Humanitarian organizations must understand both the geographical and social environments in which they operate in order to maximize the impact of their interventions. This paper describes how seasonal population movements of the GaagwangNuer, semi-nomadic pastoralists of western Ethiopia, were mapped. An understanding of these seasonal movements was requisite to the inception of a medical humanitarian program in western Ethiopia in 2010. A participatory approach using qualitative methods was used to investigate three interrelated aspects of Gaagwang Nuer society: population mobility, kinship structure, and agricultural and pastoral activities. In this study, we combined geography with anthropological frameworks to develop an operational base for the provision of health care in a remote setting inhabited by nomadic pastoralists. Our research demonstrates that nomadic migrations are generally predictable, which is conducive to health care planning and service delivery. This methodology may have applications in planning services for other nomadic pastoralist societies.

About the Author:Colin Watson is Lecturer in Health at the School of Health, Charles Darwin University, Australia


I. Introduction

Of all the information required in planning humanitarian interventions, the “where” dimension is perhaps the most important [1]. Humanitarian organizations need to understand not only the geography of their environment, but also the infrastructure, settlements, land usage, political boundaries, and a range of the social and population data that overlay it. However, this information is frequently not readily available.

This challenge was evident during the establishment of a primary-level medical intervention in the Whantoa district in the Ethiopian regional state of Gambella. Located in Western Ethiopia, the Whantoa district is very remote, with little infrastructure or public services. The district is bordered by the Baro, Pibor, and Makwei Rivers and transected by the Dhure River. At the time of the humanitarian intervention, landform maps were inaccurate and land use maps were nonexistent. The Whantoa district is home to the Nuer, one of the largest ethnic groups in East Africa, who inhabit large tracts of land in South Sudan and Western Ethiopia. The Nuer practice a mixed agriculture and pastoralist economy centered on the production of sorghum, maize, and cattle. Cattle herding activities require semiannual migrations between relatively permanent, wet season settlements and temporary dry season cattle and fishing camps, a practice referred to by the term transhumance. Transhumance describes spatially limited patterns of movement such as those made by the Nuer, in which pastoral movements and agricultural activities are confined to discrete geographical areas [2]. For most Nuer residing in the district, dry season grazing lands are in close proximity to wet season settlements. Despite these modest seasonal migrations, the movement of people across the district had profound implications for the planning and provision of health care services.

This paper describes a participatory methodology used to map seasonal migrations of the Gaagwang, a clan of the Nuer who live in the Whantoa district. Local knowledge was utilized to understand the interplay of geography, climate and human activity. This understanding proved essential for planning health service delivery, particularly outreach programs such as mobile clinics and vaccination campaigns.

II. Background

The Whantoa district is located between the Ethiopian highlands and the plains of South Sudan. Comprised predominantly of lowland savannah with some marsh and forest, there are few roads and many communities are only accessible by foot or river. The population of the Whantoa district is estimated at 23,000 inhabitants [3]. The presence of several vulnerability factors—including food insecurity, population displacement, and limited health care—prompted the intervention by Médecins Sans Frontièresin the small town of Mathar, Whantoa district’s administrative center. The intervention was comprised of a fixed-site health service providing both outpatient and inpatient care. Mobile clinics also provided outreach primary health care.

Nomadic pastoralism is a commonly practiced lifestyle in Sub-Saharan Africa. Pastoralism is a production system and economy based on the herding of livestock [4]. In Ethiopia, pastoralists are estimated to comprise 11% of the population [5].

Nomadic populations often have poor health status. Pastoralists are at risk of infectious diseases including tuberculosis, acute respiratory and gastrointestinal infections, vaccine-preventable diseases, sexually transmitted infections, and certain parasitic infections. In addition, several zoonotic diseases (animal to human transmission) occur because of the close contact between humans and their domesticated animals [6].  Pastoralists are underserved in all forms of health care. Mobility, low population densities, and dispersed communities can make the provision of health care to nomadic groups challenging [7]. These factors are often compounded by the lack of accurate maps. Service provision to nomadic populations requires knowledge not only of the geographic terrain, but also of other variables such as climate, public infrastructure and human activity. Much of this information was not available at the inception of the humanitarian project in Whantoa district. The geographical isolation of the district and limited development meant there was little knowledge of topography and transhumance patterns beyond Whantoa.

