Arid Lands Newsletter (link)No. 53, May/June 2003
Using geospatial technologies to develop
participatory tools for natural resources management

Participatory GIS-based natural resource management: Experiences from a country of the South

by Rami Zurayk

"One of the most important issues we still face is how much the local communities will be able to use the products of PGIS to improve their livelihoods after the project ends. This is a continuous source of frustration and relates to the whole development research approach rather than to PGIS alone."


Introduction

(Back to top)
In spite of its limited area and small population (10,452 km2 and 4.1 million people), Lebanon is ecologically as well as socially diverse. Located in a typical Mediterranean setting, its elevated mountain ranges (over 3,000 m amsl) contribute to a large variation in rainfall, ranging from 1,500 mm per year in the mountain area to less than 200 mm in the central plain. Lebanon's arable land is limited to 1/3 of its area, with only 68,000 ha in permanently irrigated lands and nearly 100,000 ha as fragile terraced mountain lands. Services and trading form the basis of a liberal, laissez-faire economy. Agriculture, once a prominent activity, now contributes less that 10% to the gross national product (GNP). Moreover, the country was left crippled by 17 years of war, which ended in 1991, and currently bears the brunt of the volatile Middle Eastern politics. The public sector has been incapacitated and is still unable to fully take responsibility for the sustainability of agriculture.

Over the past 50 years, the Lebanese agricultural sector has witnessed a major shift from a low-input, extensive farming system aimed at staples and some fruit production to an intensive, land-limited and horticulture-based system. It is widely acknowledged that current farming practices are unsustainable and that land degradation is advancing.

It is against this background that the Environment and Sustainable Development Unit (ESDU) of the Faculty of Agricultural and Food Sciences at the American University of Beirut has been operating its development research programs. Located in the oldest Western-styled university in the Middle East (founded in 1866), ESDU researchers have been active in rural development for over 10 years. They have developed a number of research approaches that blend technological tools such as GIS with community-based approaches such as participatory research and participatory development communication. In this article, I will briefly present a GIS project which relied heavily on community participation (hence the abbreviation "participatory GIS" or PGIS) and discuss primarily the social aspects of the lessons learned, focusing on the possibilities and difficulties of promoting real local participation in ownership of GIS.

Why did we need GIS?

(Back to top)
We set out to develop options for sustainable natural resource management. GIS was ideal for the job, as it is designed to support the capture, management, manipulation, analysis, modeling and display of spatially referenced data for solving complex planning and management problems. However, many of the team members expressed concerns that GIS would force the research away from the end users and back into the laboratory. Those among us involved in GIS-based research argued that the use of GIS was akin to using any modern tool, and that it usually is the researcher, rather than the tool, who seeks the peace and quiet of the offices. Thus, the decision to use GIS was adopted, with the proviso that it should involve the local community and that the benefits should be directly felt by the stakeholders. This constitutes the basic definition of PGIS.

What did we use PGIS for?

(Back to top)

Over the period 1995-2003, we have used PGIS to:

  • Develop geo-referenced databases for the study areas.
  • Delineate indigenous agro-ecological zones.
  • Identify optimal conditions for the occurrence of target plant species.
  • Identify and delineate land cover changes.
  • Plan orchard development.
  • Understand the dynamics of pastoralism and the threats it may pose.
  • Delineate areas of inappropriate land use (land stress analysis).
  • Understand land tenure in relation to natural resource management.
  • Plan locations for rainwater harvesting reservoirs.

Many of these projects have been presented to the international scientific community as journal articles (Zurayk, El-Awar, Hodeib et al. 2001; Zurayk, El-Awar, Hamadeh et al. 2001; Hamadeh et al. 1999). Thus, only one example -- an indigenous agroecological zoning project -- has been selected for presentation here in order to demonstrate the capabilities of the PGIS approach.

Indigenous agroecological zoning project

(Back to top)
The aim of this project was to delineate the agroecological zones in three different study areas within three different regions of Lebanon (Aarsal, Ham and Maaraboun), according to the indigenous classification. The ultimate goal was to use the indigenous agroecological zones as the basis for natural resource management planning within the three study areas.

