Scholarly article on topic 'Participatory GIS to inform coral reef ecosystem management: Mapping human coastal and ocean uses in Hawaii'

Participatory GIS to inform coral reef ecosystem management: Mapping human coastal and ocean uses in Hawaii Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Applied Geography
Keywords
{PGIS / "Coral reef management" / "Participatory mapping" / Hawaii / "Ecosystem-based management" / "Coastal marine spatial planning"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Arielle Sarah Levine, Christine Loftus Feinholz

Abstract Sociospatial information is critical to marine and coastal ecosystem management. The Hawaii Coastal Uses Mapping Project used a participatory geographic information systems (PGIS) methodology to gather local knowledge regarding the location and intensity of coastal human activities in Hawaii's priority sites for coral reef management. PGIS provided an efficient and effective means of obtaining information in a data-poor context, particularly at a scale and location where considerable local knowledge is held by community members and resource users. We detail the PGIS methods developed to collect sociospatial data on human uses in the project regions and discuss important considerations regarding the practice of PGIS that emerged from the mapping process, as well as implications for the production and documentation of spatial knowledge. Key themes include: issues of scale and appropriateness in using PGIS as a method for mapping human coastal and marine activities; data validity, authority, and the nature of local knowledge; community trust, engagement, and collaboration; and utility for coral reef management. While several factors limit local agencies' ability to use this spatial information to date, natural resource managers found the participatory mapping process to be highly valuable for stakeholder identification and engagement, and the maps provide a resource to state and federal managers to better understand the human implications of future management scenarios.

Academic research paper on topic "Participatory GIS to inform coral reef ecosystem management: Mapping human coastal and ocean uses in Hawaii"

Applied Geography xxx (2014) 1—10

Participatory GIS to inform coral reef ecosystem management: Mapping human coastal and ocean uses in Hawaii

Arielle Sarah Levine a b *, Christine Loftus Feinholz c

a Department of Geography, San Diego State University, 5500 Campanile Drive, San Diego, CA 92182-4493, USA b I.M. Systems Group, Contractor to NOAA Coral Reef Conservation Program, USA c Pacific Cartography, 66-206 Haleiwa Rd., Haleiwa, HI 96712, USA

ARTICLE INFO ABSTRACT

Sociospatial information is critical to marine and coastal ecosystem management. The Hawaii Coastal Uses Mapping Project used a participatory geographic information systems (PGIS) methodology to gather local knowledge regarding the location and intensity of coastal human activities in Hawaii's priority sites for coral reef management. PGIS provided an efficient and effective means of obtaining information in a data-poor context, particularly at a scale and location where considerable local knowledge is held by community members and resource users. We detail the PGIS methods developed to collect sociospatial data on human uses in the project regions and discuss important considerations regarding the practice of PGIS that emerged from the mapping process, as well as implications for the production and documentation of spatial knowledge. Key themes include: issues of scale and appropriateness in using PGIS as a method for mapping human coastal and marine activities; data validity, authority, and the nature of local knowledge; community trust, engagement, and collaboration; and utility for coral reef management. While several factors limit local agencies' ability to use this spatial information to date, natural resource managers found the participatory mapping process to be highly valuable for stakeholder identification and engagement, and the maps provide a resource to state and federal managers to better understand the human implications of future management scenarios.

© 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND

license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Article history: Available online xxx

Keywords: PGIS

Coral reef management Participatory mapping Hawaii

Ecosystem-based management Coastal marine spatial planning

Introduction

Scientists and policy makers are increasingly emphasizing the need for ecosystem-based approaches to managing ocean and coastal regions, recognizing that people and human behavior comprise an important, but often under-examined, component of these complex systems (De Young, Charles, & Hjort, 2008; McLeod, Lubchenco, Palumbi, & Rosenberg, 2005). The management of marine and coastal ecosystems is an inherently place-based process (Koehn, Reineman, & Kittinger, 2013), and place-based approaches are seen as critical to addressing many of the challenges inherent in managing these complex social-ecological systems (Young et al. 2007). Social data is increasingly recognized as an important component of marine ecosystem planning and management (Kittinger et al. 2014), but in practice most efforts to obtain social

* Corresponding author. Department of Geography, San Diego State University, 5500 Campanile Drive, San Diego, CA 92182-4493. Tel.: +1 619 594 5600. E-mail address: alevine@mail.sdsu.edu (A.S. Levine).

data and characterize the human dimensions of marine and coastal ecosystems remain fragmented, sectoral, and limited in scope, with a high reliance on secondary data sources (Le Cornu, Kittinger, Koehn, Finkbeiner, & Crowder, 2014), which often cannot provide the information needed to understand human—environment interactions.

Because coastal and marine ecosystem management is fundamentally based on spatial information, data regarding the locations of resources, people and problems are central to the decisions that are made (Vajjhala, 2006). In addition, incorporating local knowledge and community participation in decision-making processes are critical to obtaining more complete information and developing successful natural resource management and planning programs (Berkes, Colding, & Folke, 2000; Gadgil, Berkes, & Folke, 1993; Pimbert & Pretty, 1997; Salm, Clark, & Siirila, 2000). This is particularly important in marine settings, where scientific or species-specific data is often lacking (Aswani & Lauer, 2006; Johannes, 1998). Participatory mapping, which incorporates local participation to identify and develop spatial information, provides a means of engaging local resource users and stakeholders in data

http://dx.doi.org/10.1016/j.apgeog.2014.12.004

0143-6228/© 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

A.S. Levine, C.L. Feinholz / Applied Geography xxx (2014) 1—10

gathering and natural resource management (Craig, Harris, & Weiner, 2002; Dunn, 2007). Early local engagement through participatory mapping can also improve community trust and local buy-in regarding both the validity of the data created and of management actions, integrating local and 'indigenous' knowledge with scientific or 'expert' data (Dunn, 2007).

Participatory mapping methods vary tremendously, from using sticks, dirt, and paper to incorporating highly technical internet crowd-sourcing and 3-D modeling (Corbett, 2009). The use of geographic information system (GIS) technology in participatory mapping has become increasingly relevant over time as technologies improve and methods for incorporating community input into a GIS framework become more accessible. While many scholars have noted the tendency for GIS to leave out certain groups and favor certain types of knowledge (Aitken & Michel, 1995; Obermeyer, 1995), participatory GIS (PGIS)1 emphasizes community involvement in the production of geographic information used in spatial decision making. PGIS provides a means to integrate local or indigenous knowledge with scientifically-derived data, allowing for the greater participation of local communities and resource users in natural resource planning. The creation of 'official' knowledge maps via PGIS can assist in incorporating information from a locally specific (and evolving) context into broader understandings of ecology and resource management (DeWalt, 1994; Dunn, 2007).

