IK Monitor Articles (9-1)

Indigenous Knowledge and Development Monitor, March 2001

Contents IK Monitor (9-1) | IKDM Homepage | Suggestions to: ikdm@nuffic.nl | © copyright Nuffic-CIRAN and contributors 2001.

Renee E. Bartolo and Greg J.E. Hill

A case study from northern Australia
Remote sensing and GIS technologies as a decision-making tool for indigenous land management.

This article discusses the issues at stake when remote sensing and geographic information system (GIS) technologies are combined with indigenous peoples' intimate knowledge of the land, to develop a tool for informed land management decisions.

Approximately 85% of the Top End coastline of northern Australia is owned by Aboriginal people. A substantial part of this coastline lies within Arnhem Land, which is administered by an Aboriginal Land Trust. Wetland systems extend across this region, and their natural resources support significant local subsistence economies. The Arafura Swamp is such a wetland system.

The people of the Arafura Swamp have inhabited the region for tens of thousands of years and they have an intimate knowledge of the land. The knowledge of the elders is crucial in drawing up land management plans. It gives younger members of the community, who may be trained in Western land management principles, an insight into the state of the country before some of the recent threats began to make themselves felt. For example, some elders have noted an increase in the number of hoofed animals in the last 50 years, accompanied by the conversion of open swamp to wooded swamp dominated by paperbarks. This has important implications for natural resources obtained from open-swamp environments.

The natural resources of the Arafura Swamp are currently under threat from the spread of the weed Mimosa pigra and an increase in the number of feral pigs (Sus scrofa) within the swamp. These threats are relatively new. There are few up-to-date maps of the area on which to base informed land management decisions. The advent of geographic information system technology has facilitated analyses of changes in wetland systems (Young & Dahl 1994).

The major advantage of implementing a geographic information system (GIS) in the community is that the knowledge of the elders can be recorded before it is lost, and later analyzed for the purpose of future sustainable development.

However, the role of GIS based on Australian Aboriginal knowledge has thus far been focused on supporting Native Title negotiations (for example, Turk & Mackaness 1995; Turk 1996; Guthrie & James 1996), rather than promoting natural resource management within the indigenous lands. Crerar (1998) presents one resource-based study in which the integration of remote sensing and GIS technologies was utilized to locate crocodile nesting sites, for the commercial harvesting of the eggs for the Bawinanga Aboriginal Corporation. In other countries the use of GIS technology for indigenous natural resource management is more advanced (for example, Marchand & Winchell 1994; Marozas 1993; Tabor & Hutchinson 1994; Harmsworth 1997).

Traditional Aboriginal landowners have indicated an interest in current large-scale maps and satellite imagery of the Arafura Swamp region. This article focuses on the establishment of a baseline environmental GIS for use by the Ramingining community. It focuses on the practical applications of both the framework for indigenous knowledge and GIS outlined by Harmsworth (1998) and the approach used by Lawas (1997) in integrating remote sensing and GIS with the knowledge of farmers (in this case, rangers, who represent the various clans that own land in the study area).

Study area
The Arafura Swamp is located approximately 550 kilometres east of Darwin, the capital of the Northern Territory. The swamp is situated in the floodplain of the Gulbuwangay and Goyder Rivers, and covers an area of over 71,000 hectares (see map). The climate of the region is characteristic of the wet-dry tropics; the mean annual rainfall for the region is 1156 mm.

Most of the indigenous people of the Arafura Swamp and the surrounding area are members of Yolgnu clans. The Yolgnu are the indigenous people of northeast Arnhem Land, and the various clans have similar cultural practices and language. From a Western perspective, there are five recognized community groups inhabiting the Arafura region: Ramingining, the Northern Swamp, Murwangi, the Southern Swamp, and Goyder catchment (Wearne & White 1997).

The Arafura Swamp is vital to the livelihoods of the people. Wetland fauna such as magpie geese (Anseranas semipalmata), file snakes (Acrochordus arafurae), northern long-necked turtles (Chelodina rugusa), and barramundi (Lates calcarifer) are collected as a basic food source. The wetland flora is not only important as a food source, it is also used for medicinal purposes.

