International interest in the Eastern Arc Mountains and Coastal Forests hotspot has increased over the last three decades as the realization of its biodiversity importance and of the global crisis affecting tropical forests has deepened. Although descriptions of the wealth of biodiversity in the forests of the Eastern Arc Mountains date back to 1860 and there has been outstanding scientific work in the hotspot during the last 100 years, concerns for its conservation are relatively recent. Until about 30 years ago, nearly all the investment in the forests of the area had been in plantations, many of which were established after clearing indigenous forest.
The situation is now greatly changed and the last decade has seen a series of publications, workshops and conferences on the biodiversity and conservation of this hotspot (mostly organized by the United Nations Development Programme/Global Environment Facility (UNDP/GEF) and the WWF Eastern Africa Regional Programme Office (WWF-EARPO). These have produced a wealth of recent information on biodiversity issues (in particular on the distribution of endemic species across sites) and on forest status and management. This information has greatly reduced the time and effort needed to prepare this profile.
Current concerns for the conservation of the Eastern Arc Mountains date back to the 1978 Fourth East African Wildlife Symposium at Arusha. The conference was attended by 150 delegates, most of whom were not especially interested in forest conservation. However, a post-conference trip to Amani in the East Usambaras resulted in a report to the Government of Tanzania, drawing its attention to the biological importance of and threats to the Eastern Arc Mountains (Rodgers 1998).
In 1983, the Tanzania Forest Conservation Group (TFCG) was founded. In December 1997, there was a landmark international conference on the Eastern Arc Mountains at Morogoro, Tanzania attended by more than 250 delegates (Burgess et al. 1998a). During this conference, working groups reported on urgent issues such as the status of the remaining forest and participants presented papers on biodiversity, sociology and management. Much of the more recent conservation effort in the Eastern Arc Mountains dates from this conference, although one of the most important of these had already started with a UNDP/DANIDA project. This led in turn to a GEF Project Development Fund (PDF) Block A proposal and grant to characterize the conservation issues in the Eastern Arc Mountains in more detail.
The Block A process started after the December 1997 conference and included preliminary assessments of biodiversity values, conservation concerns, priority actions, financial constraints, sustainable financing opportunities, effectiveness of previous donor interventions and the development of preliminary proposals for GEF projects in the Eastern Arc Mountains. A three-way matrix was constructed showing levels of biodiversity and endemism, the degree of threat and the level and effectiveness of previous interventions. This enabled a ranking exercise that revealed that three of the main forest blocks (East Usambaras, Udzungwas and Ulugurus) were exceptionally diverse and that there was no major donor or public support for the Ulugurus. The Ulugurus, therefore, became a focus in the development of a PDF Block B proposal supported by UNDP and the World Bank. This PDF/B involved extensive stakeholder consultations and resulted in: 1) an outline and plan for a participatory and strategic approach to conservation and management in the Eastern Arc Mountains; 2) proposals for institutional reforms in the forest sector with a particular focus on facilitating participatory forest conservation and management; 3) a needs assessment for priority pilot interventions in the Ulugurus; and 4) the legal establishment of an Eastern Arc Mountains Endowment Fund (EAMCEF). The outcomes from this process were integrated into larger forest biodiversity concerns and into a proposed $62.2 million Tanzania Forest Conservation and Management Project.
During this time, awareness of the biodiversity values of the East African coastal forests had also grown. In 1983, a team from the International Council for Bird Preservation (ICBP, now BirdLife International) surveyed the avifauna of Arabuko-Sokoke Forest on the north coast of Kenya and drew attention to its globally threatened bird species (Kelsey & Langton, 1984). A detailed survey (Roberston, 1987) of the sacred Kaya Forests (conserved by the Mijikenda, a group of nine tribes on the Kenyan coast) highlighted their conservation importance for trees and led to a comprehensive survey of Kenyan coastal forests commissioned by WWF (Robertson & Luke 1993). This focussed on the plant species and on the status of the forests and made recommendations for their conservation.
The Frontier-Tanzania Coastal Forest Research Programme carried out a series of biodiversity surveys from 1989 to 1994 (Lowe & Clarke 2000; Clarke et al. 2000; Burgess et al. 2000; Broadley & Howell, 2000; Hoffman 2000). In 1993 a workshop on the East African coastal forests was held in Dar es Salaam. This raised the profile and conservation action in these forests and led to a series of status reports on the conservation and management of the Tanzanian coastal forests (Clarke 1995; Clarke & Dickenson 1995; Clarke & Stubblefield 1995). These and other studies are summarized in another landmark publication for the hotspot (Burgess & Clarke, 2000).