III. Methodology

Terrain Mapping

Data collection and analysis were undertaken by personnel involved in the delivery of mobile health services and included both national and international staff. Data were collected over a five-month period spanning the wet and dry seasons in order to capture environmental and social variations.

We employed a participatory approach to data collection. Participatory mapping has been used extensively over the past two decades, particularly in the areas of natural resource management, risk assessment, and indigenous cultural knowledge [8]. The social and cultural values applied to landscapes are subjective, contextual, and tend to vary spatially. Capturing these values can be difficult and can only be achieved with the participation of local inhabitants [9]. In this research, key informant interviews, focused discussions with community leaders, and document review were the principle methods used to gather and record spatial and ethnographic data. Participatory methodologies draw extensively on the expert knowledge of local inhabitants. When applied to mapping, these methods have the potential to yield detailed social and spatial representations of communities [10].

The first task of the project was to produce a map of the Whantoa district. Existing maps were inaccurate and incomplete, but were used as a reference point for determining the district boundary, the approximate location of wet and dry season settlements, and an initial charting of clan affiliations. Satellite images were used to ascertain the location and course of river systems that flow through the district. Knowledge of these waterways was crucial in understanding both settlement and migration patterns. Furthermore, existing maps did not provide information on the different types of settlements, including whether they were temporary dry season camps or permanent wet season settlements. Over a five-month period, we visited many of these camps and settlements and recorded their location and name as well as additional information on seasonal migrations to and from each settlement. We also recorded the timing and destination of seasonal migrations in order to gain an understanding of residency patterns in the district.

Understanding Clan Membership

The mapping research also investigated three interrelated aspects of Gaagwang Nuer society: population mobility, kinship structure, and agricultural and pastoral activities. Wet season residence in Gaagwang Nuer society is determined exclusively by clan membership. We documented the kinship structure of Gaagwang Nuer society in order to understand residence and migration patterns. Through key informant interviews, we investigated clan divisions and subsequently confirmed this information with senior clan members. We conducted ten interviews across the six sites where mobile health services were delivered. In addition we visited sixteen sites for the purposes of collecting information relating to clan membership, residential affiliations, and seasonal movements. As much of this information gathering was opportunistic, we were unable to ensure a diverse range of perspectives. Nevertheless we endeavored to confirm data through informant feedback. Graphic representations of data were distributed to national staff for comment. A number of adjustments were made particularly relating to seasonal movements. Hand-drawn maps were discussed with senior clan members to ensure our interpretations were valid.

Research into the history of the region was limited by a paucity of information relating to the Gaagwang Nuer. The exception was Edward Evans-Prichard’s seminal work, The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People [11]. This work informed inquiries regarding the agricultural and pastoral practices of the Gaagwang Nuer. It enabled comparisons between contemporary practices and those recorded by Evans-Pritchard in the early twentieth century. 

IV. Results

Residency and Kinship Affiliations

The political and social organization of Nuer society has been described as “a combination of kinship and residency affiliations” [12]. The Eastern Jikany is one of 11 major named territorial groupings of the Nuer nation composed of the Gaajok, the Gaajak and the Gaagwangclans. Eastern Jikany is a dialect of the Nuer language and is spoken by people living in the eastern part of the Upper Nile State in South Sudan. The Gaagwang clan lives predominantly in the Whantoa district, although their territory encapsulates a broad tract of land that extends from Western Ethiopia to South Sudan. There are three sub-clans of the Gaagwang: the Nyang, the Gatcieke and the Ciengingiey. Each of these sub-clans has further divisions. The fourth and final sub-classification is based on a place of residence, indicating the kebele (neighborhood) in which a sub-clan’s permanent or wet season residence is located.  Table 1 lists the divisions of theNyang sub-clan and their kebeles. We were also able to establish residential affiliations for all divisions of the Gatcieke and Ciengingiey sub-clans, which gave a complete record of wet season residence in the Whantoa district.