PGIS to inform place-based ecosystem management in Hawaii

In the state of Hawaii, marine managers are increasingly moving towards place-based approaches to addressing threats to coral reef ecosystems (State of Hawaii, 2010). Coral reef ecosystems in Hawaii face numerous threats from a variety of sources, including fishing, coastal development, pollution, climate change, as well as impacts from recreation and tourism (Friedlander et al., 2008). In order to design site-specific management approaches and monitor management outcomes over time, managing agencies require spatial information regarding both the biophysical conditions of these sites, as well as the human uses and activities that take place in these areas. While biophysical monitoring is in place in several locations in Hawaii, spatial information regarding human uses of coastal and marine areas in Hawaii is sparse and, for the most part, has not been systematically collected over time.

To better understand and document the variety of human activities taking place in priority sites for coral reef management in Hawaii, natural resource managers and researchers initiated the Hawaii Coastal Uses Mapping Project.2 The project used a PGIS methodology to gather locally-sourced knowledge regarding the location and intensity of a range of human uses relevant to coral reef management. Through engaging local coastal resource users and stakeholders in mapping human activities in a group setting, PGIS facilitated the digital documentation of local spatial knowledge directly within a geographic information system. The project provides a useful example of how PGIS can be used to obtain sociospatial data to inform a more complete understanding of human—environment interactions.

In this paper, we describe the Hawaii Coastal Uses Mapping project, its outcomes, and the broader implications of using PGIS for site-based coral reef management that can be learned from this case study. We begin by describing the site prioritization process and the decision to use PGIS to inform site-based management. We

1 Also frequently referred to as "public participation GIS" or PPGIS.

2 The project took place in 2010-2011 in West Hawaii and 2011-2012 in West

then describe the methods used in this particular mapping project, including prioritizing activities for mapping, stakeholder recruitment, mapping workshops, and data collection and processing. We provide a brief overview of the project outcomes and results, with select examples of the spatial data collected. We discuss important considerations regarding the practice of PGIS that emerged from the mapping process, as well as implications for the production and documentation of spatial knowledge. Key themes include: issues of scale and appropriateness in using PGIS as a method for mapping human coastal and marine activities; data validity, authority, and the nature of local knowledge; community trust, engagement, and collaboration; and the utility of PGIS as a tool for informing marine and coastal ecosystem management. We conclude with a discussion of project-specific outcomes related to coral reef ecosystem management in Hawaii, as well as the potential for PGIS to facilitate and inform site-based coastal resource management in the future.

Site-based coral reef management in the main Hawaiian Islands

The state of Hawaii's Coral Reef Working Group (CRWG) recognized that in order to implement effective coral reef management, actions need to be taken at a site-specific level to show tangible outcomes (State of Hawaii, 2010). Previous management actions had been spread across the entire state, but due to limited financial resources, state managers determined that efforts would be more successful if targeted at specific locations. Thus, the state chose to prioritize certain coral reef sites for state funding and technical support. The CRWG set an objective to reduce key anthropogenic threats to two priority near-shore coral reef sites using "ahupua'a3-based" management (State of Hawaii, 2010).

As a starting point for determining priority sites, the CRWG used areas of biological importance prioritized in an ecoregional assessment completed by The Nature Conservancy in 2009. These sites were selected using the GIS-based MARXAN computer modeling program, which dynamically analyzes geospatial data so that it can be spatially optimized and represented under different scenarios. Ecological information obtained from scientific databases and expert reviews were used in the model to determine coral reef sites with the greatest potential for long-term resiliency in the Main Hawaiian Islands (The Nature Conservancy of Hawaii,

2009). The CRWG then obtained expert input to rank these sites based on their biological value, degree of threats to coral reefs, and conservation viability, and further refined these rankings based on the sites' readiness for conservation action, urgency of threats, opportunities to collaborate across different conservation priorities, and potential for effective management (State of Hawaii, 2010). In the end, two locations were selected as priority areas for the program: Ka'anapali-Kahekili on in West Maui, and Pelekane Bay-Puako-Anaeho'omalu Bay in West Hawaii Island (State of Hawaii,

2010) (Fig. 1).

Although biophysical data was available through ongoing monitoring programs and previous scientific studies, the state of Hawaii had little information regarding human uses of the marine and coastal environment in the selected priority sites. The Coastal Uses Mapping project was proposed to document human activities, uses, and issues in the two sites and obtain baseline information that would allow management agencies to make informed decisions and monitor changes over time. Participatory mapping was selected as the most efficient and effective methodology to obtain

3 A Hawaiian term referring to traditional land and sea divisions in Hawaii, which generally encompassed the entire watershed and waters offshore, acknowledging the linked nature of human actions upstream and in the ocean.

Fig. 1. Hawaii coral conservation priority sites.

information for these sites given the data-poor context, the considerable local knowledge held by community members and resource users in the two sites, and the value of engaging the public early in the resource management processes. Project organizers chose to use PGIS to facilitate participation by multiple and diverse stakeholders, formalize and simplify data processing, and allow for the production of consistently acquired and publicly vetted spatial data products that could be easily integrated with spatial biophysical data for the priority sites.

Methods

Methods for the project were derived from Wahle and D'lorio (2010) and adapted significantly to account for the specific uses, scale, and socio-cultural context in Hawaii's priority sites. Project organizers4 began the process by defining the project areas and uses to be mapped. Because the project was oriented towards Hawaii state management programs, organizers aligned the project boundaries with the coastal and ocean regions of the priority sites over which the state has jurisdiction. This included a coastal length of 45 miles in West Hawaii and just over 14 miles in West Maui, out to 3 nautical miles offshore for both sites.

Identifying coastal uses for mapping

State managers and local stakeholders provided input regarding key human uses and activities that take place at the two priority sites. For the West Hawaii site, an exhaustive list of human uses and activities was generated, ranked according to importance, and distilled into a final list of 17 key human uses that were considered both relevant to coral reef management and capable of being mapped by local resource users and stakeholders. Each use was precisely defined, with an explanation of what the activity included and excluded.5 This was critical to ensure that no activity would be mapped in duplicate under two different categories (Wahle &

4 The project was an inter-agency collaborative effort involving representatives from multiple NOAA Offices (the Pacific Islands Regional Office, Pacific Services Center, Marine Protected Areas Center, and Pacific Islands Fisheries Science Center), as well as the Hawaii State Division of Aquatic Resources.

