Participatory GIS for Integrating Local and Expert

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning Biancamaria Torquati University of Perugia, Italy Marco Vizzari University of Perugia, Italy Carlo Sportolaro Agronomist, Perugia, Italy

ABSTRACT This chapter describes the development and implementation of an operational method for integrating expert and local knowledge with new technologies for geographic mapping and communications, to enhance cultural landscape analysis and planning. Topics include the following aspects: a) analysis of type(s) of information required to construct a geographic information system (GIS), with the landscape as a common objective; b) method of implementing and integrating various types of expert knowledge in the GIS; c) method of collecting, organizing, and structuring local knowledge in the GIS; d) method of integrating expert with local knowledge; e) exploration of GIS functions. The main aim of this work is to examine the possibility of using participatory mapping methods and GIS for comparison and integration of multidisciplinary scientifc expertise, local knowledge, and landscape project proposals. In particular, it involves specifc methods for enhancing local features of vineyard landscapes through a participatory process developed with both vineyard entrepreneurs and the local population. The case study concerns the wine-growing area of Umbria, a region of central Italy.

INTRODUCTION The European Landscape Convention (ELC), which defines the landscape as “an area as perDOI: 10.4018/978-1-60960-621-3.ch020

ceived by people, whose character is the result of the action and interaction of natural and/or human factors”, introduces two levels of evaluation, of equal importance: objective (“the area”) and subjective (“as it is perceived by people”)

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

(Lambertini, 2009). Consequently, landscape perception by locals becomes a focal point in landscape definition, analysis and interpretation (European Council, 2000). In this respect, the ELC promotes participation and raising of awareness in local populations by means of direct involvement in transdisciplinary planning (Antrop and Sevenant, 2009). Three aspects are believed to be important, if regional resource management is to be meaningful to the inhabitants involved (Brunckhorst et al., 2005): 1) to define the areal proportion of the region which inhabitants consider to be part of their “community”; 2) to select landscape units within the region with a high degree of homogeneity, which increases the coincidence of interest among the inhabitants; 3) to consider externalities of resource use. In addition, in the last twenty years, participatory GIS (P-GIS) has been widely used for participatory spatial planning (PSP), assuming that this use of GIS is a tool for better governance (McCall, 2003). This chapter investigates the validity of participatory mapping methods and GIS for landscape planning of vineyard areas in central Italy. It attempts to answer the following questions: a) what type(s) of information are required to construct a geographic information system (GIS) with the landscape as a common objective? b) how can the various types of expert knowledge be implemented and integrated in the GIS? c) how can local knowledge be collected, organized and structured in the GIS? d) how can expert and local knowledge be integrated? e) how can GIS functions be exploited in order to improve landscape understanding and communications? The chapter begins in Section 2, describing the background of participatory GIS methods. Section 3 discusses the goals, conceptual and methodological approaches, and summarizes the field surveys and specific analyses performed. Section 4 considers the procedures followed to inform and consult local people for landscape planning, integrating scientific and common local knowledge. Section 5 considers the procedures

followed to involve and collaborate with local people, in order to improve a landscape master plan. Sections 6 to 8 draw some conclusions about the potential of P-GIS to make landscape governance easier, and outline possible future research directions.

PARTICIPATORY GIS (P-GIS) METHODS Landscape planning problems are often complex, require the contribution of experts from differing fields, engage both private and public institutions, involve stakeholders in diverging interests and, generally, affect a large number of people (Voss et al., 2004). As highlighted in the European Landscape Convention (ELC), landscape perception by locals becomes a focal point in landscape analysis and interpretation (European Council, 2000). Thus, their participation becomes an increasingly important component in landscape decision-making (Vajjhala, 2006). In traditional planning, local communities are not directly involved in landscape analysis, and they may express their opinions only during the final phase of participation. Conversely, modern participatory spatial planning (PSP) is based on the involvement of citizens and other stakeholders in the various phases of the process. In the PSP approach, participants may influence the final contents of the plan and have the opportunity to enhance their awareness about the characteristics and values of “their” landscapes. Participation may have a variety of additional goals, from the building of relationships (as in communicative planning) to prescription or goal selection (as in instrumental rationality) (Talen, 2000). A multiplicity of participatory approaches exists, aiming at involving the public and integrating local experiences and preferences with experts’ scientific knowledge within the planning process. One effective method is to use mapping and spatial information technologies, such as geographic

379

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

information systems (GIS) (see, e.g.,: Craig et al., 2002, Dunn, 2007, Jankowski, 2009, Sieber, 2006, Tulloch, 2008). These powerful tools are designed to manage, transform and represent georeferenced data (Jones, 1997, Smith et al., 2007) as well as to disseminate geo-information and related services to the general public by means of transparent communications exploiting the power of the internet (Kuntu-Mensah, 2006). GIS systems have shown themselves to be very effective in many fields of local governance (Longley et al., 2005), including economic development, transportation and service routing, housing, infrastructures, healthcare, tax maps, human services, law enforcement, land-use planning, parks and recreation, and environmental monitoring. In all these contexts, proper use of GIS technology can improve good governance, enhancing legitimacy, participation, transparency, respect for rights, empowerment, information access, and equity (e.g., Corbett and Keller, 2005; Drew, 2003; Howard, 1999; McCall, 2003; Tulloch, 2008). In all these concerns, the main challenge of GIS-based spatial decision-making applications still lies in providing a reliable tool for enhancing public participation in spatial planning (Boroushaki and Malczewski, 2009). The value of basing the evaluative process of landscape planning on GIS is that it emphasizes spatial thinking, stimulating the expression of residents’ descriptions and equipping them with a more complex spatial vocabulary than that supplied by a simple paper map (Talen, 2000). According to the International Association for Public Participation (2007), there are five levels of public participation: informing, consulting, involving, collaborating, and empowering. Participatory GIS (P-GIS) methods can be developed in order to support all five forms of public participation effectively (Andrzejewska and Baranowski, 2005). Geo-information technologies (GIT) can support participation over a wide variety of strategies. On one hand, traditional GIS mapping tools and participatory sketch-maps can be applied in rapid rural appraisal (RRA), participatory rural