Approach

The investigation relied principally on a combination of participatory field methods and GIS. GIS was used primarily as a tool of analysis and presentation. Participatory field methods were selected to gather data on the kind of seamless experience and knowledge local participants have of their environment and to integrate this experience with that of the researchers, in order to broaden the common knowledge-base (Chambers 1994a, 1994b).

The team

Meetings were arranged to bring together the project team and the local focal points in each of the study areas. The group usually consisted of 2-3 researchers and an equal number of local people with good knowledge and experience of the area. The participatory investigations took the form of relaxed semi-structured interviews following a guiding framework, which outlined the sequence of events and points to remember. The researchers behaved mostly as active listeners and observers.

Link to Figure 1, sketched resource map
Link to Fig. 1, sketched resource map (~48K)

Link to Figure 2, satellite image with zones marked
Link to Fig. 2, satellite image with boundaries (~25K)

Link to figure 3, agroecological map
Link to Fig. 3, agroecological map (~110K)

The tools

The classic participatory investigations tools (maps, sketches, diagrams, ranking) were used first. The participants in each study area were provided with sheets of white paper and asked to draw resource maps of the study areas according to their perceptions. They were also provided with a satellite image of their area and asked to mark the boundaries of its different agroecological zones based on their recognition of the terrain. The information obtained included local soil types and their classification and distribution within the landscape, local agroclimates, local land use systems and indigenous agroecological zones. This information was then ground-truthed and transformed into geographic data using a Global Positioning System (GPS), entered into a GIS format (ArcInfo), and overlaid with the base maps of the three study areas in order to produce a georeferenced agroecological map.

Field validation

During the ground-truthing phase of the project, one or more of the local participants accompanied the field team to locate the boundaries of the different soil and agroecological zones in the study area. This particular step was necessary but very lengthy, as each area required an average of five visits with a team of three persons (local guide, GPS operator, and driver). Hundreds of data points were acquired in each study area in order to enable the accurate drawing of the zone boundaries.

Findings

Two kinds of products resulted from this project: hand-drawn maps (local team) and GIS maps. The PGIS approach led to a set of maps that faithfully represented the study areas and that were readily recognized and adopted by the local community. One of the main limitations of the mapping approach was uncovered when we tried to define soil units. Our approach relied on representing them as clearly delineated polygons. This fell short of fully reflecting the local knowledge, where soils form a continuum of different types intergrading into one another rather than being well-defined, clearly bounded patches on the landscape. Farmers in the project areas consider the soil units as being composed of "mostly this, with some of that." This "fuzzy" approach does not appear to be specific to Lebanon; it was also reported in Highland New Guinea (Sillitoe 1998). Its usefulness is clear as it allows farmers the flexibility to adapt to changes in the landscape, which is necessary in the dissected highland areas where the relief can create multitudes of soil types.

We also found that while it was possible for the local community to geographically locate a zone with great accuracy on a white sheet of paper as well as on a satellite image, it was not possible to rely on their perception of dimensions when drawing free-hand on a white sheet. The size (surface area) they allocated to a certain location depended directly on the importance of this location to their livelihood. For example, in one study area a map was hand-drawn by each of the adjacent communities of Ham and Maaraboun. The Ham group drew the local "mazraa" (elevated plain) as oversized and the local "sakiy" (irrigated plain) as small, while the Maaraboun group inverted the sizes of these two features (large sakiy and small mazraa). Maaraboun relies mainly on the sakiy for agricultural production while the mazraa is the main farming area of Ham.

Lessons learned

(Back to top)
PGIS proved to be a useful, practical, and cost-effective tool that delivered the product it was initially meant to deliver: accurate georeferenced information that is recognized and readily adopted by the local communities and other stakeholders. It also greatly facilitated the research process by enhancing trust-building, as the communities appreciated the fact that the researchers acknowledged the value of their own local knowledge. This increased their commitment to the project and to its outcomes. It also gave the researchers valuable insights into the way the local people perceive their environment, which was helpful in the implementation of other components of the project.

