Marshall 1 Nate Marshall Parks and Peoples: Dilemmas of Protected Area Conservation in East Africa Will Da Beast Return? The State of the Serengeti s Great Migration Parks and terrestrial mammal migrations seem mismatched. Parks, with their rigid boundaries and regulations, are inflexible. Migrations, however, are flexible by nature. They occur because species exploit inconstant resources, so their location follows resource availability rather than precise, predictable patterns. Keeping a migratory population bounded within a park, then, is like balancing a marble on a dome: an unstable equilibrium. Perhaps this is part of the reason why, while protected areas spread globally, Earth s migrations collapsed. The great buffalo migrations of the American Great Plains, the elephant migrations of northern Cambodia, and the Bornean sun bear migration, to name a few, all disappeared due to land development and other pressures. 1 Protected areas now cover 15.4% of Earth s terrestrial surface, which is encouraging news for long-term wildlife prospects, yet this still could not save the aforementioned migrations. Clearly, the mere existence of protected areas is not sufficient to save mammalian migrations; the protected areas must be well-situated and must curtail harms to wildlife. Tanzania has both an astonishing 38% of its land designated for conservation and what many consider the largest remaining migration: the so-called Great Migration. 2 Are Tanzania s protected areas sufficient to save the Great Migration from collapse? The Serengeti ecosystem exhibits stark geobiological contrasts that drive the Great Migration. The northeastern part of the Serengeti ecosystem abuts Lake Victoria, one of the largest freshwater lakes in the world. Particularly during the dry season (May-October),
Marshall 2 thunderstorms from the lake bring moisture to the dry shrubland and savanna, maintaining the quality of forage. During the wet season, the northeast receives over a meter of rain while the parched southeastern half of the ecosystem generally receives under half a meter. 3 Due to the low rainfall and hardpan soil - the legacy of the Serengeti s close proximity the East African Rift and its periodic volcanic pyroclastic deposits - the southeastern plains support few trees and appear quite barren during the dry season (image 1). 4 In the wet season, however, rains brighten the southeast, producing a short-lived pulse of green grass. The red oat grass in the southeast is anomalously nutritious, containing twice the concentration of phosphorus as the more perennial grass in the northeastern Serengeti. 1 Thus, ungulate species migrate to the southeastern Serengeti plains for the nutritious but seasonal red oat grass and then spend the dry season consuming the perennial grasses to the northwest. Image 1 A lone tree accentuates the seemingly endless plains of the southeastern Serengeti while late-dry season storms gather in the distance. While the migration itself exists on a macro scale, the species involved exhibit unique micro-scale adaptations. Wildebeest align their calving season to early in the dry season to take
Marshall 3 advantage of the rich phosphorus concentrations in the southwestern Serengeti. While lactating, wildebeest mothers require high phosphorus intake. 1 Then, the calves are born, amazingly not only during the same season, but within the same day, minimizing the risk of predators consuming unsteady newborns. 5 Wildebeest and the two other dominant migratory ungulates in the Serengeti, zebras and gazelles, also exhibit a grazing succession. Zebra, which are larger and have a hardier digestive system, first consume the tall, old grass. Wildebeest, which are intermediate in size, follow (image 2), and the small gazelles consume the remaining nutrientrich, tender shoots after the larger animals pass. In this way, the migrants create grazing niches for each other. 6 Like cascading dominoes, the grazing of zebras triggers the wildebeest to follow, which in turn tips the gazelles to chase. These strategies - and the migratory cycle itself - elevate the Serengeti ungulate populations relative to a sedentary populations because the migrants exploit inconsistent resources. Though populations vary, there are about two million migrants - 10% zebras, 60% wildebeest, and 25% gazelles. 6 In total, these migrants travel as far as 600-800 kilometers per migratory loop around the Serengeti ecosystem. 7 image 2 Dark-colored wildebeest follow striped zebras in a mixed herd, illustrating the grazing succession (though in the Great Migration itself, wildebeest often lag much farther behind.)