380

appraisal (PRA) (Loader and Amartiya, 1999) and Participatory Learning and Action (PLA) (IFAD, 2009). In these contexts, the map-making process, supported by paper maps created with GIS, aims at involving local communities by means of the commonly understood and easily recognized language of cartography (IFAD, 2009). On the other hand, more advanced geo-visualisation methods can be applied, e.g., 3D virtual rendering and animations, interactive choropleth and diagram mapping, enhanced by aerial photographs and satellite images draped on 3D models. Landscape maps can be delivered through web mapping and web-based GIS applications (WebGIS) which, through the internet, become more accessible to non-experts and have no restrictions in time or location (Jankowski and Nyerges 2001, Dragićević and Balram 2004). In particular, web mapping aims at implementing and delivering digital maps on the web, whereas WebGIS is similar to web mapping but has additional features oriented to display, analyse, process and update geo-data. Many types of web mapping platforms can be implemented: static, dynamic, distributed, animated, real-time, personalized, interactive, analytic, and collaborative (Kraak and Brown, 2001, Peterson, 2003). Collaborative Mapping (CM) is an advanced P-GIS application, which aims at aggregating web maps and user-generated contents from a group of individuals or entities. CM is generally based on a WebGIS combined with other online tools, such as discussion forums, which can provide an alternative to traditional place-based planning (e.g., public meetings/hearings) as they do not require in-person attendance (Boroushaki and Malczewski, 2009). An interesting experiment in this direction is the Argumentation Map prototype, based on a method of formalizing debates which have geo-spatial elements in the discussion (Rinner, 1999, 2001, 2006, Sidlar and Rinner, 2009, Voss et al., 2004). Concerning the implementation strategies of the above P-GIS methods, Leitner et al. (2002) describe and evaluate six types of geo-collab-

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

oration infrastructures: community-based GIS, university-community partnerships, GIS facilities in universities and public libraries, ‘‘map rooms’’, Internet map servers, and neighborhood GIS centers. The major features for successful P-GIS implementation include issue clarity, development of local knowledge, strategic actor relationships, and incremental problem resolution (Ramasubramanian, 1999); other key factors are the structure of the organization, resources available for implementation, and motivations for such implementation (Nedovic-Budic, 1997). Web-based maps are described by many authors (Boroushaki and Malczewski, 2009; Sidlar and Rinner, 2009; Voss et al., 2004) to have major benefits to the public participation process, as they allow direct contact with spatial data, although they also have some specific drawbacks, depending on the staff and hardware available for the internet map server provider, and on users with enough knowledge to interpret the maps and data (Leitner et al., 2002). In this regard, if the application is to be successful overall, much effort must be made by implementing staff to increase access and training (Kingston 2002). Many issues have been raised concerning the real role and efficacy of P-GIS in improving good governance (e.g.: Elwood, 2006b, McCall, 2003, Ventura et al., 2002). A well-structured participatory process should be oriented to geo-database building and attempting to incorporate local knowledge and value-based, traditionally intangible information, such as how residents value the places where they live and their perception about the uniqueness of landscape elements existing in a given area (see Bosworth and Donovan, 1998). Unfortunately, in some cases, the maps used in participation appear as pre-shaped representations created by experts, built on inaccurate surveys and based on a spatial language which may appear extraneous to local cultures (Schuurman, 2006). Too often the traditional top-down flux of geo-information prevails over the alternative bottom-up approach, proposed by many authors,

aimed at direct involvement of local communities in GIS database building (see, e.g., Sieber, 2006, Talen, 2000). This last approach presents many difficulties, because only certain groups or individuals are able to interact actively with P-GIS methods and much effort is required to improve the technical knowledge of the local population. Thus, many actual and potential ‘digital divides’ may exist within participatory GIS, rooted in data and technologies, related to constructions of rationality and expertise, and inscribed in the various kinds of participatory practice (Elwood, 2006a). Because all these concerns exist, a true need for practical ethics and a code of good P-GIS practice emerged at the Mapping for Change Conference (IIRR, 2005). Rambaldi et al. (2006) focused on various key rules which can help to improve ethics in P-GIS processes and the success of participatory processes. These include transparency, openness and honesty in communications, equity and political neutrality in all participatory phases, sincerity toward people’s expectations, respect for the timing of local people, flexibility and adaptability as regards schedules, and reliability in choosing the technologies that can be effectively mastered by local people.

LANDSCAPE PLANNING AND PARTICIPATION OF LOCAL PEOPLE: GOALS, CONCEPTUAL AND METHODOLOGICAL APPROACH A review of the literature shows that P-GIS can effectively support spatial planning procedures, since it allows residents’ landscape perceptions to be explored, providing a new communicative channel and encouraging the participation of residents who may be reluctant to express their opinions by means of more traditional methods. If used with proper regard for issues of transparency of procedures and sensitivity regarding the specificities of local knowledge (McCall, 2003), P-GIS can contribute to the empowerment of com-

381

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

munities and to the construction of productive links between scientific and “common” knowledge. These concepts are used to examine the validity of using P-GIS methods for landscape planning in a vineyard area in central Italy. The case study concerns the construction of a master plan, shared with the inhabitants, for the socio-economic development of a vineyard area through enhancement of its rural landscape. First, landscape units within the region with a high degree of homogeneity were defined, in order to increase the coincidence of interest among the inhabitants. Participatory GIS methods were then developed properly, in order to support all five forms of public participation effectively: informing, consulting, involving, collaborating, and empowering. From the methodological viewpoint, this required the collection of both statistical and strictly cartographic land use information, field surveys and specific analyses, by means of iterative logic procedures covering: i.

ii.

iii. iv.

v.

vi.