More recently, WWF-EARPO organised a series of workshops to develop an Eastern Africa Coastal Forest Programme covering Kenya, Tanzania and Mozambique (WWF-EARPO, 2002). Thirty-one scientists and stakeholders from these three countries attended a regional workshop in Nairobi in February 2002. It aimed at developing a regional synthesis on coastal forest resource issues and a vision, strategy and way forward for realising the coastal forest programme. There was a strong focus on country-based group work. Maps of the region were updated, threats and root causes were analyzed, country conservation targets were agreed on and preliminary logframe action plans were developed for each country. National Coastal Forest Task Force meetings in each of the three countries subsequently refined these action plans. The document resulting from the February 2002 workshop includes comprehensive annexes which list the coastal forest sites (showing their locations, areas, status, altitudes and threats) and the endemic animals, as well as the threat analysis and country action plans. A list of endemic plants, taken from Burgess & Clarke 2000, was supplied to the workshop but not included in the report.
On 12 March 2003, a CEPF workshop was held in Dar es Salaam to define the investment niche for CEPF, building on all the previous effort. Participants included 48 people from scientific and research institutions, government departments, NGOs, field projects and donor organizations, all of whom worked in or had knowledge of the hotspot. The outputs from the workshop were subsequently incorporated into a wide-ranging consultation process that helped to define the investment priorities for CEPF in this hotspot.
The Eastern Arc Mountains and Coastal Forests hotspot runs along the Tanzanian and Kenyan coasts from the border with Somalia to the north to that with Mozambique to the south (Figure 1 -
Download the full strategy to view all graphics, tables and appendices). The bulk of the hotspot is in its western expansion in Tanzania, which takes in the Eastern Arc Mountains and the water catchment system of the Rufiji River. There is a narrow hook-like extension of the hotspot near the Kenya/Tanzania border. This follows the Eastern Arc Mountains to their northernmost limits in the Taita Hills in Kenya. The hotspot also projects northwards for about 100 km in an extension that includes the forests of the Lower Tana River in Kenya. The hotspot includes the Indian Ocean islands of Mafia, Pemba and Zanzibar.
In terms of plant biogeography, the hotspot straddles two ecoregions: Eastern Arc Forest and Northern Zanzibar-Inhambane Coastal Forest Mosaic (WWF-US 2003a, b). These two ecoregions are mostly discontinuous but do meet in the lowlands of the East Usambara, Uluguru, Nguru and Udzungwa Mountains as well as in the Mahenge Plateau (WWF-US 2003a,b; Burgess pers. com.). A considerable proportion of species (e.g. nearly 60 percent of plants) are found in both ecoregions and the distinction between them has been a matter of some debate (Lovett et al. 2000). However, each of these forest types contains an impressive number of strict endemics. Lovett et al. (2000) conclude that the forests in these two ecoregions are very different, with differences in altitude and rainfall leading to a steep gradient of species replacement with elevation.
The Eastern Arc Mountains
The Eastern Arc Mountains stretch for some 900 km from the Makambako Gap, southwest of the Udzungwa Mountains in southern Tanzania to the Taita Hills in south-coastal Kenya (Figure 2) (Lovett & Wasser 1993; GEF 2002). They comprise a chain of 12 main mountain blocks: from south to north, Mahenge, Udzungwa, Rubeho, Uluguru, Ukaguru, North and South Nguru, Nguu, East Usambara, West Usambara, North Pare, South Pare and Taita Hills. The highest point (Kimhandu Peak in the Ulugurus) is more than 2,600 m in altitude, but most of the ranges peak between 2,200-2,500 m (GEF 2002; WWF-US 2003a). Geologically the mountains are formed mainly from Pre-Cambrian basement rocks uplifted about 100 million years ago (Griffiths 1993). Their proximity to the Indian Ocean ensures high rainfall (3,000 mm/year on the eastern slopes of the Ulugurus, falling to 600 mm/year in the western rain shadow) (GEF 2002). Climatic conditions are believed to have been more-or-less stable for at least the past 30 million years (Axelrod & Raven 1978). The high rainfall and long-term climatic stability, together with the fragmentation of the mountain blocks, have resulted in forests that are both ancient and biologically diverse.