Marshall 4 The Great Migration has coexisted with human habitation for thousands of years. The adjacent human civilizations, however, are rapidly changing. Population growth, development, and agricultural encroachment are changing the landscape of northern Tanzania, which could threaten the Serengeti ecosystem. To evaluate how human activity has already impacted the Serengeti ecosystem and shed light on the ecosystem s future, I will investigate the effects of changing land use and infrastructure on the Great Migration since Serengeti National Park s establishment in 1959. My research is guided by three hypotheses, which I evaluate using the scientific literature. Figure 1 The Great Migration generally follows a circular pattern, travelling south along the eastern boundary of Serengeti National Park (the outlined figure) and north along the western boundary and corridor of the park. Note the Serengeti ecosystem, generally defined by the migration extent, is larger than the park itself. The migration may leave the park to the north, east, south, or west at various times of the year. 8 Hypothesis 1: Agricultural intensification has shifted the migration path Discussion of the migration path must begin with context about categories of land protection. The strictest form of protection in northern Tanzania is the national park, which prohibits almost all human activity. Serengeti National Park encompasses the entire central part of the Serengeti ecosystem. Though humans occupied this region prior to the park establishment,
Marshall 5 humans have been excluded from this area for the duration of my study time frame (since 1959). The second tier of protection includes games reserves such as Kenya s Maasai Mara to the north, Grumeti Game Reserve to the west, and Ngorongoro Conservation Area to the south. In Ngorongoro, human settlement is permitted but importantly, farming is restricted. Game reserves prohibit farming as well. Unprotected areas also comprise parts of the Serengeti ecosystem to the west, north, and east. However, the migratory population spends only 7.5% of the year in unprotected areas. 9 That the migration occurs within protected areas for 92.5% of the time first of all validates the efforts to protect the Serengeti Great Migration throughout its extent. Because human interference is restricted within protected areas, this also restricts my investigation of where the migration has changed course to the portion of the migratory range outside of protected areas. Land use change in the unprotected portion of the Serengeti ecosystem has indeed evolved dramatically in recent decades. In the inner ranches area surrounding the Maasai Mara Game Reserve, for example, agricultural extent increased from 12,000 acres in 1975 to 120,000 acres in 1995. 10 The strip of land west of Serengeti National Park, between Lake Victoria and the park, has long been heavily cultivated, but during the 21st century, agriculture increased and expanded up the boundary of the park. 11 These examples are evidence of a troubling trend throughout the Serengeti ecosystem: with human population rising at 2.8% annually, 12 people are increasingly farming previously undeveloped savanna for subsistence. Historically, the dominant group to the north, east, and south of the Serengeti, the Maasai, herded cattle and practiced little agriculture. With land privatization and population growth, however, many have no choice but to farm. A common benchmark subsistence level for pastoralists is five total livestock units per adult unit. In the Maasai Mara ranches, however, per capita livestock units now sits at two-
Marshall 6 three. 12 More alarmingly, residents of Ngorongoro Conservation Area, where the migration spends 36% of its time, 9 have only 2.7 total livestock units per adult unit. 13 Though farming was historically banned in Ngorongoro, many now have no other choice but to cultivate for survival. 12 The issue with farming is that unlike herding cattle, farming engenders antipathy between humans and wildlife. This is most clearly evident when wildlife destroy crops, prompting retaliatory killings. In many areas throughout east Africa, the spread of farming has demonstrably damaged wildlife populations. 14 The question, then, is whether agricultural encroachment has affected the Great Migration outside of protected areas. A 2004 study compared the tracks of tagged wildebeest in 1971 and 1999 and concluded that the migration has changed course within the boundaries of the protected area, independent from human influence, but has not been affected out of the protected area. In fact, they found that the wildebeest ventured farther into Maasai Mara game reserve, where they are in particularly close proximity to people, than they previously did. 9 However, another intensive study from 2005, which analyzed a much larger number of species and individuals, noted that areas [of the Serengeti ecosystem] with settlements are poor in wildlife. This was true despite the observation that in settled areas, the availability of forage is not limiting, 15 meaning the migrants avoided settled areas because of the human population, though food was available. This implies that as the human population has increasingly settled up to the border of the protected areas, there have been fewer migrants leaving the protected areas. Ultimately, there is no long-term time series of migratory dispersion that would enable me to reach a definitive conclusion about this hypothesis. Because so much of the migration occurs within protected areas though, I do not believe human encroachment has not substantially