382

evaluation of the system of landscape resources, on both large and meso-scales, with interpretation of morphological and structural aspects and spatial organization, land use, roads and infrastructures, and historical, cultural and architectural resources; identification of the potentials, critical factors and vulnerabilities of the various landscapes examined; sharing of geo-referenced information among the research group members; preparation of thematic maps of important landscape values, focusing on communications; presence of stakeholders during meetings at cooperatives, local organizations and associations, and later collection of feedback by means of direct interviews; participation by vineyard owners, who were asked to supply their interpretation of landscape identities and to list landscape operations carried out at farm level;

vii. tourism and other cultural initiatives, already carried out or ongoing, also through associations among enterprises operating in the territory; viii. definition of the most important “quality objectives”, based on integration of expert knowledge and the expectations of the local population, to requalify and enhance the landscapes examined; ix. definition of a general framework of operations according to geo-referenced and mapped plans, assessed by local social units and administrators, in order to evaluate the degree of mutual approval of such operations as proposed by directly interested parties. Bibliographic sources available from the second half of the 15th century onwards (selected for the Umbria region, with particular focus on the “Colli Perugini” area, just south of the city of Perugia), were used to reconstruct an historical profile of land under vines in modern times. This analytical phase was particularly useful in characterizing landscape features, thanks to spatial analysis by means of GIS spatial and multivariate statistical analyses. The general framework of operations was devised as a “master plan” for the area, with an extensive study based on the Choice Models procedure with efficient orthogonal designs. In all these phases, ample recourse was made to GIS technology and web-shared tools to construct databases for landscape analysis, and to create both paper and digital maps, which were efficient in supporting all the phases of data collection from local operators. In many cases, a true process of participatory mapping was developed, thanks to which local operators contributed actively to database construction and updating of cartographic layers. In this way, the members of the research group were not only supplied with valuable information on territorial dynamics, but were also made much more aware of the characteristics and values of their own landscapes. In this context,

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

GIS showed itself to be a true “atlas of the territory”, providing elements of land structuring according to “expert” interpretations and also essential information from locals. The communicative potential of paper maps and GIS was enhanced, defining a new form of relationship with the local population. Projects for operations emerged in the construction of a link between scientific and ordinary knowledge. Within this complex informational process, sustained by GIS and the web, the various types of knowledge could meet, mingle, and be reciprocally measured and compared, in an ongoing search for effective dialogue. It is stressed that the scientific knowledge was derived by a multidisciplinary team consisting of architects, urban planners, agronomists, GIS experts, anthropologists and agricultural economists.

LANDSCAPE PLANNING AND PARTICIPATION OF LOCAL PEOPLE: INFORM AND CONSULT The first step was to identify what types of information are required to construct a geographic information system with the landscape as a common objective, and how the various types of expert knowledge can be implemented and integrated in the GIS to inform and consult local people. The information used for territorial analysis was derived from the vineyard registry for the Umbria Region, updated to 2007, which was subject to preliminary calculations aimed at the reorganization and optimization of Count. All information obtained from various sources was geo-referenced and used to produce detailed maps for the study. Databases and maps were catalogued and placed on a CMS (Content Management System) website built with a double purpose: 1) to share data and maps with all the experts involved in the project; 2) to provide information, as well-organized and complete as possible to local people, both stakeholders and

ordinary residents, to allow them to obtain feedback on analysis, alternatives and/or decisions for the socio-economic development of the area. The vineyard “parcel” was chosen as a reference because it has two important features: it expresses a geographical dimension connected with the vineyard registry, and it is a homogeneous production unit with specific technical (year of planting, type of farming, grape variety) and structural (farm and ownership) attributes. Study of the spatial distribution of vines, through information from units under vines, allowed assessment of vineyard density in the territory, and identification of the most important “wine landscapes” and of those contexts in which vineyards can play a major role in the economic development of the territory. Vineyard intensity (Figure 1) was conceived as an economic phenomenon characterized by specific configuration and spatial concentration, and this was then examined by the spatial analysis techniques offered by modern GIS. The method is based on density analysis of vineyard coverage, expressed as the sum of vineyard surface over the total area of each land registry map sheet. In detail, density was calculated by constructing a special index set at maximum value (100) for areas with the greatest vineyard intensity (land registry map sheets with maximum vineyard coverage - 64%), and gradually decreasing values for coverage percentages calculated linearly from the maximum value. Thanks to this working method, it was possible to build contour lines of vineyard intensity, clearly highlighting vineyard distribution and grouping landscapes according to vineyard intensity class. Comparing geo-processed and field survey data, and with the help of aerial photographs at regional level, three areas-under-vines with different levels of intensity were classified: 1) Vineyard-intensive areas (HIV), index greater than 30; 2) Areas of moderate intensity (MIV), index between 10 and 30; 3) Areas of low intensity (LIV), index below 10.