JHA Watson Table 1

In the Whantoa district a total of thirty-two permanent settlements were recorded. In addition, we recorded 28 temporary settlements, most of which are situated on one of three river systems that pass through the district. The Baro, Makwei, and Dhure Rivers provide year-round access to water.

Place of residence and kinship are closely associated in Nuer society. While some families are related by birth or through marriage, others become “kin” through co-residence, labor exchange or other forms of mutual cooperation [13]. Wet season settlements hold communal rights to surrounding water sources and grazing lands. In the dry season, grazing pasture is limited to river-flooded lowlands. In Whantoa district there is an abundance of such pasture that is collectively owned and controlled by clan groups. The temporary dry season cattle camps established on these pastures bring together people and cattle from across the district. Both forms of seasonal residency are underpinned by kinship networks, as are seasonal migrations. In some wet season settlements, people disperse to different dry season cattle camps only to convene again during the wet season. Knowledge of this annual process of division and union among Gaagwang Nuer communities was fundamental to our understanding of the movement of people and their cattle across Whantoa district.

Seasonal Migrations

In addition to the settlements themselves, we also mapped seasonal movements between all permanent and temporary settlements in Whantoa district. These movements are determined by a number of factors with cattle welfare as the primary driver. In February and March, drought conditions force households to leave their permanent cieng (village). They generally move closer to rivers or permanent waterholes, where they establish a temporary wech (camp). The wech is established in the same location each year in order to secure dry season pasture. The decision to leave the wech and return to the cieng is also based on cattle welfare and is initiated when mosquitoes become bothersome to cattle and flooding is imminent. Knowledge of settlement patterns and seasonal flooding across Whantoa district was central to health service planning. During the Nuer seasons of Tot and Reul (June to September) widespread flooding makes vehicular access impossible and many settlements can only be accessed by river. The interplay between geography, climate, and pastoralism emerged as the principle determinant for how mobile clinics could operate in Whantoa district.

Figure 1 shows the annual migrations of the Gaagwang Nuer in Whantoa district from temporary cattle camps to permanent settlements. All of the temporary camps are located along permanent water sources. These are abandoned for the permanent villages in July and early August.

JHA Watson Fig1

Figure 1. Seasonal Population Movements of the GaagwangNuer, Whantoa District, Ethiopia


Agricultural activities also contribute to population movements, though to a lesser extent. If maize has been planted at the wech, the Gaagwang move their cattle to an intermediate location, usually with noshelter. They wait until the maize has been harvested before moving to their wech, which ensures that cattle do not eat the maize. The staged approach to seasonal migration is determined by the need to plant maize. Hutchinson states that those Eastern Jikany Nuer living along the upper Sobat and BaroRivers are able to produce two harvests annually as the riparian land yields abundant and fast maturing maize crops [14]. Table 2 outlines the seasonal agricultural and pastoral endeavors of the Gaagwang Nuer. Both activities are closely linked to annual rainfall patterns. Our information on the agricultural and pastoral practices of the Gaagwang Nuer was found to be similar to descriptions recorded by Evans-Pritchard in the 1940s. Evans-Pritchard described two seasons of equal duration, stating that the “seasonal dichotomy combined with pastoral interests” were the principal determinant of political organization in Nuer society [15]. While animal husbandry was a year-round activity, agriculture was a seasonal practice generally restricted to elevated ground on which the permanent settlements are located. Heavy rainfall from June through August results in widespread flooding of low-lying land, making it unsuitable for habitation, grazing, and agriculture [16].

JHA Watson Table 2

The movement of people and their cattle can also occur for reasons other than agriculture or pastoralism. Famine, marriage, or inter-family conflict may require extensive travel. In such circumstances, people usually travel to the territory of their own sub-clan. In theory, people are free to settle where they choose, as long as they do not infringe on the territory of others. However, most prefer to live with their fellow clansmen[17].

