Serengeti National Park

Surface water is scarce throughout Serengeti in the dry season, the only permanent source being the Mara River in the north. A continuous flow in the Mara River is essential for the migration during the dry season and especially during droughts. The Mara river is threatened by unsustainable activities as evidenced by a decrease in the minimum observed flow of the Mara River in the Serengeti National Park during droughts compared between 1972 (1 m3/s) and 2016 (0.16 m3/s). The decrease can in part be attributed to deforestation of the Mau forest in Kenya (Gereta et al., 2002). Water extraction from the Mara River is increasing, especially via large-scale irrigated farms in Kenya e.g. irrigation farming now withdraws about 80% of the available Mara River water during a drought (Kihwele et al., 2020). This is in addition to all other water usage in Kenya.

At the local level, there are very few perennial springs in Serengeti. The most important one for the Park's biodiversity and its Outstanding Universal Value is at Bologonja. Although the spring has very little water flow during the dry season, in October 2017, plans were already in place to lay a substantial pipe to divert some of that water to Seronera.  Based on the size of the water pipe, >50% of that water could be extracted. No EIA can be found but impacts on Outstanding Universal Value is likely.

The proposals for several more dams upstream of Serengeti raises great concern for the cumulative impacts on water and sediment flow.

The Loliondo Game Conservation Area is increasingly overgrazed and unsustainably managed (TAWIRI, 2017; Veldhuis et al., 2017). This area drains westward in small streams to the Serengeti National Park. The flow in those streams now occurs as very short flash floods that cause severe gully erosion; the rest of the time these streams are dry (Kihwele et al., 2020). The eroded sediment is carried by the streams during intense flash floods to Serengeti National Park where it is predicted to silt the few river holes on the eastern side of the Park, and thus make them unavailable as a source of drinking water; these water holes are now an important source of water for resident and migrating animals in the dry season.

Anti-poaching efforts have been strengthened in recent years, and its operations have been successful at the site. The number of elephants poached is in decline (16 elephants poached in 2014/15 compared to 8 in 2016/2017) and no rhinos have been poached. There has been an increase in the elephant population, likely due to a compression effect (e.g. immigration from surrounding areas and from Kenya; Morrison et al., 2018). However, the situation can change quickly and could mount beyond low risk. 

Subsistence poaching for meat targets the migration of wildebeest and constitutes a steady drain on the animal population especially during a drought when several tens of thousands of wildebeest wander out of protected areas in search of fodder and are killed (Gereta et al., 2009; UNESCO and IUCN, 2010). Thousands of wire snares (10 tons of snares) set by poachers have been removed and destroyed by rangers and NGOs during the last 3 years  (FZS, 2019).  Previous to this, studies by Rentsch et al (2014) estimated more than 100,000 wildebeest were being harvested illegally per year (ie 6-10% of the population). Clearly the poaching pressure is un-ending; the recent declines in poaching illustrates the need for sustained anti-poaching effort throughout the Serengeti ecosystem (Hilborn et al 2006). However, as populations of most target species within the Park are stable or increasing, this does not appear to be affecting their overall viability.  Several species are caught as by-catch in snares aimed at migratory wildebeest and zebra.  Unpublished data from road transects by the Serengeti Biodiversity Program (TAWIRI) suggest declines of several populations of resident herbivores but primarily bushbuck across the system and hyena in the western corridor. Snare injuries in the lion and hyena populations remain high as these predators follow snare-lines in search of scavenging opportunities. In the buffer zones, animal diversity decreases outside the protected area with increasing distance from the Park boundaries and with increased land-use (e.g. Kyando et al., 2019) so that in some places only rodents are now found (Shilereyo et al., 2019). Targeted snaring for giraffe (i.e. heavy cables attached to the canopy of trees) has led to declines of the population in many regions of the ecosystem from 1.5-2.6 individuals / km^2 to 0.3-0.4 / km^2 (Strauss et al., 2015).  Dried giraffe meat is likely sold as wildebeest meat.  The focus on giraffe is likely because a large amount of meat can be harvested from a single carcass making it very profitable, however giraffe are Tanzania's national animal and therefore the penalties for poaching are much larger.  Circling columns of vultures scavenging from carcasses at poacher camps are a clear flag used by rangers and other law enforcement personnel to identify illegal activity (Rija et al., 2020).  In response to this, poachers are poisoning vultures en masse.  Furthermore, livestock killed by predators are often laced with poison as a way of removing problem animals, however these carcasses are also eaten by vultures which subsequently die. Combined with an increasing trade in vulture parts for traditional medicine, many populations of vultures have declined in the Serengeti despite being one of the key breeding locations (Ol Karion gorge) in East Africa (Ogada et al., 2016). 