affected the migration route. The rising human population poses additional threats to the migration besides changing its route, and the next hypothesis will explore these challenges. Marshall 7 Hypothesis 2: Increasing population of adjacent human settlements has decreased the migratory population The human population in the Serengeti region is growing 2.8% annually, but this growth is not spatially uniform. At a distance between 20 and 60 kilometers from the protected area boundary, growth is lower - about 2.6% each year. Surprisingly, the growth rate spikes to 3.5% in close proximity (>20 km) to the boundary. Though the human population density currently decreases with increasing proximity to protected areas, density is increasing most rapidly where wildlife are most likely to be found. 11 Intriguingly, wildlife also tend to congregate along the borders of protected areas. Though the migratory population spends only 7.5% of the year outside of protected areas, it is found within 10 km of the boundary 33% of the time. 9 It is as if the protected area boundary is a battle line and two armies - the migratory wildlife and the human population - are assembling on each side. The key question: are these armies confrontational? Or are they nonviolent, passively living on their respective sides of the line? Though instructive, this battle line metaphor fails to address some of the historical complexity of the human-wildlife relationship. The Wakuria people, predominant inhabitants of the agricultural strip between the Serengeti and Lake Victoria, have long relied on the migration as a seasonal source of meat. Oral interviews of Wakuria individuals revealed that 92.7% would be willing to kill wildlife for subsistence and 93.2% would kill to sell as bushmeat. 16 The hunting data corroborate these interview statistics. Nearly twenty years ago (the most recent data that I could uncover,) about 40,000 wildebeest were killed annually. At that time, researchers predicted
Marshall 8 that population collapse would occur at 80,000 annual kills. 17 Unfortunately, there is no consensus on the magnitude of the harvest. 15 Furthermore, the migratory population is not solely determined by the magnitude of human harvest in the first place. Various other factors including food availability, climate, predator populations, and disease, act as confounding variables, preventing us from concluding that human activity has caused a decline in migratory populations. Were a correlation to exist between a growing human population and a declining migratory population, we could not definitely conclude the human population causes the migratory decline. With this in mind, the critical question: how has the migratory population changed over time? is still worthwhile to raise. Image 3 Despite their slender body, wildebeest provide an important source of meat and sustenance for some native peoples. Imagine 1.3 million of these animals migrating in a vanishing line across the Serengeti plains. Since 1960, the wildebeest population (wildebeest have the highest population among the Serengeti migrants and are thus the most intensively studied) exhibits a dramatic increase followed by three decades of oscillation (Figure 4). 18 The rapid increase depicts the population recovering from a rinderpest epidemic, while the oscillation could be explained by human
Marshall 9 activity or by any of the aforementioned confounding factors. Significantly, there is not a worrying trend of long-term population decline - the modest decline in recent decades may be a natural adjustment to the population briefly exceeding the ecosystem s carrying capacity after the rinderpest recovery. 17 Thus, I conclude that there human hunting is not producing demonstrably negative impacts on the migratory population, though this analysis is confounded by ecological and environmental factors independent of the nearby human population. Figure 2 Wildebeest population from 1959 to 2006. 18 Hypothesis 3: Construction of roads and fences is detrimental to the Great Migration The increasing human population near the Serengeti directly causes an increase in road and fence construction. This could, on one hand, be a beneficial trend because roads can facilitate economic development and thus reduce reliance on hunting migratory species for subsistence or bush meat. On the other hand, roads could also make it easier to sell bushmeat. If people build roads or fences in the path of the migration, collisions and conflict are inevitable. A decade ago, the discussion of roads and migrations grabbed international headlines when
Marshall 10 Tanzania proposed paving a road across the entire northern Serengeti plains that would effectively cut the ecosystem in half. Forecasters predicted that by 2035, the road would average one car every thirty seconds, making collisions a tremendous hazard during the migration. 19 Animals are reluctant to cross a paved road, so many scientists concluded that if the road were built, it would spell the end of the migration. Another simulation anticipated that, assuming no habitat would actually be lost, simply cutting the ecosystem in half would cause a third of the migratory population to disappear. 20 Fortunately for the Great Migration, the Tanzanian government canceled the project in 2011 under immense international pressure, then briefly revitalized the idea as a dirt road, and has since taken no action. Because human populations cannot settle within the protected areas, roads and fences are not a pressing threat to the migration today. In other migrational corridors, however, this is not the case. One migratory corridor that suffers from advancing human infrastructure is the Kaputei Plains in southern Kenya. Like the Serengeti, these plains host an ungulate migration, albeit on a much smaller scale. During the wet season, wildebeest and other ungulates venture from Nairobi National Park into the dry Kaputei Plains and then return to the park for the dry season. In recent decades, the Kaputei Plains became heavily privatized, and privatization has resulted in fences now isolating 20% of the plains. 18 A 2013 report concluded that though only 20% of the land is fenced, very few wildebeest now return to the national park in the dry season due to the fencing as well as the associated expansion of roads and agriculture. 21 Fortunately, this is not the case in the Serengeti because paved roads and enclosures do not exist within the protected areas where migratory populations spend 92.5% of the year. Thus, I confidently reject my third hypothesis and conclude that with the paved northern road defeated, roads and fences are not a historical or an imminent threat to the Great Migration. They may, however, present a distant peril.