However, the matter of true ownership of the product of PGIS is still a major preoccupation of the ESDU. In our examination of PGIS over the course of the numerous projects, a number of issues were highlighted. Some of these apply to all participatory research. For instance, in the early projects we implemented, as well as in some of the recent ones, the process of participation was criticized as we found that in many instances, the local people were "being participated" by the researchers rather than wholeheartedly participating in the process. This was primarily due to our researchers' biases, which often imposed a priority that was not always important for the local people. This was the case, for instance, with projects related to the conservation of agrobiodiversity, still an elusive concept for farmers in the South (Zurayk and Haidar 2002). In other instances, however, successful participation was achieved, as the local communities had themselves expressed their interest in the outcomes of the analyses. This was the case, for instance, when researchers and farmers together planned future orchard development or determined the location of rainwater harvesting reservoirs in Aarsal (Zurayk, El-Awar, Hodeib et al. 2001; Zurayk, El-Awar, Hamadeh et al. 2001).

Our self-examination also led us to identify issues specific to PGIS. One of these derives from its highly technical aspects. Unlike other participatory field tools, this one cannot be, at present, fully transferred to local communities, due to its technological and human resources requirements. Very few people in the rural areas of the South have enough know-how in information technology to be able to be trained to take over the whole process. Thus, for the foreseeable future, PGIS will require an umbilical cord linking researchers with the local communities.

At the beginning of the study period, the local people were unfamiliar with maps and with geographic representations such as contour lines and land cover polygons. As a result, throughout the process, the researchers were required to act in varying degrees as interpreters of the maps.

Comprehension of visual presentation material by non-specialists has been identified as a major issue in the use of GIS. Draping 3-D models with particular digital coverages (soils, land cover) has been suggested as one means to achieve such comprehension (Theocharopoulos et al. 1995). Our experience in this project was that a 3-D model remained poorly comprehensible by the local people. We found, however, that the landowners could visualize their lands as cadastral parcels with clearly defined boundaries, as such parcels related to their perceptions of land ownership and tenure.

One of the most important issues we still face is how much the local communities will be able to use the products of PGIS to improve their livelihoods after the project ends. This is a continuous source of frustration and relates to the whole development research approach rather than to PGIS alone.

Over the past decade, we have, with the help of local communities, accumulated a large volume of information on the sustainable management of natural resources. Most of this information was never translated into knowledge on the ground, as very few rural development interventions have resulted from our research results. The piecemeal nature of development projects in the South and its highly politicized donor-government-communities relationship are certainly to blame. We may also share part of the responsibility for engaging in participatory development research without any way of insuring that this research will lead to any tangible improvement in the livelihoods of the concerned communities. Our presence always has the effect of raising expectations and in most -- if not all -- cases these expectations remain unmet. This commonly leads local communities to reject both the products of participatory research and any further research endeavors. Little can we blame them, as by the project's end both researchers and funding agencies will have fulfilled their agendas through publications and recognition, while all the local community has to show for their participation is a set of maps with fading colors.

The way forward: Embedded research

(Back to top)
We recognize that appropriate research is a major driving force for development. However, our experience with PGIS shows that, in its present form, local participation in research may lead to valuable information but too often results in few changes in the livelihoods of the target communities. This was the case with the project described here. This has the effect of leaving dedicated researchers (like us) with the uneasy feeling of abusing the trust of the community by being the sole beneficiaries of the research process (through the papers we publish).

Most institutions such as ours focus mainly on carrying out pertinent and relevant applied participatory research and then publishing the outcomes. Implementation of the research results is usually not accounted for in the grants that fund our work, which leads to frustration and fatigue of the local communities. Local people remain poor and feel used. This explains why, after working for decades on combating poverty, and after thousands of pertinent documents, we have not moved very far. There is a pressing need for a new paradigm in which development interventions are planned as an integral part of a process led by development research. Not until we "embed" research into development are we going to be able to make a difference, whatever technological tool we adopt. This is the approach we have recently come to rely upon in ESDU.