5 Complete definitions for each use are available in the final map books created for each project site.

Townsend, 2013); for example, the activity of "swimming" would not be mapped a second time under "diving and snorkeling," as the act of swimming with a mask and snorkel would only be mapped once under "diving and snorkeling." For the West Maui site, initial stakeholder meetings utilized the list generated in West Hawaii as a starting point to obtain feedback regarding relevant uses in the priority site. The types of activities included on West Maui's final list and their definitions were then adapted to cover 17 relevant uses in that priority site (with the goal of being as comparable as possible with the West Hawaii data). Tables 1 and 2 list the uses mapped in each priority site.

Stakeholder recruitment

For each project site, a locally-based workshop coordinator was hired to begin recruiting participants a few months prior to each of the planned mapping workshops. Workshop participants were recruited who had significant knowledge and experience regarding at least one, if not multiple, of the activities being mapped. It was critical to ensure that there were a range of participants who, together, could map the entire project region, as well as all types of activities that might fall under a single use category. A preliminary list of local stakeholders, many of whom were already engaged in collaborative resource management, was obtained through input from local agency representatives, NGOs, and community groups.

Table 1

West Hawaii coastal uses.

Extractive Non-extractive

Aquarium collecting Camping

Gill nets Charter boating and mammal watching

Net fishing from a boat Charter diving and snorkeling

Pole and line fishing from a boat Non-charter diving and snorkeling

Pole and line fishing from shore Non-motorized, non-charter boating

Spearfishing Surfing

Shoreline gathering Swimming

Throw nets Thrill craft and high-speed activities

Culturally significant areasa

a The category "culturally significant activities," was included in the initial list for West Hawaii, but project participants and organizers decided not to create final maps for this category given the challenges involved in mapping it (discussed in Section PGIS as a method for mapping human coastal and marine activities: Scale and appropriateness).

A.S. Levine, C.L. Feinholz / Applied Geography xxx (2014) 1—10

Table 2

West Maui coastal uses.

Extractive

Non-extractive

Aquarium collecting Bait netting

Free diving and SCUBA-based

harvesting Net Fishing from a boat

Net Fishing from shore or nearshore Pole and line fishing from a boat Pole and line fishing from shore Shoreline and nearshore

gleaning/gathering Throw nets

Camping

Commercial diving and snorkeling Commercial, motorized boating and marine mammal watching Non-commercial, non-motorized boating

Recreational diving and snorkeling

Surfing

Swimming

Thrill craft and high-speed activities

This list served as a starting point for recruitment, and additional participants were contacted through referral (snowball sampling), at their place of business (in the case of tour and fishing charter operations), or in some cases in person while they were engaging in coastal activities in the area.

Gaining trust and cooperation from potential participants was another critical component of the recruitment process. Potential participants needed to believe in the usefulness and validity of the mapping project so that they would be willing to take time to participate and share accurate information. Recruitment often took multiple conversations to explain the project rationale, goals, and mapping process. To maximize opportunities for participation, community members were allowed to select one day to participate out of three possible days (two weekdays and one weekend day to accommodate different schedules). Participants were provided with breakfast, lunch, and reimbursed for any travel expenses, but no financial incentive was provided. In the end, the total number of community participants for each project was 48 for West Hawaii and 47 for West Maui.

Fig. 2. Workshop setup.

area" was defined as an area routinely used by most users most of the time, within the seasonal patterns for the activity. Dominant use areas would be considered high-intensity sites for the activity. Each group had to reach internal consensus regarding the boundaries of each dominant use area before their map was finalized, with deference given to participants with greater topical or regional expertise for each location or use.

Workshop participants were also asked to share additional relevant information, or "supplemental use data," about each use being mapped. This supplemental data included both spatial and non-spatial information. Examples of spatial data include distance offshore or a particular depth limitation relevant to the use; non-spatial data included information regarding seasonal patterns, special "pulse" events (such as regattas), species-specific data, and any other aspects of the activity that local residents and resource users deemed important.

Technology and mapping process

Mapping workshops

Spatial data regarding the location and intensity of each activity was obtained through three one-day workshops held at each site. Participants were separated into small mapping groups of 6—12 individuals, plus facilitators. The order in which the uses were mapped varied between groups. Each group started by mapping an activity that was considered uncontroversial and familiar to the majority of participants (eg. "swimming"). The order in which additional uses were mapped was based on the participants' knowledge within each group; topics with which the group participants held unique expertise were mapped early in the day to ensure that they would be covered. Each group mapped as many of the uses as they could, with all group members providing information about an activity if they had knowledge of it.

Each mapping group was assisted by mapping facilitators: one "process facilitator" and one "technical facilitator." The process facilitator was responsible for explaining the mapping process, defining the activities to be mapped, encouraging broad participation, and diffusing conflict. The technical facilitator was responsible for scaling and displaying data layers on demand, capturing data and spatial attributes, and technical troubleshooting. A note-taker worked with each facilitator to capture spatial and supplemental notes for later use.

In each group, a "general use area" and a "dominant use area" were mapped separately for each of the coastal uses depicted in Tables 1 and 2. The general use area was defined as all areas in which the use was known to occur with some regularity over the past 5 years, regardless of frequency and intensity. A "dominant use

Basemaps of both project sites were created to help orient participants to the mapping region and visualize the project area. Place names and feature markers were selected that were specific to different categories of stakeholder knowledge. For example, buoys and bathymetry helped fishermen and boaters orient themselves, whereas snorkelers were familiar with the local names for bays and promontories. The United States Geological Survey (USGS) topographic placenames data layer was used to label all features deemed relevant by local experts. When the official USGS nomenclature didn't match local common usage, the maps were edited accordingly. All site basemaps included data regarding: bathymetry, moorings, resorts, golf courses, bays and promontories, recreation facilities, boating facilities, ocean-sports establishments, parks, roads, topography, water bodies, land cover, placenames, and Quickbird multi-spectral satellite imagery acquired in 2004 with a spatial resolution of 2.5 m.