383

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

Figure 1. Regional contour lines of vineyard intensity (intensity index), showing vineyard distribution in Umbria

Analysis of landscape functions, considered by experts as a key stage in modern planning processes, is based on three phases: i) analysis of the relationship between primary resources, land use and ecology; ii) relationships which are developed between settlement systems and hydrogeo-morphological structures; iii) identification of integrated functional areas (Clementi, 2002). On this assumption, the relationships between the spatial distribution of vineyards and other physical components of the territory (natural, historical, cultural and socio-symbolic) were analysed. A large number of indicators was calculated by GIS, with land map sheets as basic map units, and extensive use was made of geo-referenced map data and at-

384

tributes extracted from the Regional Master Plan and other studies on Umbria (digital terrain model, regional ecological network map, phytoclimaticgeobotanical maps, soil class maps). Some spatial analysis procedures were used in the GIS environment to calculate indicators by means of specific mathematical functions (density or distance modeling) which, besides the mere presence of the various landscape resources, enabled examination of their density or relative distance. In particular, point elements (e.g., archaeological sites, churches, castles, etc.) and some linear features (e.g., roads) were calculated as densityoriented features of individual types of resources.

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

For other elements, the average Euclidean distance between them was calculated and attached as attributed data to map-sheet polygon units. Climate data were subjected to basic interpolation procedures (IDW - Inverse Distance Weighted) which were accurate enough to yield information on regional coverage. Altogether, 63 indicators were identified and divided into 5 classes: i) pedo-climatic; ii) natural and environmental; iii) cultural and historical; iv) land use; v) structural and vineyard. Most of these indicators (54, to be exact, unrelated) were used to identify “homogeneous vineyard units” by cluster analysis. The purpose of cluster analysis was to combine the information aggregated by map-sheet units into classes (clusters) according to criteria of similarity, i.e., to determine a number of classes in order to show, if possible, homogeneity of data within classes and maximum data diversity between classes. The concept of homogeneity is specified in terms of distance or by hierarchical methods alternative to those based on distance. Initially, we chose to use the Ward method as a hierarchical method which can look at each step in an aggregation of classes, so that the variance within the class is minimal. Later, following the results obtained, the K-means algorithm was used, dividing groups of objects into K partitions according to their attributes. This algorithm, following an iterative procedure, minimizes total intra-cluster variance. Initially, in fact, it creates a number of K partitions and assigns entry points to each partition; it then calculates the centroid value of each group and builds a new partition by associating each entry point to the cluster whose centroid is closest to it. The centroids are then recalculated for the new cluster, and so on, until the algorithm converges. Clusters identified by K-Means Cluster Analysis were called “wine production zones” and characterized according to the distribution of mean, maximum and minimum values taken from the 54 indicators within the cluster. In this

way, we identified the variables which most characterized the clusters both negatively (-) and positively (+). Mapping of vineyard areas was made by aggregating the map units (map sheet polygons) falling within the individual land areas. Identified areas were then overlapped with DOC and DOCG zones, and also with local landscapes identified in the regional classification made according to the new regional landscape plan. The whole process resulted in a cartographic representation of regional vineyards and spatial characterization, taking into account soil and climate, natural-environmental, historical and cultural resources (Figure 2). Starting on analysis conducted at regional level, a pilot area was chosen in which to test an effective procedure for local planning, through bottom-up participatory management. The DOC zone called “Colli Perugini” was chosen, both for its large wine production and for the presence of vineyard areas with some potential for socioeconomic development. In the pilot area, it was considered essential to gather information on three points: the wine production system, historical and cultural resources, and the system of integrated projects promoted by local groups involved in the territory. The wine production system was identified and mapped with data both from the vineyard registry, through which it was possible to identify grape producers, and the list of wine-producing estates available on the websites of associations of wine routes and wine guides (Figure 3). A further study was conducted by analysing the files of accession to Measure 1.1.1. (A) “Investments in agricultural holdings” and Measure 1.2.1.g “Improving the processing and marketing of agricultural products” of the Rural Development Plan 2000-2006, which helped to identify the wine enterprises which made infrastructural or commercial investments in the early years of the new century.

385

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

Figure 2. Clustering of land units according to wine production, and environmental and cultural parameters

For a deeper view of cultural and historical resources, consistent with the commercial system of wine production, some preferred lines of study were initially identified. They were found in the three avenues of historical vehicular traffic which have been defined as the “Vineyard Circuits”. The identified research fields of interest were the following: the settlement system, the architectural heritage, religious architecture, pictorial art, the

386

hydrographic network, garden art, architectural ”language” and materials. The system of integrated regional projects promoted by associations operating in the territory was reconstructed by analysing the projects submitted under the Single Programming Document (SPD) 2000-2006 and Leader programs, and by contacting local associations with activities promoted on the internet.

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

Figure 3. Historical routes and landscape in the “Colli Perugini” area: cultural, natural and landscape resources

In order to reach the maximum level of participation estimated possible for the project, the experts who made up the working group assessed suitable techniques for insertion of geographical data. The local people were informed of the data collected, preliminary results (thematic mapping of land-under-vines data at regional level, landscape analysis, socio-economic data clustering,

etc.) which were shown during public meetings organized at municipal offices and at local associations (Local Action Groups, Associazione Terre Burgarelli, Associazione Porta Eburnea, Agricultural Foundation, Mountain Communities wine cooperative, etc.). After meetings, participants were asked if they would like to consult the Web-GIS through which, after registration, they could access and download the material produced

387

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

by the working group (although they were not able to make any changes to it). The meetings were also useful for identifying qualified stakeholders (specific public authorities, wine producers in private and cooperative forms, those promoting public and private tourism, etc.) who were granted extended access to non-published data through the said CMS. This allowed stakeholder profiling and access to different levels of geographical information (from unpublished raster maps to editable geographical databases, according to their level of expertise and reliability).

