CORDIS - EU research results

Avoiding the socio-ecological collapse of remnant evergreen forests in drylands: the case study of northern Kenya

Final Report Summary - ASEC-DRYLAND-FORESTS (Avoiding the socio-ecological collapse of remnant evergreen forests in drylands: the case study of northern Kenya)

Project context and objectives:
More than 1/3 of the Earth surface consists of drylands [1]. One in five people in the world live in already degraded or desertification-prone drylands; the world’s poorest and most marginalized citizens being among them [1]. Only in Africa, 325 million people live in drylands [1]. Drylands are characterized by a wide range of vegetation cover from semi-desert to moist evergreen remnant ‘oasis’ forests, typically found in mountains. These evergreen remnant forests in drylands provide essential livelihood resources to surrounding communities [2]. Despite their importance, such remnant forests remain over-exploited [3] and understudied. Predicted changes in climate [4] and current trends in population growth and development are likely to tester more pressure on the ecological functioning and ecosystem service delivery of these already fragile ecosystems. Therefore, there is an urgent need to assess the functioning and spatial use of these ‘forest oasis’ to develop better management strategies. In order to contribute towards a better understanding and an improved management of these forests the following research was carried out, with four principal research objectives: (1) to identify, measure and value the ecosystem services generated by these forests; (2) to reveal the inter-relationship between forests and local climate; (3) to determine if current use of important ES is sustainable; and (4) to develop management strategies which promote sustainable resource provision. This project focused on three remnant forests in dryland northern Kenya: Mt Marsabit, Mt Kulal and Mt Nyiro forests. The selected study sites provided key insights and information of wide relevance for the sustainable use of natural resources in dryland pastoralist communities both in Africa and in similar environments worldwide.

Work performed and main results achieved:
First, a participatory bottom-up approach was used for identifying and estimating the value of forest ecosystem services (ES). Twelve focus-group discussions were organised around each mountain (n=36) and 120 semi-structured interviews were carried out in different villages around Mt Marsabit, the mountain inhabited by several ethnic groups. Results indicated that water was always identified as the most important ES provided by the forest. However, the second most important ES differed between mountains and ethnic groups, being firewood for Borana in Mt Marsabit, fodder during droughts for Samburu in Mt Marsabit and medicine resources for Samburu in Mt Kulal and Mt Nyiro. Some ES were only mentioned by one ethnic group or in one location (e.g. place for hiding during conflict). Preferred plant species for different providing ES (food, fodder, medicine, poles and firewood) also differed between mountains and ethnic groups. As ethnicity, location and local taste preferences affect ES perception and importance ranking, this should be taken into account by decision-makers, e.g. restricted access and regulated extraction is likely to affect people differently.

Secondly, a novel approach combining remote-sensing, historical meteorological data and field measurements of rainfall and fog was used to study inter-relationships between these forests and local climate. Vegetation changes in the forest were studied using earth observation data (Landsat images) calibrated with >900 geo-referenced field observations. Measurements of carbon stocks were carried out in 24 permanent plots established in the three mountains following RAINFOR protocols. Historical meteorological data was gathered from Marsabit meteorological station, and field measurements of rainfall and fog in Mt Marsabit and Mt Kulal for the period of 15 months were collected using 15 rain gauges and fog collectors. Results indicated that between 1986 and 2014, Mt Nyiro forest lost 30% of its forest cover, Mt Kulal forest 20% and Mt Marsabit forest 8%. While in Mt Nyiro and Mt Kulal fire to produce fresh pasture was the main driver of forest loss, clearing land for cultivation was the main driver in Mt Marsabit. The different forests were found to have large amounts of live tree carbon stocks, with Mt Nyiro having particularly large carbon stocks related to large specimens of Podocarpus and Faurea saligna trees, some of which were >1m diameter and >30m high. Historical data revealed a significant decrease in rainfall and fog during the past three decades. Field measurements indicated that fog was an important source of water input in these ecosystems, and that important variation in fog and rainfall within one mountain could be observed.

Thirdly, to determine if current use of important ES was sustainable 150 water samples of different streams, boreholes, wells, rain and fog were collected; and the abundance, population structure and regeneration of important plant species for provisioning ES was determined using the 24 permanent vegetation plots established. Results indicate that different aquifers exist in each mountain, and that the water being extracted from boreholes is from ancient aquifers not being replenished by rainfall or fog, highlighting the long-term unsustainability of this extraction. With regard to plant species, main findings indicate that non-timber forest product harvesting is not necessarily destructive for the species concerned; species’ intrinsic characteristics, number of uses and the timing of the harvest are of key importance. However, the extraction was found to be unsustainable for certain plant species, especially for Olea europaea and Rinorea convallaroides in Mt Marsabit; and effective conservation plans are needed for these species.

Finally, another round of focus-group discussions was organised around each mountain (n=36) to show project findings and discuss plausible forest management alternatives, including the potential establishment of a carbon project. Semi-structured interviews were also conducted with key stakeholders and project partners. Both communities and key stakeholders stated that a greater involvement of communities in forest management decisions was necessary. Communities were aware of forest degradation and highlighted the need to help them find alternative livelihood strategies to reduce pressure on forests. Enrichment planting of key canopy species was discussed as a potential intervention. The establishment of a carbon project could be an option, especially in Mt Nyiro and Mt Kulal, which have higher forest carbon stocks. This research produced five publications in international peer-reviewed journals (two more are in preparation) and three oral presentations in two international conferences: Association for Tropical Biology and Conservation (ATBC) meeting in Hawaii in July 2015 and ATBC meeting in Montpellier in June 2016.

The expected final results and their potential impact and use:
This innovative multidisciplinary research has not only significantly advanced scientific understanding the functioning and spatial use of evergreen montane forests in drylands but also tested some novel approaches at the interface between natural and social sciences, which can be used by other researchers in the future. This improved understanding has provided evidence for more informed management decisions, with regard to the three forests studied in northern Kenya, and other remnant forests in Kenya (e.g. project findings were shared and discussed with Kenya Forest Service and Kenya Water Towers Agency). We hope that our findings can also help management decisions outside Kenya. Local communities in the study area also benefited from participating in this study, as information from other ethnic groups and mountains was shared with them, together with project findings. They felt they understood better what was happening to their forests and also, they had their voice heard at higher management levels, as they contributed to the final report which was shared with regional managers. This project also promoted knowledge transfer between leading European and African research institutions and capacity building: the lead researcher taught in two courses (one in Nairobi, one in Copenhagen), and several students, from Africa and Europe, participated in the fieldwork campaigns, learning from the lead researcher and her collaborators. Raising awareness about these biodiversity rich unique ecosystems; about the people that live there (semi-nomadic pastoralists with traditional lifestyles with little formal education but great knowledge about their environment); and about this type of multidisciplinary research was another main objective of the project. Several outreach activities were organised, from environmental education programs in primary schools, to public lectures, Facebook posts and online short films. We can only protect what we know exists, and raising awareness was an important objective of this project. We believe this successful project has already benefited a number of people: researchers, managers, students, local communities and the general public in Africa and Europe, and we hope it will continue to do so as two more publications come out, the lead researcher attends the ATBC meeting in Montpellier in June 2016 and a longer film about the project is made in the coming months and is presented in two film festivals.

(1) UNDC and CFC 2009. African drylands commodity atlas. 2 UNDP 2004. Sharing innovative experiences. 3 Miles et al. 2006. Journal of Biogeography 33:491-505. (4) Boko et al. 2007. Africa. Cambridge University Press, Cambridge, 717–743.