A remote interactive whiteboard6 was used to project an interactive display of the site basemap from an ArcGIS desktop map document onto a vertical flat white surface during the mapping workshops. There are major advantages to using this technology for participatory mapping. Unlike paper maps, participants can ask the GIS technician to zoom in or out and show more or less detail, making the basemap fully scalable and dynamic. Participants could also request different layers be displayed to better orient themselves to the projected map. This was particularly helpful for participants who were unaccustomed to viewing satellite imagery.

6 Luida eBeam Edge.

Using a pen-like stylus that functioned as a remote mouse, participants were able to digitize ESRI ArcGIS geodatabase features on-the-fly during the mapping workshops. Each feature class was developed to store additional, supplemental attribute data provided by the participants. Feature attributes, like the depth or seasonality of an activity being mapped, were stored as additional information within the geodatabase to supplement the spatial information regarding that use. This provided a major advantage over paper maps; the intermediate step of transcription was not necessary and participant-provided ancillary information was stored directly with each mapped feature.

Two GIS technicians facilitated the technology within each mapping group. Facilitators provided a short demonstration of how to use the stylus to digitize polygons, and the participants quickly became comfortable with the technology, even those with no prior computer experience. Fig. 2 illustrates a typical workshop setup. One technician piloted the ArcGIS map document while the other took notes on participant comments regarding any additional spatial information associated with each mapped use. The piloting technician worked with participants and the process facilitator to display the desired map view, turn base layers on and off, scale the data, and enable participant input via the freehand drawing tool. Supplemental data discussed by the group during the creation of each polygon was recorded by the note takers.

Data processing

The resulting data underwent a four-step post-processing methodology, illustrated in Fig. 3 and developed by NOAA's MPA Center (Wahle & D'lorio, 2010). During the initial data review (steps 1 and 2), a coastline layer was used to clip out the land from polygons where uses were confined to the ocean. Remnant drawing artifacts (slivers and holes) for each mapped polygon were manually repaired. Polygons were then extended or trimmed to align with the spatial notes. The two most common types of spatial adjustments regarded either distance from shore or a particular isobath. These steps resulted in a set of cleaned and adjusted polygons representing each coastal use from each mapping group.

Data from each of the mapping sessions were merged into a single layer for the general use and for the dominant use areas for each category mapped (step 3). General use areas were considered inclusive, and the outer boundary of all of the merged polygons served as the boundary for the final general use area. For the dominant use category, we used thresholds of agreement based on the availability and amount of data. When there was broad common knowledge about a use and the dominant use area was mapped multiple times, a majority threshold was established for the final dominant areas which included only those areas mapped as dominant by a majority of mapping groups. In some cases, where a use or location was so specialized that only one or two people were able to provide spatial information, we mapped the final dominant use area in an inclusive way.

The resulting data underwent a spatial generalization process using vector-based grids (step 4). This provided a consistent means to display overlapping data as well as decreased spatial specificity for potentially sensitive information. We used a 400-m square grid for West Hawaii and a 100-m hexagonal grid for West Maui to consolidate and generalize the data. The merged polygons were reprocessed into the gridded basemap; a grid cell was considered inclusive if a polygon fell into the majority area of a cell. This process resulted in a gridded representation of each coastal use for each of the project sites. Each final gridded dataset utilized a binary attribute scheme to determine the presence or absence of each coastal use within each gridded cell.

Fig. 3. Illustration of data processing methodology.

A.S. Levine, C.L. Feinholz / Applied Geography xxx (2014) 1 —10

Digital and paper versions of the final maps were presented back to stakeholders for a final review and vetting of the data. Additional revisions based on participant feedback were incorporated into the final data layers.

Results

The West Maui and West Hawaii coastal uses mapping workshops resulted in stakeholder-vetted maps of human uses of the coastal and ocean areas of each of the priority sites. Final map books were published which contain individual maps depicting the spatial footprint of each use across the project region, additional maps zoomed-in for finer scale resolution, and a brief narrative description of supplemental use data. The final processed spatial data and metadata for each site is hosted online,7 as are interactive web map applications from which users can generate maps and statistical information regarding coastal uses in these areas, as well as generate heat maps indicating the total number of overlapping uses in a region. An example map derived from the coastal use data is presented in Fig. 4, depicting commercial, motorized boating and marine mammal watching areas for the West Maui priority site.

Discussion

The past two decades have seen a significant increase in efforts to obtain sociospatial data that is relevant to ecosystem-based planning and management (McLain et al., 2013), and participatory mapping has played a key role in obtaining this type of information. As technologies for recording spatial information become more accessible and widely available, opportunities for community participation in mapping will continue to increase (McCall & Dunn, 2012). PGIS, which involves local communities in collaboratively creating and using geographic information, has become an increasingly common means to obtain spatial data regarding the social seascape of the marine environment. Still, there are formidable social and technical challenges involved in successfully implementing PGIS (Barndt, 1998; Weiner, Harris, & Craig, 2002). The case example presented here provides insight regarding critical themes in PGIS practice, as well as theoretical considerations regarding the use of PGIS for the production and/or documentation of community-based knowledge.

PGIS as a method for mapping human coastal and marine activities: Scale and appropriateness

PGIS has been heralded with both enthusiasm and caution (Dunn, 2007), and the stated purpose and methods for undertaking PGIS vary tremendously. PGIS applications range from urban community planning (Vajjhala, 2006), decision support for water protection and allocation (Jankowski, 2009), designing marine protected areas (Aswani & Lauer, 2006; Scholz, Steinback, Kruse, Mertens, & Silverman, 2011), marine spatial planning (St Martin & Hall-Arber, 2008), participatory land use planning (Wang, Yu, Cinderby, & Forrester, 2008), documenting and protecting local knowledge and indigenous tenure rights (Bond, 2002; Neitschmann, 1995), conflict resolution (Kwaku Kyem, 2004), and have a wide range of goals and desired outcomes. The methodologies used to achieve these objectives, both for fostering

7 The final products for West Hawaii are available at: http://marineprotectedareas. noaa.gov/dataanalysis/hi_coastal_use/, and for West Maui at: http://www. hawaiicoralreefstrategy.com/index.php/westmaui#West_Maui_Coastal_Use_ Mapping_Project.

community participation in mapping and for making effective use of spatial technologies, vary as much as the objectives themselves.