The original forest cover (2,000 years ago) on the Eastern Arc Mountains is estimated at around 23,000 km², of which around 15,000 km2 remained by 1900 and a maximum of 5,340 km² remained by the mid-1990s (Newmark 1998; GEF 2002). At that time the Udzungwas contained the largest area of natural forest (1,960 km²), followed by the Nguru, Uluguru, Rubeho, East Usambaras, South Pare, West Usambaras, Mahenge, Ukaguru, North Pare and Taita Hills (6 km²). These and the following estimates of forest status and losses in the Eastern Arc Mountains are all taken from Newmark 1998. Losses were greatest, relative to original cover, in the Taitas (98 percent), Ukaguru (90 percent), Mahenge (89 percent) and West Usambaras (84 percent). The forests had become highly fragmented, with mean and median forest patch sizes estimated at 10 km² and 58 km², respectively. By 1994-96, the Udzungwas and the West Usambaras contained the largest numbers of patches (26 and 17) and only one mountain block (Ukaguru) had more or less continuous forest. At that time there were an estimated 94 forest patches in the Eastern Arc Mountains. Within forest patches there was considerable degradation. Of the closed forest that remained, only 27 percent had closed forest cover. With the exception of a few sites where there has been active intervention, the situation at present is far more likely to have deteriorated than improved since 1996.
The area defined by the Coastal Forests of Tanzania and Kenya in the hotspot includes the intervening habitats between the coastal forest patches. Although the main biodiversity values are concentrated in the forests there are a significant number of endemics (especially plants) in non-forested habitats. This part of the hotspot is therefore a mosaic, which stretches from the border of Kenya with Somalia, to the border of Tanzania with Mozambique, including the islands of Zanzibar, Mafia and Pemba. This part of the hotspot is, largely for practical reasons, partly defined by national boundaries; coastal forests in Somalia (very little left) and Mozambique (large areas) are poorly known and are excluded. Northern Mozambique could be included with further survey work. With the exception of Somalia, the mosaic, as defined here, corresponds to the WWF ecoregion known as the “Northern Zanzibar-Inhambane Coastal Forest Mosaic” (WWF-US 2003b). This falls within the “Zanzibar-Inhambane Regional Mosaic,” which is one of 18 distinct biogeographical regions that White (1983) recognized for Africa.
In Kenya, the Northern Zanzibar-Inhambane Coastal Forest Mosaic is mostly confined to a narrow coastal strip except along the Tana River where it extends inland to include the forests of the lower Tana River (the northern-most of which occur within the Tana Primate National Reserve) (Figures 1, 2). In Tanzania, the Mosaic runs from border to border along the coast, contracting in the Rufiji Delta region. There are also some outliers located up to ca. 300 km inland at the base of a few of the Eastern Arc Mountains (Udzungwa, Mahenge, Uluguru and Nguru) (WWF-US 2003a). Much of the Mosaic has been converted to subsistence agriculture, interrupted by plantations and human settlements, including the large cities of Mombasa and Dar es Salaam (populations of more than 700,000 and 3 million, respectively).
Geologically, the coastal forest strip has been subject to considerable tectonic activity and to sedimentation and erosion associated with movements of the shoreline (Clarke & Burgess 2000). Most coastal forests are found between 0-50 m and 300-500 m, although in Tanzania they occur up to 1040 m (Burgess et al. 2000). Rainfall ranges between 2000 mm/year (Pemba) to 500 mm/year (northern Kenya and southern Tanzania) (Clarke 2000). There are two rainy seasons (long, April-June; short, November-December) in the north, but only one (April-June) in the south. Dry seasons can be severe and El Niño effects dramatic. Climatic conditions are believed to have been relatively stable for the last 30 million years (Axelrod & Raven 1978), although variation from year to year can be considerable, leading to droughts or floods.
By the early 1990s, there were about 175 forest patches in the Coastal Forest Mosaic (Kenya 95, Tanzania 66) covering an area of 1,360 km² (Kenya 660 km², Tanzania 700 km²) (Burgess et al. 2000). Mean patch size was 6.7 km² in Kenya and 10.6 km² in Tanzania. Modal patch-size classes were 0 - 1 km² in Kenya and 5-15 km² in Tanzania. The two largest coastal forests are both in Kenya (Arabuko-Sokoke, minimum area 370 km²; Shimba, minimum area 63 km²) (WWF-EARPO 2002), while in Tanzania there are no coastal forests larger than 40 km² (WWF-US 2003b). There is some uncertainty with these figures because of differences in criteria for patch inclusion in the data set (e.g., the exclusion of all but a few small patches (<2 km²) from the Tanzanian data set and their full inclusion in the Kenya data set). The available information is also somewhat out of date and the current situation is, again, far more likely to have deteriorated than improved. No reliable estimates are available for the coastal forest with intact and contiguous canopies or for the extent of forest loss in recent history.
Previous: Introduction / Next: Biological Importance