V. Discussion

Many of the nomadic pastoralist communities of East Africa are both geographically and socially marginalized [18]. The provision of health services to nomadic populations is logistically complicated and more expensive than fixed-site services [19], [20]. Nevertheless, our research demonstrates that nomadic migrations are generally predictable, which is conducive to healthcare planning and service delivery.  By understanding settlement and population movement patterns we were able to formulate a seasonal strategy for delivering mobile health services. We identified potential sites for mobile clinics and determined how and when our team might access these sites and which modes of transport would be required. Pastoralists tend to dwell in remote and often insecure environments beyond the reach of mainstream health services. Disease control activities in such circumstances are often difficult to implement. Being able to follow a particular group of people throughout the year is essential for planning health services such as vaccination programs. The ability to adapt services to pastoralists’ mobile lifestyle optimizes the design of service delivery, thus ensuring health interventions are sustainable.

In this study, we combined geography with anthropological frameworks to develop an operational base for the provision of healthcare in a remote setting inhabited by nomadic pastoralists. We were able to develop a detailed understanding of landforms and the social and cultural customs around land use, which were crucial to the implementation of outreach programs such as mobile clinics and vaccination campaigns. Mobile clinics targeted those settlements along the Baro and Dhure Rivers from June to September, during which widespread flooding across the district limited vehicular access and services could only be delivered by boat. From October through April, vehicular access was generally possible and the focus of service delivery shifted from temporary camps to permanent villages. It is important to note that this shift did not guarantee access for all people because most temporary camps were an amalgam of family groupings from different settlements, some of which were beyond the reach of dry season mobile clinics.

Accurate spatial and social data is essential in planning humanitarian interventions. In many developing countries spatial data is either absent or inaccurate. In particular, the temporal resolution of spatial data may lag behind the situation on the ground [21]. Producing a map of the Whantoa district contributed to our understanding of landforms; however, it did not explain land use nor did it reveal the social and cultural practices linked to agriculture and pastoral activities of the GaagwangNuer. We used the concept of ground truth validationto complement existing geographic maps with a range of human activity indicators [22]. Ground truth is a well-established principle used in cartography and meteorology that refers to information that is collected on the earth’s surface to complement data collected through remote sensing techniques [23]. Through our interviews we were able to access local knowledge regarding land use activities, settlement patterns, and seasonal movements. This knowledge enabled an understanding of both the physical and social environment.

Understanding the annual migrations of people across the district required knowledge beyond that of departure and destination points. As discussed earlier, the Gaagwang Nuer practice a form of migration known as transhumance, the annual movement of people and livestock between wet and dry season pastures. In transhumance, the patterns of migration are generally predetermined under normal environmental circumstances [24]. In the Whantoa district, the physical environment, both landform and climate, emerged as the primary determinant for the movement of people and their cattle, which ultimately influenced the type, location, and means by which health services could be delivered throughout the district.

Lastly, understanding the interplay between people and the landscape as well as the complex social and cultural associations that shaped patterns of livelihood and habitation allowed us to establish an operational base in the Whantoa district. The final district map with landform, settlement, and migration patterns enabled planning for year-round service delivery through mobile clinics.

Humanitarian interventions are increasingly complex and costly to fund. A number of writers have claimed that development projects are often implemented with insufficient knowledge of the social and cultural environments in which they operate [25], [26]. The humanitarian sphere, in particular, does not emphasize the role of culture and the value of the discipline of anthropology [27]. Shared social learning—an ethnographic approach to field work in which understandings of social and cultural realities are actively sought—may be particularly important in humanitarian work [28]. This approach may be constrained by a lack of training in research and analysis in combination with ethnocentric attitudes and inherent difficulties in understanding the complex multiple realities of humanitarian environments [29].