LANDSCAPE PLANNING AND PARTICIPATION OF LOCAL PEOPLE: INVOLVE AND COLLABORATE The second step in our study was to ascertain how local knowledge could be collected, organized and structured in the GIS, how expert and local knowledge could be integrated, and how GIS functions could be exploited in order to improve landscape understanding and communications. The members of the research group made available their own technical expertise and organizational capacities in order to start up constructive relations with locals, using and implementing the geographic database with local knowledge. They then faced the problem of access to information and their own role in creating and activating a Community Information System (CIS) (Ghose, 2001) through the GIS system. In this regard, the GIS offered not only a set of mapping tools but also various methods in support of the whole participatory process of gathering spatial information and making maps (Vajjhala, 2006). With the aim of stimulating project proposals from the interaction of local with expert knowledge, and in order to turn the Colli Perugini area “into a wine district with its own identity”, attempts were made to involve local people directly.

388

For example, to understand the relationship between socio-economic and historical-cultural components as perceived by “the locals”, a survey was conducted among wine-producing enterprises in the area, by means of: 1) a structured questionnaire; 2) detailed maps of the area for each firm, containing all the information collected during the first phase. Entrepreneurs were interviewed, the aim being to collect information on company history, firm size, reputation on domestic and international markets, sensitivity of the entrepreneur toward rural landscape conservation, participation in events related to wine tourism, firm orientation to innovation and investment, and relations with other wine companies in the territory. After the interview, each producer and sometimes family members involved in management were asked to look at the maps, prepared using all the information collected on the area of interest of the firm. By examining these maps, respondents were able to identify places of historical and cultural interest in their area of production, and any structural problems of greater concern, or damage, which were duly drawn or annotated directly on the paper map of the area surveyed. This resulted in a “living map” of sites as they were perceived by local economic operators, interested in sharing a draft for socio-economic development of the area with public administrators (Figure 4). Besides annotating information on the paper maps, sites were rated according to present criteria. First, each site was mapped as a polygon (e.g., natural park or conservation site), point (e.g., historical buildings) or line (e.g., stretch of ancient or historical road). Interviewees were asked if each site, in their view, had historical, landscape and/or symbolic significance. Information about the level of maintenance, degree of accessibility, usability, etc. of such sites was also sought. The result was an accurate survey of the main sources of environmental and cultural heritage, intended as part of a complex system of coordi-

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

Figure 4. Annotations by wine producers of sites, information, and historical or cultural data on paper maps

nation oriented toward enhancement of the area, both internally and externally. All information gathered on sites was used to implement the base maps, and was made available on a GIS Web site, with free access to the public and selective access to working group members. At this point, based on recommendations made by those with both expert and local knowledge, the construction of a regional project could be started by identifying single actions and interventions. Attention focused on exploiting local resources in order to create integrated development involving both the tourist potential of the territory and the interests of local producers, in synergy among natural and cultural heritage, agriculture, arts and crafts and local traditions. In particular, actions covered the following themes: landscape and the environment; the environment and landscape history; landscape and culture; landscape and history; hospitality; rural roads and leisure, and tourist signs. Several measures were identified for each theme, including:

1) public footpaths in the territory, 2) footpaths through vineyards; 3) the gardens of historical villas; 4) documentation centers in the area; 5) arts and arts and crafts workshops; 6) church architecture; 7) castles, fortresses and historic towns; 8) agri-tourism, food and wine itineraries; 9) alternative viability; 10) tourist signs. The point at which each measure was taken was identified, together with the public and private actors, and the three levels - maintaining the status quo; basic improvements to the existing situation, and significant improvements to the existing situation - were added to the mapping system (Figure 5). A fact-finding survey was carried out regarding the satisfaction produced by the above measures, in order to incorporate preferences expressed by local people in public decisions. Participation was achieved by a direct survey, with questionnaires and maps of planned operations, constructed according to the Choice Experiments procedure.

389

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

The Choice Experiments procedure, based on the preferences (SP) approach, was chosen because decisions are made by subjects with reference to a “hypothetical market” and are not based on data from actual behavior recorded on the real market. Choice Experiment methods include techniques for elicitation of preferences (choices, rankings, ratings), expressed by interviewees who express decisions on a fixed number of alternatives, each of which is described by a set of attributes. The context and characteristics of decision alternatives are the stimuli, and the decision is the response of the interviewee (Adamowicz et al., 1998), which represents the data to be analysed. In our case, attributes were represented by 10 identified project measures, and alternatives were given by a different combination of measures implemented at different levels of action, arising from the experimental design. Attributes

included the item price, which becomes one of the characteristics to be assessed. The method allows estimation of which interventions are most appreciated by respondents and their degree of willingness to pay for them, so this method is clearly very “powerful” in processes of public participation to conceive solutions and incorporate suggestions made by citizens collaborating of their own free will. The survey was conducted in two phases. In the first phase, which used the first experimental design, local people were interviewed directly during the Focus Group meetings held on the premises of public and private associations operating in the territory. The second phase used the second experimental design, which emerged from the results obtained during the first phase. The survey was conducted by means of on-line questionnaires, implemented by PHP (server-side), MySQL (data-

Figure 5. Identified project measures, with alternatives in different combinations arising from the experimental design