Human-wildlife conflicts are frequently recorded in adjacent villages, of which about half involved elephants (State Party of Tanzania, 2012). The State Party of Tanzania reported in 2014 that land use plans are being implemented in affected communities, research on best practice human-elephant conflict control approach through the use of chili fences has been implemented, and communities adjacent to the property are being assisted (State Party of Tanzania, 2014). Carnivores also attack cattle in the buffer zones (Blackburn et al., 2016; Lyamuya, 2017) and this is making some of the population hostile to the Park. It should be noted that many positive initiatives to engage and bring benefits to the local communities do exist however, such as introducing conservation-compatible livelihoods. In 2019 the Grumeti Fund built 35 km of electrified fencing on the northern border of the Ikorongo Game Reserve in western Serengeti in an attempt to reduce human-wildlife conflicts (primarily elephants), and to keep livestock out of the reserve as well as to ensure re-introduced rhino do not enter village lands. This is a controversial intervention as Tanzania has been consistently opposed to the use of fencing in protected areas. In addition, the fence poses some threat to migratory animals that may get stuck on the opposite side. The fence was erected in December 2019 so its long-term impacts have yet to be assessed.

The central part of Serengeti (around Seronera) appears to be overwhelmed by visitors (State Party of Tanzania, 2011), partly because tourism infrastructure and facilities are concentrated in this relatively small area. A recent survey by Serengeti National Park staff suggests there are now 3,766 beds available which is a 6-fold increase in the last decade.  This is comprised of 7 lodges, 9 permanent tented camps, 9 public campsites, 217 ‘premium’ and ‘special’ campsites, 4 rest houses and a youth hostel. The construction of the permanent (i.e. not a seasonal camp) Belabela Lodge in the Western Corridor in the Msabi-Kirwaira area is underway as well as Melila Lodge in the Mbalageti Valley. The EIA for Belabela Lodge (State Party of Tanzania, 2019) presents the location of the lodge in the high use zone, whereas previously the exact location was presented as a low use zone (i.e. where such permanent structures would not be permitted) according to the Project Brief. The 2014-2024 GMP has not been submitted to UNESCO but all sources report this GMP to have changed much of the zonations within Serengeti to give way to tourism development (IUCN Consultation, 2020), which creates clear discrepancies between the ecological priorities of the park and the pressure to build new tourist infrastructure. Further, there are already several tourists tented camps and bandas lodges nearby including Kirawira, Handajega, Serena Kirawa and Grumeti River, so the low use zones of the Western Corridor will likely become crowded in the near future. Finally, many of the sites for construction are in the middle of the migration route. 

The majority of the new lodges and tourist infrastructure occur in important wildlife areas (i.e. areas with water access or in areas where the forage quality is exceptional) (unpublished data, Serengeti Biodiversity Program). Perhaps the most noteworthy development is the recent construction of a large lodge adjacent to the Moru Rhino Protection Zone which stands to displace these secretive animals from the area. In addition, the sudden increase of tourism in the north along the Mara River (Kogatende) has led to an insurgence of permanent tented camps built on the river bank and extending the road network. These tourist facilities have recently been re-positioned away from the river bank to allow free access by migrating herds that cross the river en mass during the dry season. There are 9 airstrips in Serengeti, of which 6 are regularly used by tourist traffic. The Seronera airstrip has been upgraded and extended. The air traffic has increased significantly (i.e. more planes of larger size) raising concerns that the Tanzania Aviation Authority will need to establish an air traffic control to manage the demand. Overall the infrastructure such as water, toilets, food, waste, road maintenance and staff required to maintain the tourism demand is placing very large pressure on the ecosystem overall and is changing the way animals are distributed and move through the ecosystem (IUCN Consultation, 2020).

The park receives ca.160,000 non resident visitors per annum, (350,000 travel through, but mostly just crossing the park without visiting) (State Party of Tanzania, 2011), which would not be excessive if they were well distributed around the park (Larsen et al., 2020). However, visitors tend to concentrate around Seronera where there is an attractive visitor centre and other facilities. There are few opportunities for visitors to get out of their vehicles elsewhere, and the number of game viewing tracks is limited, so those that exist become quickly over-crowded. The SNP General Management Plan 2014-2024 (State Party of Tanzania, 2019) shows that 3,728 beds are needed to cater for increasing visitors. As of 2020 there are currently 3,766 beds. There are 217 seasonal campsites in the Park with a capacity of approximately 3,000 beds.

A recent analysis of 20 years of GPS collaring data from migratory wildebeest suggests the migration is spending less time in areas with a large tourist footprint, despite many of these areas having prime grazing conditions; this suggests that animals are being displaced and forgoing access to critical resources (Morrison, Torney and Hopcraft, in prep).  The most extreme case of this pattern is seen in the Masai Mara where tourist density is 10 times that of the Serengeti; the migration now spends about 35 days less per year compared to the previous decade, which amounts to over 25% of Kenya's prime wildebeest viewing time. Thus, more tourism pressure will likely threaten the Outstanding Universal Value of the Park. Nevertheless the State Party of Tanzania (2019) informed that the EIA's were certified through the Ministry of Environment Vice President Office and that the State Party will monitor and mitigate any potential adverse impact. How that would be done is not explained.  With increasing tourist numbers, increasing road kills and road erosion are being observed. Stricter speed limit enforcement is needed to avoid fatal collisions.
The road network used for game viewing is limited, and much of it is heavily used, especially around Seronera and more recently in the north along the Mara River. There are efforts to distribute tourism pressure more evenly in the Park, however recent changes that restrict multiple-entry in a single day on the same pass have seriously undermined the viability of tourist facilities on community land outside the park, while increasing the demand for tourist infrastructure inside the park.