Marshall 11 Conclusion The prognosis for the Great Migration is cautiously optimistic. Despite ongoing human hunting and population increase, the wildebeest population is relatively stable. A proposal to block the migration with a paved road through the Serengeti was shelved. Agricultural intensification in the Serengeti region has not posed an existential threat to the migration itself because of the network of protected areas that maintain open lands for migrants to pass through. This sanguine outlook, though, depends on the migration remaining within the boundaries of the protected areas. The wild card is climate change. The climate outlook in northern Tanzania is rather sobering. 1.5 degrees Celsius of warming are expected by 2050 alone. 22 Droughts are likely to be more extreme, making water - already the limiting resource throughout the Serengeti ecosystem - even more scarce. 23 It is impossible to predict how these changes will affect the migration. Sinclair et al. (2008) already suggests that juvenile survivability will fall due to pressure on water and food, lowering the population. 23 Will a warmer, drier climate cause the migration to change its location permanently? Will year-to-year variations in the migration path become more variable? These questions remain to be answered, but one thing is clear: the increase in agriculture, population, roads, and fences right up to the protected area border is cutting off any chance of future flexibility. And history indicates that a migration in an inflexible area is soon a missing migration. For the Great Migration to continue, the climate must remain stable enough to permit the migration to continue occurring within the protected areas anchored by Serengeti National Park. That is an issue that transcends the Serengeti and requires all of us to contribute.
Marshall 12 Image 4 In the late dry season, the first migratory zebras begin to arrive in the central part of the Serengeti Plains. Image 5 A group of wildebeest seek shade on a warm day. Not all wildebeest migrate. Some herds remain the residents of a specific region year-round.
Marshall 13 1 Wilcove (2008) 2 United Republic of Tanzania. 3 Norton-Griffiths et al. (1975) 4 Pers. comm. with Lesikar Naalasi, Hoopoe Safaris 5 Estes & Estes (1979) 6 Gayuka et al. (2012) 7 Berger (2004) 8 Burrard-Lucas (2011) 9 Thirgood et al. (2004) 10 Serneels (2001) 11 Estes et al. (2012) 12 Lamprey & Reid (2004) 13 Boone et al. (2006) 14 Reid (2012) 15 Rusch, et al. (2005) 16 Kalterborn et al. (2005) 17 Mduma et al. (1998) 18 Norton-Griffiths (2007) 19 Saving the Great Migrations 20 Holdo et al. (2011) 21 Ogutu et al. (2013) 22 Craparo et al. (2015) 23 Sinclair et al. (2008) Notes Works Cited Berger, J. (2004), The Last Mile: How to Sustain Long-Distance Migration in Mammals. Conservation Biology, 18: 320 331. doi:10.1111/j.1523-1739.2004.00548.x Boone, R., K. Galvin, P. Thornton, D. Swift, M. Coughenour. (2006). Cultivation and Conservation in Ngorongoro Conservation Area, Tanzania. Human Ecology, 34(6): 809-828. https://doi.org/10.1007/s10745-006-9031-3 Burrard-Lucas, W. (2011). Great Migration. Burrard-Lucas Wildlife Photography. Retrieved from blog.burrard-lucas.com/2011/03/great-migration/ Craparo, A. C. W., Van Asten, P. J. A., Läderach, P., Jassogne, L. T. P., & Grab, S. W. (2015). Coffea arabica yields decline in Tanzania due to climate change: Global implications. Agricultural and Forest Meteorology, 207, 1-10. Estes, A. B., T. Kuemmerle, H. Kushnir, V. Christian, R.Herman, H. Shugarta. (2012). Land-cover change and human population trends in the greater Serengeti ecosystem from 1984 2003. Biological Conservation. 147: 255-163. https://doi.org/10.1016/j.biocon. 2012.01.010 Estes, R., R. Estes. (1979). The birth and survival of wildebeest calves. Ethology, 50(1): 45-95. doi: 10.1111/j.1439-0310.1979.tb01015.x