390

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

base) and Javascript (client-side effects) software applications. The software infrastructure formed a survey platform for the management of choice experiments which allowed adopted in the next administrative step. The second phase used an efficient design built on Bayesian prior, estimated from information collected in the first phase (Scarpa et al., 2007) and determined by the criterion of D-efficiency (Scarpa and Rose 2008). The first wave was used prior deriving from a pilot survey, based on the construction of an orthogonal fractional design and including only main effects. In the first and second waves, nine card choices were shown, but while the first scenarios included combinations of all ten measures identified, the number of combinations fell to eight in the second step, because two of them turned out not to be significant. The on-line questionnaires were made available through a dedicated web domain as a closed survey, accessible through login and password; results were shared by the CMS project.real-time descriptive statistics and data export of text or database format. Sophisticated data processing and assessment of willingness to pay was carried out by the NLOGIT software application. The overall structure of the survey (including construction of the experimental design and the phase of questionnaire administration) consisted of two separate phases or “waves”. At the end of the first wave, collected data were compiled according to a specific base multinomial logit, in order to estimate the relative priority of the weightings of the attributes necessary to build an efficient design,

SOLUTIONS AND RECOMMENDATIONS Although a simplified application of the P-GIS was applied, the study successfully confirmed the consistency of these methods in supporting

landscape analysis and planning. The participatory approach, supported by GIS and web interfaces, allowed the construction of a virtual link between scientific and local knowledge. The limited knowledge of modern technology of the local population, on one hand, and the limited time and financial resources at the disposal of the research team, on the other, did not allow the development of a more sophisticated web platform oriented to true collaborative mapping. The collection of landscape information based on paper maps, although it was a reliable tool for landscape communication to local communities, turned out to be of little efficiency for GIS database implementation. Thus, the development of a more robust and dynamic web interface, which should ideally combine an effective GIS mapping system (partly based on Google maps, for example) with a forum system, will be of prime importance. The former can map the landscape elements of interest collected by experts or suggested by locals; the latter can support geo-referenced discussions on landscape, for better comprehension of perception of landscape features by locals. This platform, with a user-friendly interface, properly promoted and enhanced, can also support the management of local landscapes through the collection of information and suggestions sent by local people. As the above “digital divide” can undoubtedly constrain this kind of participation process, proper teaching of selected local people must support the development of these systems, in order to facilitate technology access and utilization. The power of maps, SIT and modern communication technologies call for a great sense of responsibility on the part of all those involved in applying P-GIS methods (Rambaldi et al., 2006). It will be essential to apply basic ethic principles in PGIS approaches, focusing mainly on transparency, openness and honesty in communications, and equity and neutrality in all participation phases. P-GIS still represents a reliable strategy for participation, although the above issues - related to the efficiency of technologies used for implementation

391

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

and to the real level of locals involved – may still arise. In Italy and in European counties in general, P-GIS methods are still under-exploited. Thus, national and local government should stimulate recourse to them within participatory spatial planning with properly constructed policies.

tions. Thus, research on landscape should develop more effective methods, partly based on P-GIS, aimed at analysis and evaluation of landscape elements and their contribution to landscape quality.

CONCLUSION FUTURE RESEARCH DIRECTIONS Landscape communication is still in its infancy. Much effort should be made to develop more effective communications strategies aimed at informing and involving local people in PSP. Experts in landscape analysis and planning are often not trained in communications, and recourse to external consultants is not cost-effective, as communications experts are usually not very knowledgeable about landscapes and how to study them. However, the need for more advanced - but user-friendly - web-P-GIS platforms requires a research effort in the field of web technologies oriented to geo-collaboration. The actual interfaces used in collaborative mapping are still experimental, and appear too little intuitive and stimulating for users. So research in this field should be oriented toward further development of such interfaces. In addition, the integration of web interfaces with GIS-database platforms is still poorly known. GIS software houses and open-source communities should seriously consider this need for web-platforms oriented toward participatory mapping in landscape analysis and planning. The traditional web-GIS tools should be developed in order to include more advanced tools, supporting bi-directional communications, in mapping with proper technical strategies for reliable database updating. Landscape quality and its perception by locals takes on a central role in contemporary processes of landscape analysis, planning and design. It is closely bound to landscape components and features and to their spatial and functional interrela-

392

The example of the “Colli Perugini” wine-producing area of Umbria offers an illustration of the many issues with which landscape governance must deal. It confirms that P-GIS represents a good option in spatial planning problems, in order to explore residents’ landscape perceptions, providing a new channel of communications and encouraging residents to participate by expressing their opinions. It can also contribute to the empowerment of communities and to the construction of productive links between scientific and “ordinary” knowledge. The construction of a master plan, shared with inhabitants of a vineyard area, for the socioeconomic development of that area through enhancement of its rural landscape, shows that Participatory GIS methods can effectively support the four forms of public participation: informing, consulting, involving and collaborating. To achieve the fifth form of public participation- “empowering” - it is necessary to activate the “social process” through greater visibility of information and decision-making.

ACKNOWLEDGMENT The analysis presented here draws on data from an interdisciplinary research project funded as part of the ongoing activities of the of Umbrian Regional Directorate of Agricultural and Forestry. The authors’ thanks go to all the researchers, interview partners, and funding organizations of this project. We are particularly grateful to Anna Lambertini, Luciano Giacchè and Andrea

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

Pochini, and to Gabriel Walton for her revision of the English text.

REFERENCES

Corbett, J. M., & Keller, C. P. (2005). An analytical framework to examine empowerment associated with Participatory Geographic Information Systems (PGIS). Cartographica, 40(4), 91–102.

Adamowicz, W., & Louviere, J. S. (1998). Introduction to attribute-based stated choice methods. Final Report to NOAA, Advansis.

Craig, W. J., Harris, T. M., & Weiner, D. (Eds.). (2002). Community participation and Geographic Information Systems. London, UK: Taylor & Francis.

