profitability. GIS analysis is a rigorous approach to land evaluation that the hotel industry can use to considerable benefit. Operators in a wide range of indus- tries have already applied GIS to identify appropriate operating sites. Retailers and restaurant operators, for instance, are increasingly using GIS as part of their market research in siting new properties.

As this technology continues to be developed, the cost of the associated hardware and software will decrease. Moreover, digital data are also becoming more readily available, thus fostering GIS analysis. For example, automobile navigation systems are driving the collection of GIS data. Improvements in the technology have decreased the learning curve for GIS operators and made GIS analysis more user friendly. Thus, we predict that GIS will in the near future become an integral component of real-estate, tourism, and other hospitality-industry siting and planning decisions. CQ

AuQust 19% l 59

Applying Geographic Information Systems: Siting of Coastal Hotels in Costa Rica

Documents