In the past decade, a number of geospatial technologies have enabled humanitarian organizations to represent, analyze, and integrate geographical data for use in operational planning. Maps are undoubtedly important tools for the design and coordination of humanitarian interventions. While sophisticated computer-based tools are available for analyzing and representing geographical and demographic data, these may not be available in low-resource or remote settings. Volunteered geographical information in emergency contexts has been found to be useful; however, this approach is dependent upon established information technologies in field sites [30]. While humanitarian organizations rely on mapping to pursue their objectives, they often lack the tools and skills to acquire data[31]. Geographic and social data for remote locations is often incomplete and can be difficult to obtain. A participatory approach to data collection and map production as outlined in this research is a powerful method for generating such data. It is a method that is sensitive to variation within discrete geographic and socio-cultural populations. Such approaches have been used in a variety of settings including forestry [32], indigenous land management [33] risk identification [34] and biodiversity conservation [35], generating both the geospatial and social data necessary to understand local contexts. Smith, Barrett, and Box state that the value of participatory methods is realized when research is concerned with “spatial and temporal variation in perceptions” [36]. It is an approach to fieldwork worthy of consideration by the humanitarian community. There are both normative and practical arguments for inclusive approaches to humanitarian planning. Ideally, people affected by development strategies, regardless of the context, should be involved in the planning of programs that are likely to impact their lives. On a practical level, local people bring to the planning process expert local knowledge. This expertise encompasses the social, cultural and material conditions of life—factors that are likely to enhance relevance and sensitivity to the context in which humanitarian organizations operate [37].

VI. Conclusion

This paper describes the methods used to map the seasonal population movements of the Gaagwang Nuer in the Whantoa district of western Ethiopia. Mobility is a key dynamic of many pastoral communities. While the seasonal movement of people and their livestock has enabled adaptation to semiarid environments, it has also made the provision of healthcare challenging. However, our research suggests that the movement of nomads is generally predictable, which facilitates healthcare planning and service delivery. The very factors that determined the seasonal movement of the Gaagwang Nuer—landforms, climate, and kinship—become key considerations in planning and coordinating service delivery. The participatory methodology used in this research not only detailed a geographic representation of Whantoa district, but also the social and cultural relationships that inhabitants of this region have with their environment. Such participatory methodologies may have applications in planning services for other nomadic pastoralist societies.



[1] MapAction, “Field guide to humanitarian mapping,” MapAction 

http://www.humaninet.org/mapactionfieldguide.pdf (accessed February 8 2014).

[2] Sharon Hutchinson, “A guide to the Nuer of Jonglei State,” http://www.cmi.no/file/1962-Nuer.pdf (accessed October 20 2014).

[3] Fabienne Nackers, “Cross-sectional survey on nutrition, measles vaccination coverage and retrospective mortality in Whantoa Woreda, Gambella Region, Ethiopia.” (Paris: Epicentre, Medecins Sans Frontieres 2010).

[4] World Health Organization, Map of Migratory Routes of the Nomads in Northern and Southern Red Sea Zobas, (WHO-Eritrea, 2007): 12.

[5] Tewodros Dubale and Damen Haile Mariam, “Determinants of conventional health service utilization among pastoralists in northeast Ethiopia,” Ethiopian Journal of Health Development 21 (2007): 142-147.

[6] Avinoam Meir, “Nomads, development and health: Delivering public health services to the Bedouin in Israel,” Geografiska Annaler 2 (1987): 115-126.

[7] Jakob Zinsstag, Moustapha Ould Taleb and Criag Philip “Editorial: health of nomadic pastoralists: new approaches towards equity effectiveness,” Trop Med Int Health 11 (2006): 565-568.

[8] Robert Chambers, “Participatory mapping and geographic information systems: Whose map? Who is empowered and who is disempowered? Who gains and who loses?,” Electronic Journal on Information Systems in Developing Countries 25 (2005): 1-11.

[9] Nora Fagerholm and Niina Kayhko, “Participatory mapping and geographical patterns of the social landscapes of rural comunities in Zanzibar, Tanzania.” Fennia 187 (2009): 43-60.

[10] Michael Goodchild, “Citizens as sensors: The world of volunteered geography,” Geojournal 69 (2007): 211-221.

[11] Edward Evan Evans-Pritchard, The Nuer : a description of the modes of livelihood and political institutions of a Nilotic people (New York; Oxford, Oxford University Press, 1969), 16-93.

[12] Sharon Hutchinson, Nuer Dilemmas: Coping with Money, War and the State (Berkeley, University of California Press 1996), 22.