Andrzejewska, M., & Baranowski, M. (2005). Geovisualisation methods for participatory spatial planning. Paper presented at the meeting of the International Cartographic Conference, La Coruña, Spain.

Dragićević, S., & Balram, S. (2004). A Web GIS collaborative framework to structure and manage distributed planning processes. Journal of Geographical Systems, 6, 133–153. doi:10.1007/ s10109-004-0130-7

Antrop, M., & Sevenant, M. (2010). Transdisciplinary landscape planning: does the public have aspirations? Experiences from a case study in Ghent (Flanders, Belgium). Land Use Policy, 27(2), 373–386. doi:10.1016/j.landusepol.2009.05.005

Drew, C. (2003). Transparency - considerations for PPGIS research and development. URISA Journal, 15(1), 73–78.

Boroushaki, S., & Malczewski, J. (2009). ParcitipatoryGIS.com: A WebGIS-based collaborative multicriteria decision analysis. Submitted to URISA Journal. Retrieved January 29, 2010, from http://www.urisa.org/files/ Bosworth, M., & Donovan, J. (1998). A mapmaker’s dream: Public involvement applications utilization of GIS. Position paper presented at Project Varenius Specialist Meeting: Empowerment, Marginalization, and Public Participation GIS, Santa Barbara, CA. Brunckhorst, D., Coop, P., & Reeve, I. (2006). ECI-civic optimisation: A nested framework for planning and managing landscapes. Landscape and Urban Planning, 75, 265–281. doi:10.1016/j. landurbplan.2005.04.001 Clementi, A. (2002). Interpretazioni di paesaggio. Roma, Italia: Meltemi.

Dunn, C. E. (2007). Participatory GIS: A people’s GIS? Progress in Human Geography, 31, 616–637. doi:10.1177/0309132507081493 Elwood, S. (2006a). Critical issues in Participatory GIS: Deconstructions, reconstructions, and new research directions. Transactions in GIS, 10(5), 693–708. doi:10.1111/j.1467-9671.2006.01023.x Elwood, S. (2006b). Negotiating knowledge production: The everyday inclusions, exclusions, and contradictions of participatory GIS research. The Professional Geographer, 58(2), 197–208. doi:10.1111/j.1467-9272.2006.00526.x European Council. (2000). European Landscape Convention. Strasbourg. Ghose, R. (2001). Use of information technology for community empowerment: transforming geographic information system into community information system. Transactions in GIS, 582, 141–163. doi:10.1111/1467-9671.00073

393

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

Howard, D. (1998). Geographic information technologies and community planning: Spatial empowerment and public participation. Retrieved from http://www.ncgia.ucsb.edu/ varenius/ ppgis/ papers/ howard.html IAP2 - International Association for Public Participation. (2007). Spectrum of public participation. Retrieved from http://www.iap2.org/ associations/ 4748/ files/ IAP2% 20Spectrum_vertical.pdf IFAD. (2009). Good practices in participatory mapping. International Fund for Agricultural Development. Retrieved from www.ifad.org/ pub/ map/ PM_web.pdf IIRR. (2005). Mapping for change. International Conference on Participatory Spatial Information Management and Communication, International Institute for Rural Reconstruction, Nairobi, Kenya, 7-10 Sept. Jankowski, P. (2009). Towards participatory geographical information systems for communitybased environmental decision making. Journal of Environmental Management, 90, 1966–1971. doi:10.1016/j.jenvman.2007.08.028 Jankowski, P., & Nyerges, T. (2001). Geographic Information Systems for group decision-making: Towards a participatory, geographic information science. New York, NY: Taylor & Francis. Jones, C. B. (1997). Geographical Information Systems and computer cartography. Essex, UK: Addison Wesley Longman Limited. Kingston, R. (2002). Web-based PPGIS in the United Kingdom. In W. J. Craig, T. M., Harris, & D. Weiner (Eds.), Community participation and Geographic Information Systems (pp. 101–112). London, UK: Taylor and Francis. Kraak, M. J., & Brown, A. (2001). Web cartography - developments and prospects. New York, NY: Taylor & Francis.

394

Kuntu-Mensah, P. (2006). Geographic Information Systems: Indispensable ingredient for good governance in Ghana. Promoting land administration and good governance. Paper presented at the 5th FIG Regional Conference, Accra, Ghana, March 8-11. Lambertini, A. (2009). Valorizzazione dei paesaggi viticoli: produzioni di qualità – paesaggi di qualità. Italy: Unpublished Research Report, Department of Economics and Food Science, University of Perugia. Leitner, H., McMaster, R. B., Elwood, S., McMaster, S., & Sheppard, E. (2002). Models for making GIS available to community organizations: Dimensions of difference and appropriateness. In Craig, W., Harris, T., & Weiner, D. (Eds.), Community participation and geographic information systems (pp. 37–51). London, UK: Taylor & Francis. Loader, R., & Amartiya, L. (1999). Participatory rural appraisal: Extending the research methods base. Agricultural Systems, 62(1-3), 73–85. doi:10.1016/S0308-521X(99)00056-6 Longley, P. A., Goodchild, M. F., Maguire, D. J., & Rhind, D. W. (2005). Geographic Information Systems and science (2nd ed.). Chichester, West Sussex, England: John Wiley & Sons Ltd. McCall, M. K. (2003). Seeking good governance in participatory GIS: A review of processes and governance dimensions in applying GIS to participatory spatial planning. Habitat International, 27, 549–573. doi:10.1016/S0197-3975(03)00005-5 Nedovic-Budic, Z. (1997). GIS technology and organisational context: Interaction and adaptation. In Craglia, M., & Couclelis, H. (Eds.), Geographic information research: Bridging the Atlantic (pp. 165–184). Bristol, PA: Taylor and Francis. Peterson, M. (Ed.). (2003). Maps and the Internet. Elsevier.