The people of the Arafura Swamp and the surrounding area have a great spiritual affinity with the land, which is divided into different 'countries', on the basis of clan. Further distinctions are often based on tribal divisions. Another concept central to the Yolgnu understanding of the landscape is 'dreaming'. Dreaming is inherited from the father, and the resulting patrilinear line entitles the recipient to ownership of that country.

Case study context
The Northern Land Council (NLC) is an indigenous statutory authority that promotes the land rights and economic self-determination of various Aboriginal clans in the Top End of Australia. In order to facilitate land management decisions, an indigenous ranger programme has been implemented with the aid of the Caring for Country Unit of the NLC. This initiative, which is known as the Wanga Djakamirr Rangers project, aims to provide the Yolgnu with the skills which will enable them to manage their countries efficiently. The research project was carried out within the framework provided by the ranger project.

Methodology
The methodology adopted is outlined in Figure 1. The issue of scale is crucial in determining the optimum resolution of data to be used for mapping purposes. The most suitable scale was determined after collecting the available source data on the study area. The GIS was established using digital topographic data (GEODATA) from the Australian Surveying and Land Information Group (AUSLIG). Satellite imagery was enhanced and interpreted using standard procedures.

In conducting fieldwork, appropriate research protocols for working in Aboriginal communities were followed. The protocols adopted in this study were developed by the Northern Territory University, Center for Indigenous Natural and Cultural Resource Management (CINCRM 1998).

Results and discussion
Baseline environmental data that the local community can use in land management decisions have been collected. This information is now being compiled and stored within a Geographic Information System (GIS). Using the enhanced Landsat TM image, a vegetation map containing 20 vegetation communities and a geomorphological map with 10 categories were produced. The AUSLIG GEODATA provided a basic framework for the GIS, but the data were outdated. Similar problems related to positional accuracy were encountered by Sirait et al. (1994) and Bestow (1993). The GIS is currently in the process of being updated.

Working together
Here we would like to make some observations on the research collaboration involved. During the fieldwork there was extensive collaboration with the Wanga Djakamirr Rangers, while discussions with community members confirmed the validity of the scale analysis and data selection. Joint fauna and flora surveys were conducted in parallel, in connection with the first biological inventory of the region. The people of Ramingining were interested in Western reporting methods and their relevance for the self-management of their own lands. The scientists, in turn, were eager to learn from the knowledge systems and practices of the local people (see photo).

During the actual mapping, collaboration between scientists and rangers was aimed at compiling a map that was culturally suitable for land management by the community. The rangers were asked for their input in order to determine the appropriate symbology for the map. They expressed a strong desire for the map to represent the landscape as closely as possible. In order to achieve this, photos of land cover types were attached to the map. The vegetation map was reconfigured with the guidance of the indigenous rangers, to make it suitable for interpretation by the people of the region. For example, the Melaleuca species were allocated tones of grey to match the approximate colour of the bark. The floodplain grasses were colour-coded in tones of brown, to resemble their appearance during the dry season.

Aboriginal names for cover types were recorded, to add to the legend. Class categories were also renamed in one of the common spoken dialects. The naming of class categories was a complicated process. The rangers stated that they do not distinguish between certain species unless they have some cultural importance. Table 1 presents a number of generic Yolgnu terms for broad groupings of vegetation. It should be noted that the collection of Aboriginal names for purposes of a map is a long process, due to the hierarchical nature of knowledge within Aboriginal society.

At this stage, it is not possible to examine the geomorphological map in the same manner. From the understanding gathered from the rangers, the Aboriginals do not name geomorphic features in a Western sense. While names were recorded for coastal geomorphic features such as freshwater springs and sandbars, it could not be determined whether these were dreaming names. Therefore, for considerations of cultural sensitivity, they were not included on the map.

Issues at stake
The purpose of the Arafura Swamp GIS is to aid the local community in making informed land management decisions with regard to environmental problems that are new to both the traditional owners and the region itself. The natural resources of the region support the local subsistence economy, and natural resource management and the economic development of the swamp are of vital importance (Storrs 1997). The implementation of the GIS for land management has been well received by the Ramingining community. They view the combined use of Western science, traditional knowledge and GIS as a contribution to better land care. They would like to have control over the GIS, thereby increasing their self-sufficiency when it comes to land management decisions. This would also ensure privacy, in connection with sensitive cultural information (Turk 1996). Other relevant issues related to the implementation of a GIS at Ramingining include training the Wanga Djakamirr rangers in the use of GIS software and the storage of traditional knowledge for future generations. This would involve the customization of the GIS user interface, in order to address cultural issues such as those examined by Campari and Frank (1993) and Gould (1991). Further research is needed, in order to incorporate indigenous land classification systems into the GIS.