This study utilized PGIS for the purpose of obtaining baseline information regarding the spatial extent and intensity of human activities in priority sites for coral reef management. We found that our selected methodology was conducive to better understanding human uses of the coastal and marine environment. This was particularly the case in locations such as West Hawaii and West Maui, where there is very little existing spatial data regarding human activities. It would be both cost-prohibitive and time-consuming to undertake regular, regional long-term human use monitoring of coastal activities, particularly in locations such as Hawaii Island where many coastal areas are remote and challenging to access. In this context, local resource users are arguably the best source of information about activities that they engage in and observe regularly. The participatory mapping process allows resource users to convey not only the locations of activities, but also contextual details regarding the limits and drivers of activities (such as depth limitations, wind and weather constraints, or social pressures), details that traditional monitoring data generally lacks.

However, PGIS methods must vary based on local context, and not every context is suitable for the use of PGIS to obtain spatial data. Considerations of scale are critical in deciding who can participate, how people participate, what information can be obtained, what specific PGIS methodology should be used, and whether PGIS methods are even appropriate. The scale of each project region in our study was such that most participants were familiar with the majority of each site being mapped, although the degree of participants' expertise varied by location and use-type. The scale also was small enough that multiple uses could be mapped in detail over the course of one day. These considerations led to our choice of a multi-day workshop format with different mapping groups. This allowed multiple groups to map each coastal use, providing redundancy in the data that served as a validity check. Each mapping group was composed of a variety of types of resource users, which also allowed different users to validate (or contest) the information provided by others in their group.

While our methodology was appropriate for mapping the spatial footprint of the majority of selected ocean and coastal activities, it was not appropriate for all types of uses. "Culturally significant areas" was a use-category deemed important by local stakeholders and managers, and the category was included for mapping during the West Hawaii workshop. However, the category proved challenging to map, and participants approached this challenge in diverse ways. While some participants in the West Hawaii workshop chose to map the entire coastal and marine environment as significant, others preferred to restrict points to locations that had already been mapped (heiau, fish ponds, burial sites, etc.) for confidentiality reasons. Still others stated that they could only speak for their own family's area, but that each family had a story and significant sites that could not be represented via the workshop process. Others opted not to create any polygons at all rather than risk misrepresenting the significance of Hawaiian cultural uses.

Mapping culturally relevant information is important for resource management decisions, but it challenges the epistemo-logical limits of conventional GIS (Bond, 2002; Laituri, 2002). PGIS has been demonstrated to be a useful tool for representing indigenous knowledge and claiming areas important for traditional use (Neitschmann, 1995), and maps can assist in eliciting stories relevant to cultural significance (Grenier, 1998), but it is important to note that indigenous knowledge is unique, and cultural values placed on land and place are often manifested and communicated in fuzzy, emotional, and holistic terms which may not fit neatly into a GIS format (Dunn, 2007; McCall & Minang, 2005). After obtaining feedback from workshop participants, particularly cultural

Fig. 4. Commercial, motorized boating and marine mammal watching in West Maui, Hawaii.

practitioners, we decided not to include the "culturally significant areas" category in the final map products, nor was it included in the list of uses to be mapped in West Maui. While participants felt this was an important category, the multi-stakeholder workshop process was not appropriate for accurately obtaining and representing information regarding many of the important non-spatial elements critical to Hawaiian culture.

Data validity, authority, and the nature of local knowledge

In a data-poor context, PGIS provides a means to obtain the best information available from local knowledge-holders and represent it in a format that is consistent with the presentation of traditional scientific data. Inscribing, digitizing, and combining community-driven data with remotely sensed imagery in a GIS format gives this information greater "authority" (Dunn, 2007; MacNab, 2002), formalizing anecdotal data collection in a way that enables natural resource decision-makers to effectively incorporate this information into existing geospatial data management structures where it can be combined and compared with other data. Community-members, in turn, are more likely to accept this data as valid because local resource users provided the spatial data used to create the maps and understand how they were created.

The converse of this, however, is that resource managers may question the validity of anecdotal information obtained from local community members. While representing exact spatial boundaries and information with this type of geographic knowledge is, for the most part, impossible (Couclelis, 2003), PGIS can reveal results that are surprisingly consistent with other more traditional mapping methods for natural resource planning and conservation. Brown (2012) found that information on native species distribution obtained through PGIS had relatively low spatial error when compared with vegetation maps. Cox, Morse, Anderson, and

Marzen (2014) also found high levels of public accuracy in identifying suitable habitat for threatened species conservation.

To avoid creating a false sense of precision with imprecise boundaries, as well as mitigate potential concerns about mapping sensitive data, we used an equal-area grid cell format to represent the final data. The West Hawaii site was mapped using 400 m gridded squares, and the smaller West Maui site used 100 m gridded hexagons. We found that using hexagonal tessellation as a gridding scheme was more appropriate, particularly when following a non-linear geographic feature such as the coastline (Carr, Olsen, & White, 1992). This scale was appropriate to minimize errors due to participants' uncertainty about exact boundaries and was considered to be adequate spatial resolution for most anticipated management planning needs.

While the methodology provided a useful approach to obtaining spatial information regarding human coastal and ocean uses, there are a number of challenges inherent in representing this type of data. Translating spatial data from participants' mental maps into GIS layers poses the potential problem of reifying boundaries that are actually dynamic, shifting, and fuzzy (McCall & Dunn, 2012). Mental maps contain additional spatial and non-spatial information regarding social relations, histories, events, and patterns (McLain et al., 2013), as well as "day to day contexts of lived experiences" (Aitken, 2002). It is challenging to translate this complex spatial and non-spatial contextual information into standard maps or geodatabases, and indistinct boundaries derived from PGIS cannot be viewed in the same way as precise boundaries depicted in technical surveys (Craig et al., 2002). Creating GIS layers through a PGIS process runs the risk of over-simplifying the information or portraying it in a shallow way. We attempted to resolve many of these issues through the use of gridded cells, linked attribute tables, and supplemental qualitative text, but accurately mapping the range and complexity

of dynamic, contextual human uses in a visually accurate way remains a significant challenge.