Marshall 14 Gakuya F, Ombui J, Heukelbach J, Maingi N, Muchemi G, et al. (2012). Knowledge of Mange among Masai Pastoralists in Kenya. PLOS ONE, 7(8): e43342. https://doi.org/10.1371/journal.pone.0043342 Holdo, R., J. Fryxell, A. E. Sinclair, A. Dobson, R. Holt. (2011). Predicted impact of barriers to migration on the Serengeti wildebeest population. PLOS ONE, 6(1): e16370. doi:10.1371/journal.pone.0016370 Kalterborn, B. P., J. W. Nyahongo, K. M. Tingstad. (2005). The nature of hunting around the Western Corridor of Serengeti National Park, Tanzania. European Journal of Wildlife Research. 51: 213-222. Lamprey, R.H., and Reid R.S., 2004. Expansion of human settlement in Kenya's Maasai Mara: What future for pastoralism and wildlife? Journal of Biogeography 31(6) 997 1032. Mduma S. R., Hilborn R., Sinclair A. E. (1998) Limits to exploitation of Serengeti wildebeest and implications for its management. In: Newbury DM, Prins HHT, Brown P (eds) Dynamics of tropical communities. Blackwell Science, Oxford, 243 265 Norton-Griffiths, M., Herlocker, D., Pennycuick, L. (1975). The patterns of rainfall in the Serengeti Ecosystem, Tanzania. African Journal of Ecology, 13: 347 374. doi:10.1111/j.1365-2028.1975.tb00144.x Norton-Griffiths, M. (2007). How many wildebeest do you need? World Economics, 8(2): 41-64. Ogutu, J., Owen-Smith, N., Piepho, H., Said, M., Kifugo, S., Reid, R., Gichohi, H., Kahumbu, P., Andanje, S., 2013. Changing Wildlife Populations in Nairobi National Park and Adjoining Athi-Kaputiei Plains: Collapse of the Migratory Wildebeest. Open Conservation Biology Journal, 7, 11-26. Reid, R. (2012). Savannas of Our Birth: People, Wildlife, and Change in East Africa. Oakland: University of California Press. Rusch, G. M., Stokke, S., Røskaft, E. Mwakalebe, G., Wiik, H., Arnemo, J. M. & Lyamuya, R. (2005). Human-wildlife interactions in western Serengeti, Tanzania. Effects of land management on migratory routes and mammal population densities. Norwegian Institute for Nature Research Report 85. 47 pp. "Saving the Great Migrations: Declining Wildebeest in East Africa? (2013). United Nations Environment Programme. Retrieved from na.unep.net/geas/ getuneppagewitharticleidscript.php?article_id=107 Serneels, S., M. Y. Said & E. F. Lambin. (2001). Land cover changes around a major east African wildlife reserve: The Mara Ecosystem (Kenya). International Journal of Remote Sensing. 22:3397-3420. http://dx.doi.org/10.1080/01431160152609236 Sinclair, A. E., C. Packer, S. R. Mduma, J. M. Fryxell. (2008). Serengeti III. Chicago: The University of Chicago Press. Thirgood, S., A. Mosser, S. Tham, G. Hopcraft, E. Mwangomo, T. Mlengeya, M. Kilewo, J. Fryxell, A. Sinclair & M. Borner. (2004). Can parks protect migratory ungulates? The case of the Serengeti wildebeest. Animal Conservation. 7: 113-120. doi:10.1017/s1367943004001404
Marshall 15 Wilcove, D. S. (2008). No Way Home. Washington DC: Island Press. United Republic of Tanzania. Protected Planet. Retrieved from www.protectedplanet.net/country/tza