We were fortunate, in ESDU, to obtain large grants for development research from the International Development Research Centre in Canada to study the sustainability of farming systems of drylands in Lebanon, and, simultaneously, a service contract for development interventions in the drylands of Lebanon: the provision of technology transfer and extension, on behalf of the Lebanese Ministry of Agriculture. The grants were significant and allowed a lot of flexibility.

Halfway through the implementation of the research grant, we faced the problems of frustration and near-rejection by the local community, and of the uneasy feeling on our part that our research could be inconsequential. We then linked the services contract with the research program, and things became smoother. Our research improved as it was built on real needs, and we were able to tailor it and adjust it through a feedback loop. Through the services contract, we were able to implement, at farmer field scale, the results of our research. We initiated an iterative process which is still going on and which truly comes close to full participation by the local community.

Then came the Iraq war, and the notion of "embedding" took a whole new different meaning in the Middle East--provoking, for us, yet further questions about where development research stands with regard to the realities of the world...

References

(Back to top)

Chambers, R. 1994a. Participatory rural appraisal (PRA): Analysis of experience. World Development 22(9): 1253-1268.

Chambers, R. 1994b. Participatory rural appraisal (PRA): Challenges, potentials, and paradigm. World Development 22(10): 1437-1454.

Hamadeh, S., R. Zurayk, F. Awar, S. Talhouk, D. Abi Ghanem and M. Abi Said. 1999. Farming systems analysis of dryland agriculture in Lebanon: an analysis of sustainability. Journal of Sustainable Agriculture 15(2/3): 33-44.

Sillitoe, P. 1998. Knowing the land: Soil and land resource evaluation and indigenous knowledge. Soil Use and Management 14(4): 188-193.

Theocharopoulos, S. P., D. A. Davidson, J. N. McArthur, and F. Tsouloucha. 1995. GIS as an aid to soil surveys and land evaluation in Greece. Journal of Soil and Water Conservation 50: 118-124.

Zurayk, R., F. El-Awar, H. Hodeib, C. Sayegh, and S. Hamadeh. 2001a. A GIS-based methodology for soil degradation evaluation. In Sustaining the global farm: Selected papers from the 10th International Soil Conservation Organization (ISCO) Conference, May 23-28, 1999, Purdue University, West Lafayette, Indiana, USA, eds. D. Stott, R. Mohtar and G. Steinhardt, 1082-1089. International Soil Conservation Organization. Online: http://tucson.ars.ag.gov/isco/isco10/SustainingTheGlobalFarm/P027-Zurayk.pdf (Accessed 7 July 2003).

Zurayk, R., F. El-Awar, S. Hamadeh, S. Talhouk, C. Sayegh, A. Chehab and K. al Shab. 2001. Using indigenous knowledge in land use investigations: a participatory study in a semi-arid mountainous region of Lebanon. Agriculture, Ecosystems and Environment 86(3): 247-262.

Zurayk, R. and M. Haidar. 2002. Whose basic needs? Farmers and funders priorities in biodiversity conservation. Regional Workshop on Best Practices in Biodiversity Conservation, organized by the Third World
Academy of Sciences and the Global Environment Facility, Muscat, Oman,
April 2002.

 

bar denoting end of article text

Author information

(Back to top)
Dr. Rami Zurayk is Professor of Land and Water Resources in the Environment and Sustainable Development Unit, American University of Beirut, Lebanon. You can reach him for comment at:
rzurayk@aub.edu.lb

Additional web resources

(Back to top)
Environmental and Sustainable Unit (ESDU), American University of Beirut
http://staff.aub.edu.lb/~webeco/ESDU/index.html
Established in 2001 within the Faculty of Agricultural and Food Sciences, ESDU focuses on "action-centered, policy-relevant and participatory research."

E-cosystem web site projects page
http://users.aub.edu.lb/%7ewebeco/projects.htm
This web site, created by the Faculty of Agricultural and Food Sciences at AUB, currently provides information on various long-term ecological change research projects being carried out by AUB faculty.

About the Arid Lands Newsletter

Link to ALN home page Link to index page for back web issues Link to index page for pre-web issue archive Link to ALN #53 table of contents