In addition to the challenge of representing imprecise boundaries, another difficulty lies in ensuring that the information obtained is accurate. Participants can lie. Even when participants are providing their best knowledge, it is impossible to have absolutely certain or precise boundaries with geographic data regarding coastal human activities (Couclelis, 2003); the coast itself is, by nature, a shifting and imprecise boundary (Bartlett, 2000). While the redundancy in mapping efforts and necessity of obtaining group consensus for the final shapefiles helped minimize the risk that any single participant misrepresent information and undermine the mapping process, there was still room for intentional or unintentional error. This was particularly the case with potentially sensitive information like fishing grounds. Aquarium fishermen, in particular, were reluctant to participate in the mapping activities given recent restrictions on their fishing activities and the lack of support for aquarium fishing amongst the Hawaiian population (Rossiter & Levine, 2014). Additionally, participants' knowledge of the region was not always complete, particularly for the West Hawaii site which contained long, remote stretches of coastline that lack access.

Community trust, engagement, and collaboration

Establishing and maintaining community trust is essential for successful PGIS implementation, and meaningful community participation is a critical, yet challenging, component of participatory mapping (Barndt, 1998; Craig et al., 2002). The careful selection of participants can help ensure that key voices are included and that information is representative of a range of user-groups and demographic characteristics (McLain et al., 2013). Recruiting participants for the Coastal Uses Mapping Project was an intensive process, involving a good deal of time seeking out appropriate and knowledgeable resource users, explaining the purpose of the project, gaining participant trust, answering questions, and addressing concerns. Many participants were still skeptical or concerned about how the data would be used going into the mapping workshops, and the project lead spent considerable time addressing these concerns at the beginning of each workshop day. Each group was facilitated by carefully selected, trained individuals, most of whom were experienced with the unique cultural context in Hawaii, to help ensure a workshop environment that was conducive to smooth and meaningful participant engagement. The time devoted to participant recruitment, engagement, and trustbuilding helped increase their comfort-level and cooperation; participants willingly shared spatial information with the facilitators, and the majority of participants expressed satisfaction with the workshop process. In the workshop evaluations, the vast majority of participants (94% for West Hawaii and 82% for West Maui) stated that they found the workshops either "very useful" or "useful," and the majority of participant feedback was extremely positive.

Establishing a relationship of trust between community members and resource managers during the participatory mapping process also facilitated the process of engaging communities in natural resource management, where establishing trust and respect is critical (Roth, 2004; Thornton & Scheer, 2012). Participatory research provides a means to help communities and government agencies to learn about each other, establish a foundation for cooperation, and initiate collaborative management planning (Fox, 1990). The PGIS workshops assisted in this process by bringing representatives from government agencies together with a wide variety of local community members and resource users to map human uses of the coast, serving as a means to "bring everyone to

the table" when many new management processes were starting in the two priority sites. Local managers and NGO representatives frequently stated that the mapping workshops facilitated the process of stakeholder identification for later conservation planning. For remote managers who were based off-island in Honolulu, facilitating and note-taking during the PGIS process provided an opportunity for them to learn directly from resource users and local residents, providing them with a better understanding of the scope of uses and types of investments that stakeholders had in the resources.

The workshop process provided an opportunity for participants to engage not only with managers, but also with other community members. The involvement of different types of resource users, many of whom do not regularly interact, provided an opportunity for these individuals to meet and engage with each other in new ways, learn from each other, and revealed different perspectives and shared concerns held by different stakeholder groups. This facilitated community engagement in natural resource management in the priority sites, enabling collaborative discussions regarding human coastal uses.

As has been observed in other PGIS activities for marine management (MacNab, 2002), the mapping process was in some ways more valuable than the maps themselves. Managers consistently stated that because the PGIS process was "resource focused, but neutral," it provided a good entry point into involving communities in natural resource management planning. While the use of GIS for mapping is never a truly neutral enterprise (Aitken & Michel, 1995; McCall, 2003; Pickles, 1995), the workshop was focused on identifying the locations of activities to inform management rather than to regulate activities. Frequently, government-sponsored stakeholder engagement processes and public meetings have been flawed by soliciting community input at the final stage of planning and decision-making processes rather than during the early stages (Jankowski, 2009). Engaging local stakeholders before specific management mandates are defined provides an opportunity for local voices and priorities to be expressed and provides a forum to explore and understand diverse perspectives regarding what problems exist (Ramsey, 2009). Early community involvement also reduces the perception that affected groups must defend their activities or territories from specific management actions, allowing greater room for public input in a non-combative, collaborative atmosphere.

Utility for coral reef management

The Human Coastal Uses Mapping Project was seen by local managers as an important first step to inform management actions in Hawaii's coral reef priority sites. While state managers have been very positive about the PGIS outcomes of community engagement in management planning, most have said that the maps themselves have had only limited use for management planning to date. This could be due to a number of reasons.

Although managers initially anticipated that spatial zoning might be necessary as a management action in the two priority sites, this has not been the case. Ocean zoning remains controversial in the state of Hawaii, and managers do not consider it to be a current priority in the two sites, particularly given the lack of observed user-conflict confirmed by the mapping workshops. Collecting baseline data before management plans were in place helped to inform management directions and assisted with building community trust and engagement in management processes. However, once management plans are established, the generalized spatial information collected during mapping may not be what is necessary to inform more specific management plans and needs. There is a potential conflict between the ostensibly "neutral"

A.S. Levine, C.L. Feinholz / Applied Geography xxx (2014) 1—10

mapping that happens with no specific management or regulatory agenda versus mapping for specific management purposes, which inevitably places priorities or values on certain uses, creating winners and losers and potentially alienating certain user groups. While mapping with specific management activities in mind may provide more targeted information, it might not achieve the same levels of community participation or the benefits of community engagement observed in this project.

While the maps created during the PGIS process are available to inform management activities in the state of Hawaii, the state's natural resource agencies lack the funding and capacity to undertake an ambitious management agenda. Hawaii's Division of Aquatic Resources has been short-staffed for some time, and funds have not been available to share the mapping results beyond an initial presentation back to the communities. In addition, the political environment around spatial planning or establishing additional marine protected areas has been too sensitive for state agencies to take steps in that direction. Management priorities in the West Maui site have been oriented towards mitigating land-based sources of pollution and sediment, so little attention has been directed towards coastal and ocean management policies in that region to date.

While the maps have helped to inform general management priorities and directions for the two sites, it may be too early to determine the full utility of the PGIS mapping products for coral reef management in Hawaii. The state is still determining management actions for the two sites, and the utility of the maps may depend on management directions. Combining human use data layers with maps of biological priorities could provide a more complete picture of the social-ecological implications of selecting certain sites as marine protected areas, inform managers of the human implications of management actions, and allow for more informed decision-making and targeted outreach strategies.