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

Ramasubramanian, L. (1999). GIS implementation in developing countries: Learning from organizational theory and reflective practice. Transactions in GIS, 3(4), 359–380. doi:10.1111/1467-9671.00028 Rambaldi, G., Chambers, R., McCall, M., & Fox, J. (2006). Practical ethics for PGIS practitioners, facilitators, technology intermediaries and researchers. [London, UK: IIED.]. Participatory Learning and Action, 54, 106–113. Rinner, C. (1999). Argumentation maps: GISbased discussion support for online planning. GMD Research Series No. 22. Sankt Augustin, Germany: GMD German National Research Center for Information Technology. Rinner, C. (2001). Argumentation maps: GISbased discussion support for on-line planning. Environment and Planning. B, Planning & Design, 28(6), 847–863. doi:10.1068/b2748t Rinner, C. (2006). Argumentation mapping in collaborative spatial decision making. In Balram, S., & Dragicevic, S. (Eds.), Collaborative Geographic Information Systems (pp. 85–102). Hershey, PA: Idea Group. doi:10.4018/9781591408451.ch005 Scarpa, R., Cambell, D., & Hutchinson, W. G. (2007). Benefits estimates for landscape improvements: Sequential Bayesian design and respondent’s rationality in a choice experiment. Land Economics, 83(4), 617–634. Scarpa, R., & Rose, J. (2008). Design efficiency for non-market valuation with choice modeling: How to measure it, what to report and why. Australian Journal of Agricultural Economics, 52, 253–282. doi:10.1111/j.1467-8489.2007.00436.x Schuurman, N. (2006). Formalization matters: Critical GIS and ontology research. Annals of the Association of American Geographers. Association of American Geographers, 96.

Sidlar, C., & Rinner, C. (2009). Utility assessment of a map-based online geo-collaboration tool. Journal of Environmental Management, 90(6), 2020–2026. doi:10.1016/j.jenvman.2007.08.030 Sieber, R. (2006). Public participation Geographic Information Systems: A literature review and framework. Annals of the Association of American Geographers. Association of American Geographers, 96(3), 491–507. doi:10.1111/j.14678306.2006.00702.x Smith, M. J., Goodchild, M. F., & Longley, P. A. (2007). Geospatial analysis: A comprehensive guide to principles, techniques and software tools. Leicester, UK: Troubador Publishing. Talen, E. (2000). Bottom-up GIS. Journal of the American Planning Association. American Planning Association, 66(3), 279–294. doi:10.1080/01944360008976107 Tulloch, D. (2008). Public Participation GIS (PPGIS). In Kemp, K. (Ed.), Encyclopedia of GIS (pp. 351–353). Sage Publications. Vajjhala, S. P. (2006). Ground truthing policy, using participatory map-making to connect citizens and decision makers. Resources, 162, 14–18. Ventura, S. J., Niemann, B. J., Sutphin, T. L., & Chenoweth, R. E. (2002). GIS-enhanced landuse planning. In Craig, W. J., Harris, T. M., & Weiner, D. (Eds.), Community participation and geographic information systems (pp. 113–124). London, UK: Taylor & Francis. Voss, A., Denisovich, I., Gatalsky, P., Gavouchidis, K., Klotz, A., Roeder, S., & Voss, H. (2004). Evolution of a participatory GIS. Computers, Environment and Urban Systems, 28, 635–651. doi:10.1016/j.compenvurbsys.2003.12.003

395

Participatory GIS for Integrating Local and Expert Knowledge in Landscape Planning

ADDITIONAL READING

KEY TERMSAND DEFINITIONS

Goodwin, M. (1998). The governance of rural areas: some emerging research issues and agendas. Journal of Rural Studies, 14(1), 5–12. doi:10.1016/ S0743-0167(97)00043-0

Choice Experiment: Techniques for elicitation of preferences (choices, rankings, ratings), expressed by interviewees who express decisions on a fixed number of alternatives, each of which is described by a set of attributes. Cluster Analysis: Statistical techniques that can be applied to data to find relatively homogeneous groups of similar elements. Geographical Information Systems: GIS, informatics systems designed for managing, analyzing, modeling and representing georeferenced data. Governance: The exercise of formulates and implement policies to manage society’s problems and affairs. Landscape: An area as perceived by people, whose character is the result of the action and interaction of natural and/or human factors (Definition of the European Landscape Convention). Landscape Planning: The process aimed at the formulation and the implementation of multiple actions to enhance, restore or create landscapes. Participatory GIS: The use of mapping and spatial information technologies for involving the public and for integrating local experiences and preferences with experts’ scientific knowledge.

Görg, D. (2007), Landscape governance. The “politics of scale” and the natural” conditions of places, in Geofurum 38, 954-966. Marsden, T., & Murdoch, J. (1998). Editorial: The shifting nature of rural governance and community participation. Journal of Rural Studies, 14(1), 1–3. doi:10.1016/S0743-0167(97)00042-9 Penker, M. (2009). Landscape governance for o by the local population? A property rights analysis in Austria. Land Use Policy, 26, 947–953. doi:10.1016/j.landusepol.2008.11.007 Tress, B., & Tress, G. (2003). Scenario visualisation for participatory landscape planning—a study from Denmark. Landscape and Urban Planning, 64(3), 161–178. doi:10.1016/S01692046(02)00219-0

396