PURPOSE
The future of many pastoral societies is in the hands of policy makers who attempt to balance the needs of resident pastoralists with national priorities, including conservation. One of the ways this is attempted is through the establishment of multiple-use protected areas. The Maasai pastoralists of the Ngorongoro Conservation Area (NCA) in northern Tanzania believe that local policies have favored tourism and wildlife conservation interests to the detriment of their well being (Homewood and Rogers 1991). Policies that may affect human welfare include: 1) Direct and indirect restrictions on traditional livestock management practices and movement patterns, and 2) Limitations on crop agriculture. The Loliondo Game Controlled Area provides a control for the study of policy effects upon NCA Maasai. Although the NCA and Loliondo regions appear to differ primarily because conservation policy is less restrictive in Loliondo, ecological differences must be measured, accounted for, and incorporated into policy effects analyses. The NCA and Loliondo areas have similar ecological zones. However landscape structure is different. The proportional representation of lowland, midland and highland zones may be different. These differences are important because seasonal livestock management and crop agriculture use particular areas preferentially. Accessibility of preferred areas may be affected by both landscape structure and policy restrictions on use. NCA Conservation policy was implemented to preserve local wildlife populations and natural resources. Policy has been effective to this end. However, the economic benefits of wildlife conservation have been felt mainly at the national level through tourism gains. Major costs are felt at the local level where both wildlife and conservation policy inhibit pastoral land use (Campbell, 1981). Our goal is to determine what costs are being absorbed by NCA Maasai in the name of conservation. This question may be raised for this and many other conservation areas in East Africa and around the world where traditional human land use is being altered to conserve the resources residents depend on.

OBJECTIVES

• Determine if the human welfare status of NCA Maasai is worse than, better than, or similar to that of their Maasai counterparts in the Loliondo Game Controlled area.
• Determine if current conservation policy may be contributing to existing patterns of land use and human welfare.
• Identify if and how ecology affects Maasai land use and human welfare.
• Quantify the proportional representation and use of the lowlands, midlands and highlands zones in each study area and investigate accessibility of these areas.
• Determine if movement patterns are controlled primarily by ecology or conservation policy.

STUDY AREA
The NCA was formed in 1959 when it was separated from the Serengeti National Park and converted into a multiple-use conservation area designed to balance the needs of resident wildlife and Maasai pastoralists. The NCA and Loliondo Game Controlled Area are both located in north-central Tanzania (Figure 1). They form a portion of the Greater Serengeti ecosystem, which is in part defined by the movements of migrating wildebeest. Maasai-inhabited zones range from approximately 1100m (4000 feet) to 2591m (8500 feet) in elevation. Vegetation includes grassland, bushland, savanna, and highland forest. The variety of wildlife is staggering. The NCA is a popular tourist destination and has been given the status of ‘World Heritage Site’ by the United Nations.

METHODS
I collected data on land use, human welfare and local ecology through interviews, conversation, observation and ecological surveys in both study areas from May through December 1998. Local Maasai were hired as interpreters and guides. Geographic coordinates of all households interviewed and all locations of ecological significance were taken with a Geographic Positioning System (GPS). Statistical analyses were conducted on interview data using the SPSS statistical package. The Kruskal-Wallis non-parametric test was used to accommodate non-normal data. Livestock numbers were converted to Tropical Livestock Units (TLUs), a standard equal to a 250 Kg cow. Conversion factors used are: 1 cow = .71 TLU, Goat or Sheep = .17 TLU A Geographic Information System (GIS) was used to investigate the proportional representation of Maasai-defined lowland, midland and highland ecological zones.

RESULTS
Statistical analyses indicate that Loliondo Maasai are significantly better off than NCA Maasai in terms of per-capita livestock numbers and cultivated acreage (Figure 2, Figure 4). These indicators were used because the most important policy controls on land use appear to be those affecting herd management practices and cultivation. These are also important indicators of food security and human welfare. Significant human welfare differences exist between lowland, midland and highland households in both NCA and Loliondo, with midland residents faring the best. Livestock:human ratio tendencies for each zone are similar between the two sites (Figure 3a). Tendencies are also similar for per-capita crop agriculture (Figure 3b). However, NCA highland Maasai presented an interesting case because most of their cultivation is done outside the highland zone, in the midlands. GIS analyses indicate landscape structure differences between the two areas, with the highly productive midlands much more prevalent in Loliondo than in the NCA (Figure 5).

CONCLUSIONS AND IMPLICATIONS
Human welfare differs significantly between Maasai who live inside and those who live outside the conservation area. This leads to an initial conclusion that conservation policy is the driving force behind well-being. The implication of further analysis is that landscape variations also contribute to differences in human welfare. However, local ecology is reflected in the within-site variation, rather than between-site variation (see Figure 3). Loliondo Maasai appear to be concentrated in the productive midlands, whereas NCA Maasai appear to be concentrated in the highlands. Midlands in the NCA are scarce, and many of those that exist are on a steep slope that is not conducive to habitation, cultivation, or livestock grazing. Additionally, the threat of disease transmission from wildebeest to cattle (malignant catarrhal fever) is great in the NCA lowlands due to policy restrictions on prevention practices such as building fences and other separation methods.

This study indicates that welfare of NCA Maasai is a function of both landscape structure and policy, with policy acting to magnify the effects of ecological factors through restrictions on traditional Maasai management strategies. As a result, NCA Maasai appear to be losers in the game of conservation. The Serengetti National Park is now off-limits to grazing, and land use within the NCA is restricted to a degree that is reflected in the welfare of it’s inhabitants. The vast majority of interviewees stated that it is possible to have wildlife conservation and livestock at the same time. It is interesting that management entities usually believe otherwise. It is critical for policy makers to consider local ecological conditions, human management strategies, and the true impact of pastoralists on wildlife prior to implementation of conservation policy. This will insure that traditional management techniques will not be disturbed, and human welfare will not be compromised.

ACKNOWLEDGEMENTS
The first author would like to extend thanks to everyone who made this research project possible and complete. Terry McCabe for all of his expert advice in the field. Erik Martinson for all of his assistance and support in the field. Randy Boone, for all of his valuable assistance with GIS. All of the officials and local leaders who allowed me to work in Ngorongoro and Loliondo. My interpreters: Joseph OleNgay, Gabriel OleSaitoti, Kois OleTundan, and Gaspar Leboy, without whose help I would have understood nothing. My field guides: Julius OleSitoy, Christophe, and others, without whose help I would have gone nowhere. This project was funded by NSF grant SBR-979762 and the USAID Global Livestock CRSP.

REFERENCES
Campbell, D.J. 1981. Land use competition at the margins of the rangelands: An issue in development strategies for semi-arid areas. In: Planning African Development. Norcliffe, G. and Pinfold, T., eds. Westview Press, Boulder, Colorado.

Dodd, J. 1994. Desertification and degradation in sub-Saharan Africa: the role of livestock. Bioscience 44:28-34.

Homewood, K., and W.A. Rogers. 1991. Maasailand Ecology: Pastoral Development and Wildlife Conservation in Ngorongoro, Tanzania. Cambridge University Press, Cambridge, England.

McCabe, J.T. 1990. Turkana pastoralism: a case against the tragedy of the commons. Human Ecology 18:81-103.