Conclusion

PGIS has broad potential for involving local stakeholders in natural resource management. We found that our PGIS methodology assisted government representatives in identifying and engaging with community members and local resource users in the early stages of coral reef management planning for small-scale priority sites in the Hawaiian Islands. PGIS also allowed for the collection of information regarding human activities in a context where data on human uses were largely unavailable and challenging to collect. While not appropriate for all types of human activities, PGIS workshops provided a useful methodology for documenting the geographic extent of a number of key human uses in relatively small-scale priority coral reef management areas. This facilitated the spatial representation of data, using GIS to better visualize and communicate information regarding human activities in the coastal and marine environment, both during participatory data collection as well as through the final map products.

PGIS proved useful as a method to obtain sociospatial data to provide a more complete understanding of human—environment interactions in coastal and marine ecosystems. While this information has significant potential to inform place based marine ecosystem management in Hawaii, several factors limit its use to date. A reluctance to engage in formal ocean zoning or marine spatial planning, as well as limited funding and capacity to extend PGIS methodologies or incorporate the spatial information obtained into coastal and ocean management planning, limit the management applications of the spatial data in Hawaii. The methods used in this study also restricted our ability to obtain certain types of social data, including in-depth information on non-spatial aspects of human activities, as well as nuanced cultural and

historical data. While this type of information is compatible with PGIS, in-depth interviews and ethnographic approaches conducted at a smaller spatial scale than our mapping workshops would be better suited to obtaining this type of information to gain a more complete picture of the socio-cultural seascape of human interactions with the marine environment. In spite of these limitations, natural resource managers found the participatory mapping process to be highly valuable for obtaining sociospatial information relating to place based management of priority coral reefs sites in Hawaii, and the maps provide a resource to state and federal managers to better understand the human implications of future management scenarios.

Acknowledgments

Many individuals, too numerous to cite individually, assisted with the Hawaii Coastal Uses Mapping Project. Megan Lamson and Krista Jaspers played an essential role as local project coordinators for each site, recruiting participants and organizing workshop logistics. Staff from NOAA's MPA Center, including Mimi Diorio, Nick Hayden, and Jordan Gass, served as technical leads for the West Hawaii site and generously assisted with data collection and processing for West Maui. Jamie Carter, Kalisi Mausio, and Gabby Fausel from NOAA's Pacific Services Center helped lead the technical component for the West Maui site. The project would not have been possible without the assistance of our many skilled facilitators for both sites, including Petra MacGowan, Emma Anders, Luna Kekoa, Mike Lameier, Laurie Richmond, Kosta Stamoulis, Matt Ramsey, and Danielle Bamford. We would like to thank Petra MacGowan, Emma Anders, Chad Wiggens, and Tova Calendar for their insight regarding management outcomes and implications of the project. Financial support for the project was through NOAA's Coral Reef Conservation Program and the Hawaii Coral Reef Initiative (through the State of Hawaii's Division of Aquatic Resources). The Ka'upulehu Interpretive Center and Kaunoa Senior Center provided the use of their facilities for the workshops in West Hawaii and West Maui, respectively. Most importantly, we extend our sincere thanks to all of the workshop participants in West Hawaii and West Maui who shared their time and knowledge during the workshop process — mahalo nui loa!

References

Aitken, S. (2002). Public participation, technological discourses and the scale of GIS. In W. Craig, T. Harris, & D. Weiner (Eds.), Community participation and geographical information systems (pp. 357—366). New York, London: Taylor and Francis.

Aitken, S., & Michel, S. M. (1995). Who contrives the "real" in GIS? geographic information, planning and critical theory. Cartography and Geographic Information Systems, 22(1), 17—29. Aswani, S., & Lauer, M. (2006). Incorporating fishermen's local knowledge and behavior into Geographical Information Systems (GIS) for designing marine protected areas in oceania. Human Organization, 65(1), 81—102. Barndt, M. (1998). Public participation GIS—Barriers to implementation. Cartography and Geographic Information Systems, 25(2), 105—112. Bartlett, D. J. (2000). Working on the frontiers of science: applying GIS to the coastal

zone. Marine and Coastal Geographical Information Systems, 11—24. Berkes, F., Colding, J., & Folke, C. (2000). Rediscovery of traditional ecological knowledge as adaptive management. Ecological Applications, 10(5), 1251—1262. Bond, C. (2002). The Cherokee nation and tribal uses of GIS. Community Participation

and Geographic Information System, 283—294. Brown, G. (2012). An empirical evaluation of the spatial accuracy of public participation GIS (PPGIS) data. Applied Geography, 34, 289—294. Carr, D. B., Olsen, A. R., & White, D. (1992). Hexagon mosaic maps for display of univariate and bivariate geographical data. Cartography and Geographic Information Systems, 19(4), 228—236. Corbett, J. (2009). Good practices in participatory mapping. International Fund for

Agricultural Development (IFAD). Couclelis, H. (2003). The certainty of uncertainty: GIS and the limits of geographic knowledge. Transactions in GIS, 7(2), 165—175.

Cox, C., Morse, W., Anderson, C., & Marzen, L. (2014). Applying public participation geographic information systems to wildlife management. Human Dimensions of Wildlife, 19(2), 200—214.

Craig, W. J., Harris, T. M., & Weiner, D. (2002). Community participation and geographical information systems. CRC Press.

De Young, C., Charles, A., & Hjort, A. (2008). Human dimensions of the ecosystem approach to fisheries: An overview of context, concepts, tools and methods. FAO.

DeWalt, B. (1994). Using indigenous knowledge to improve agriculture and natural resource management. Human Organization, 53(2), 123—131.

Dunn, C. E. (2007). Participatory GIS—a people's GIS? Progress in Human Geography, 31(5), 616—637.

Fox, J. (1990). Keepers of the forest: land management alternatives in Southeast Asia. In M. Poffenberger (Ed.), Keepers of the forest: Land management alternatives in Southeast Asia (pp. 119—133). Manila: Ateneo de manila University Press.

Friedlander, A., Aeby, G., Brainard, R., Brown, E., Chaston, K., Clark, A., et al. (2008). The state of coral reef ecosystems of the main hawaiian islands. In J. E. Waddell, & A. M. Clarke (Eds.), NOAA technical memorandum NOS NCCOS 73The state of coral reef ecosystems of the United States and pacific freely associated states: 2008 (p. 569). Silver Spring, MD: NOAA/NCCOS Center for Coastal Monitoring and Assessment's Biogeography Team.