[13] Hutchinson, “A guide to the Nuer of Jonglei State,” 10.

[14] Ibid., 9.

[15] Edward Evan Evans-Pritchard, “The Nuer of Southern Sudan”, http://instruct.uwo.ca/anthro/301/eppoli.htm  (accessed October 20 2014).

[16] Ibid., 273

[17] Hutchinson, “A guide to the Nuer of Jonglei State,” 8.

[18] Zinstagg, Ould Taleb and Philip (2006).

[19] Naseem Qureshi, Muzamil Abdelgadir, Aladin Hadi Al-Amri, Talal Hussain Al-Beyari and “Strategies for enhancing the use of primary health care services by nomads and rural communities in Saudi Arabia,” Eastern Mediterranean Health Journal 2(1996): 326-333.

[20] Abdikarim Sheik-Mohamed and Johan Velema, “Where health care has no access: the nomadic populations of sub-Saharan Africa,” Tropical Medicine & International Health 4(1999): 695-707.

[21] Andrew Crooks and Sarah Wise, “GIS and agent-based models for humanitarian assistance,” Computers Environment and Urban Systems, 41(2013): 100-111.

[22] Rashid Ansumana, Anthony Malanoski, Alfred Bockarie, Abu James Sundufu, David Jimmy, Umaru Bangura, Kathryn Jacobsen, Baochun Lin and David Strenger, “Enabling methods for community health mapping in developing countries.” International Journal of Health Geographics 9(2010): 56-63.

[23] George Garrity, “Ground Truth,” Standards in Genomic Sciences 1(2009): 91-92.

[24] World Health Organization, (2007): 12.

[25] Munzoul Assal, “A discipline asserting its identity and place: Displacement, aid and anthropology in Sudan,” Eastern Africa Social Science Research Review 18(2002): 63-96.

[26] Michel Thieren, “Health information systems in humanitarian emergencies.” Bulletin of the World Health Organization 83(2005): 584-589.

[27] Pierre Minn, “Toward an anthropology of humanitarianism.” Journal of Humanitarian Assistance(2007).

[28] Raymond Apthorpe and Phillipa Atkinson, “Towards shared social learning for humanitarian programmes,” Active Learning Network for Accoutability and Performance in Humanitarian Assistance, http://www.hapinternational.org/pool/files/towardssharedsociallearning.pdf (accessed February 7, 2014)

[29] Ibid.,  6

[30] Wilfried Tissot and Yann Rebois, “Using Geomatics to increase the exchange of information between humanitarians- an idea worth looking into,” Humanitarian Aid on the Move, http://www.urd.org/IMG/pdf/URD_HEM_3_UK.pdf (accessed February 7, 2014)

[31] Matthew Zook, Mark Graham, Taylor Shelton and Sean Gorman, “Volunteered geographic information and crowdsourcing disaster relief: A case study of the Haitian earthquake, World Medical and Health Policy 2(2012): 7-33.

[32] Peter Cronkleton, Marko Antonio Albornoz, Grenville Barnes, Kristen Evans and Wil de Jong, “Social Geomatics: Participatory forest mapping to mediate resource conflict in the Bolivian Amazon,” Human Ecology 38(2010): 65-76.

[33] Richard Chase Smith, Margarita Benavides, Mario Pariona and Ermeto Tuesta, “Mapping the past and the future: Geomatics and Indigenous Territories in the Peruvian Amazon,” Human Organization 62(2003): 357.

[34] Kevin Smith, Christopher Barrett and Paul Box, “Participatory risk mapping for targeting research and assistance: With an example from East African pastoralists,” World Development 28(2000): 1945-1959.

[35] Peter Poole, “Land-based communities: Geomatics and biodiversity conservation,” Cultural Survival Quarterly 18(1995): 74.

[36] Kevin Smith, Christopher Barrett and Paul Box, “Participatory risk mapping,” 1957.

[37] Maria Julia, Maria Kondrat, ‘Health care in the social development context”, International Social Work 48(2005): 537-552.

Comments are closed.