Although the Serengeti ecosystem is adapted to fire, hot burns (which occur at the end of the dry season and are often lit by poachers to detract attention from their activities; other fires originate in the buffer zones and spread in the Park) can cause extensive long-term damage to woody vegetation (State Party of Tanzania, 2011) and they have been shown to have altered the savannah landscape within the Park (Probert et al., 2019).  Interestingly the number of fires around the edges of the National Park are significantly reducing primarily as a result of illegal grazing by surrounding communities that effectively removes the fuel load. In these areas, fire frequency is an accurate metric of the grazing pressure (Veldhuis et al., 2017).

Three invasive alien weed species are already established in the park (Argemone mexicana, Datura stromonium and Opuntia species) and two others are reported in neighbouring parts of the wider ecosystem. The highly aggressive and damaging weed Parthemium hysterophorus is already found in Ngorongoro Conservation Area and Masai Mara National Reserve, while Chromolaena odorata is in Grumeti Game Reserve (UNESCO and IUCN, 2010). Fire regulates the abundance of invasive species near roads and settlements (Bukombe et al., 2018) but there are insufficient data for areas away from roads and settlements; this study also shows that to effectively control invasive species other measures than just using fires are necessary, including physical removal and biological control agent. The role of cars in dispersing seeds including those of invasive alien species is well known, and therefore with increased visitor numbers and pressures to expand roads, the risk of spreading invasive plant species increases.  In addition, the recent expansion of two native invasive species (Bidens schimperi and Gutenbergia cordifolia) have been associated with changes in the grazing pressure that alters the competitive balance between grasses and forbs. It is likely that high grazing pressure as a result of livestock incursions and high densities of wild animals that were displaced from the edges of the ecosystem has removed grass and allowed these invasive forbs to rapidly expand across the range.

Disease transmission between domestic stock and wildlife can be catastrophic. The present population of wildebeest fluctuates inter-annually between 1.2 and 1.4 million (Hopcraft et al., 2015), and is the result of long-term recovery starting around 1960 from a rinderpest outbreak, transmitted by cattle, which had decimated the population. Wild dogs disappeared from the park in 1991, when a rabies outbreak killed three packs (UNEP-WCMC, 2011). Over 1000 lions, a third of the population, were killed in a canine distemper virus epidemic in 1993/94, and this virus also affected uncounted hyaenas, bat-eared foxes and leopards (Roelke-Parker et al., 1996). In addition, several gastrointestinal parasites are shared between livestock and wild grazers as well as Foot and Mouth Disease (FMD) primarily between buffalo and cattle. A livestock survey conducted by TAWIRI in 2017 established there were over 1,200,000 cattle within 20km (ie within 2 days walk) of the park boundary.  This is the first time in recorded history that the number of livestock is almost equivalent to the number of wildebeest.  The wildebeest alone remove approximately 4,400 tons of grass / day. If one adds to this the removal of grass by cattle, it is evident that increasing competition for grazing resources will likely lead to domestic and wild animals overlapping more in the near future.  Aside from the demographic repercussions of reducing the amount of food per capita, this increased rate of contact raises serious concerns about the potential for disease transmission. 

Human and cattle population densities are increasing very rapidly in all areas around the protected area complex, limiting wildlife migration routes in the buffer zones and exacerbating human-wildlife conflicts, and the effect is felt even within the Park for charismatic animals such as lions and cheetah (Blackburn et al., 2016; Durant et al., 2017; Snyder et al., 2019; Walelign et al., 2019). The human population on the perimeter of the Serengeti is growing at twice the national average which has led to rapid changes in land use (Estes et al., 2012; Veldhuis et al., 2019).  The border conflicts with agriculturalists is perhaps most evident on the Kenyan side of the ecosystem in the conservancies to the north of the Masai Mara.  In these areas approximately 50% of the landscape has been fenced with high tensile agricultural wire (often 11 strand and electrified).  This has led to the complete collapse of the Loita-Mara wildebeest migration from 140,000 animals in the mid 1990s to under 10,000 now (Ogutu et al., 2016) which once shared the dry season range with the Serengeti migration.  Currently the Serengeti migration encounters these fences along the northern edge of their dry season range.  These areas have the highest quality grass during the dry season; the continued agricultural fencing in the Mara conservancies could preclude the main migratory herds from accessing these resources as well as removing the prospect of generating revenue from selling tourism opportunities. The recent construction of 35 km of electrified conservation fence along the northern border of the Ikorongo game reserve (an area heavily used by the wildebeest and zebra migration as they transit from the Western Corridor towards the northern dry season refuge) has unknown consequences currently. Construction was completed in January 2020 and this year will be the first time the migration encounters it.  The fence is linear along the boundary (i.e. open at both ends) and therefore the risk is that groups of migratory animals get stuck on the opposite side in areas occupied by agropastoralists. There are plans to extend this fence for an additional 35 km to separate village grazing lands from conservation areas.