Gadgil, M., Berkes, F., & Folke, C. (1993). Indigenous knowledge for biodiversity conservation. AMBIO, 22(2/3), 151—156.

Grenier, L. ( 1998). Working with indigenous knowledge: A guide for researchers. IDRC.

Jankowski, P. (2009). Towards participatory geographic information systems for community-based environmental decision making. Journal of Environmental Management, 90(6), 1966—1971.

Johannes, R E. (1998). The case for data-less marine resource management: examples from tropical nearshore finfisheries. Trends in Ecology & Evolution, 13(6), 243—246.

Kittinger, J. N., Koehn, J. Z., Le Cornu, E., Ban, N. C., Gopnik, M., Armsby, M., et al. (2014). A practical approach for putting people in ecosystem-based ocean planning. Frontiers in Ecology and the Environment, 12(8), 448—456.

Koehn, J. Z., Reineman, D. R., & Kittinger, J. N. (2013). Progress and promise in spatial human dimensions research for ecosystem-based ocean planning. Marine Policy, 42, 31—38.

Kwaku Kyem, P. A. (2004). Of intractable conflicts and participatory GIS applications: the search for consensus amidst competing claims and institutional demands. Annals of the Association of American Geographers, 94(1), 37—57.

Laituri, M. (2002). Ensuring access to GIS for marginal societies. In W. Craig, T. Harris, & D. Weiner (Eds.), Community participation and geographic information systems (pp. 270—282). London and New York: Taylor and Francis.

Le Cornu, E., Kittinger, J. N., Koehn, J. Z., Finkbeiner, E. M., & Crowder, L. B. (2014). Current practice and future prospects for social data in coastal and ocean planning. Conservation Biology, 28(4), 902—911.

MacNab, P. (2002). There must be a catch: participatory GIS in a Newfoundland fishing community. In W. Craig, T. Harris, & D. Weiner (Eds.), Community participation and geographical information systems (pp. 173—191). London and New York: Taylor and Francis.

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(4), 549—573.

McCall, M. K., & Dunn, C. E. (2012). Geo-information tools for participatory spatial planning: fulfilling the criteria for 'good'governance? Geoforum, 43(1), 81—94.

McCall, M. K., & Minang, P. A. (2005). Assessing participatory GIS for community-based natural resource management: claiming community forests in Cameroon. The Geographical Journal, 171(4), 340—356.

McLain, R., Poe, M., Biedenweg, K., Cerveny, L., Besser, D., & Blahna, D. (2013). Making sense of human ecology mapping: an overview of approaches to

integrating socio-spatial data into environmental planning. Human Ecology, 41(5), 651-665.

McLeod, K., Lubchenco, J., Palumbi, S., & Rosenberg, A. (2005). Scientific consensus statement on marine ecosystem-based management. Signed by 221.

Neitschmann, B. (1995). Defending the Miskito Reefs with maps and GPS. Cultural Survival Quarterly, 18(4), 34-37.

Obermeyer, N. J. (1995). The Hidden GIS technocracy. Cartography and Geographic Information Systems, 22(1), 78-83.

Pickles, J. (1995). Ground truth: The social implications of geographic information systems. New York, London: Guilford Press.

Pimbert, M. P., & Pretty, J. N. (1997). Parks, people and professionals: putting "participation" into protected area management. In K. B. Ghimire, & M. P. Pimbert (Eds.), Social change and conservation: Environmental politics and impacts of national parks and protected areas (pp. 297-330). London: Earthscan Pulications Ltd.

Ramsey, K. (2009). GIS, modeling, and politics: on the tensions of collaborative decision support. Journal of Environmental Management, 90(6), 1972-1980.

Rossiter, J. S., & Levine, A. (2014). What makes a "successful" marine protected area? The unique context of Hawaii' s fish replenishment areas. Marine Policy, 44, 196-203.

Roth, R. (2004). Spatial organization of environmental knowledge: conservation conflicts in the inhabited forest of northern Thailand. Ecology and Society, 9(3), 5.

Salm, R. V., Clark, J., & Siirila, E. (2000). Marine and coasa protected areas: A guide for planners and managers. Washington DC: IUCN.

Scholz, A. J., Steinback, C., Kruse, S. A., Mertens, M., & Silverman, H. (2011). Incorporation of spatial and economic analyses of human-use data in the design of marine protected areas. Conservation Biology, 25(3), 485-492.

State of Hawaii. (2010). Hawaii coral reef strategy: Priorities for management in the main hawaiian islands 2010-2020. Honolulu, HI.

St Martin, K., & Hall-Arber, M. (2008). The missing layer: geo-technologies, communities, and implications for marine spatial planning. Marine Policy, 32(5), 779-786.

The Nature Conservancy of Hawaii. (2009). Appendix C: overview of the marine ecoregional assessment for the main hawaiian islands. In State of Hawaii (Ed.), Hawaii coral reef Strategy: Priorities for management in the main Hawaiian islands. Honolulu, HI.

Thornton, T. F., & Scheer, A. M. (2012). Collaborative engagement of local and traditional knowledge and science in marine environments: a review. Ecology & Society, 17(3).

Vajjhala, S. P. (2006). "Ground truthing" policy: using participatory map-making to connect citizens and decision makers. Resources for the Future Summer, 162, 14-18.

Wahle, C., & D'Iorio, M. (2010). Mapping human uses of the ocean: Informing marine spatial planning through participatory GIS. Silver Spring, MD.

Wahle, C., & Townsend, J. (2013). A common language of ocean uses. Silver Spring, MD.

Wang, X., Yu, Z., Cinderby, S., & Forrester, J. (2008). Enhancing participation: experiences of participatory geographic information systems in Shanxi province, China. Applied Geography, 28(2), 96-109.

Weiner, D., Harris, T. M., & Craig, W. (2002). Community participation and geographic information systems. In W. Craig, T. Harris, & D. Weiner (Eds.), Community particiaption and geographic information systems (pp. 3-16). London and New York: Taylor and Francis.

Young, O. R., Osherenko, G., Ekstrom, J., Crowder, L. B., Ogden, J., Wilson, J. A., et al. (2007). Solving the crisis in ocean governance: place-based management of marine ecosystems. Environment: Science and Policy for Sustainable Development, 49(4), 20-32.