Renee E. Bartolo and Greg J.E. Hill
Faculty of Science, Information Technology and Education
Northern Territory University
Darwin, Australia 0909
Fax: +61-89-46 7088
E-mail: r_bartolo@site.ntu.edu.au

Acknowledgements
The authors would like to thank the following people for their assistance: the Wanga Djakamirr Rangers, the traditional owners of the land in the Arafura Swamp, and Wayne Barbour; the Cooperative Research Centre for the Sustainable Development of Tropical Savannas; the Northern Land Council; the Parks and Wildlife Commission of the Northern Territory; and Chris Devonport.

References
Bestow, I. (1993) 'Developments in the GIS of the Central Land Council'. Paper presented at the North Australian Remote Sensing and Geographic Information Systems Forum 1993. Darwin: Australian Government Publishing Service.

Campari, I. and A.U. Frank (1993) 'Cultural differences in GIS: A basic approach'. Paper presented at the Fourth European Conference and Exhibition on Geographical Information Systems, EGIS '93. Genoa.

CINCRM (Center for Indigenous Natural and Cultural Resource Management) (1998) 'Guiding principles for ethics policies and research protocols'. Darwin: Northern Territory University.

Crerar, J.M. (1998) 'The use of remote sensing and GIS by indigenous people for natural resource management'. Paper presented at the 9th Australasian Remote Sensing and Photogrammetry Conference. Sydney: University of New South Wales.

Gould, M.D. (1991) 'Elicitation of spatial language to support cross-cultural Geographic Information Systems', pp. 435-448 in D.M. Mark and A.U. Frank (eds) Cognitive and linguistic aspects of geographic space. The Netherlands: Kluwer Academic Publishers.

Guthrie, K. and S. James (1996) 'The native title information system', The Australasian Geographic Information Systems Applications Journal 18: 30-33.

Harmsworth, G. (1997) 'Maori values and GIS. The New Zealand experience'. GIS Asia Pacific (April): 40-43.

Harmsworth, G. (1998) 'Indigenous values and GIS: a method and a framework', Indigenous Knowledge and Development Monitor 6(3)

Lawas, M.C.M. (1997) The resource user's knowledge, the neglected input in land resource management. The case of the Kankanaey farmers in Benguet, Philippines, ITC Publication 52.

Marchand, M.E and R. Winchell (1994) 'Tribal implementation of GIS', Cultural Survival Quarterly (Winter): 49-51.

Marozas, B.A. (1993) 'Strategies for the implementation of geographic information systems in Indian Country'. Paper presented at the GIS '93 Symposium. Vancouver.

Siriat, M., S. Prasodjo, N. Podger, A. Flavelle and J. Fox. (1994) 'Mapping customary land in East Kalimantan, Indonesia: a tool for forest management', Ambio 23(7): 411-417.

Storrs, M. (1997) 'The Top End Indigenous peoples' wetland programme'. Paper presented at the Workshop Proceedings entitled 'Research and Management in the Arafura Swamp Region, Central Arnhem Land'. Arafura Swamp: CINCRM.

Tabor, J.A and C.F. Hutchinson (1994) 'Using indigenous knowledge, remote sensing and GIS for sustainable development', Indigenous Knowledge and Development Monitor 2(1).

Turk, A.G. and W.A. Mackaness (1995) 'Design considerations for spatial information systems and maps to support native title negotiation and arbitration', Cartography 24(2): 17-28.

Turk, A. (1996) 'Presenting Aboriginal knowledge', Alternative Law Journal 21(1): 6-9.

Wearne, G. and N. White (1997) 'Supporting natural and cultural resource management in the Arafura Wetlands and Catchment: A community-based approach'. Paper presented at the Workshop Proceedings entitled 'Research and Management in the Arafura Swamp Region, Central Arnhem Land'. Arafura Swamp: CINCRM.

Young, R.W. and T.E. Dahl (1994) 'Use of GIS in assessing areas of rapid wetland change', GIS/LIS: 851-859.

Websites that may interest readers of this article.