Are coffee agroforestry systems suitable for circa situm conservation of indigenous trees? A case study from Central Kenya

• Published in: Biodiversity and conservation 2014-02, Vol.23 (2), p.467-495
• Main Authors: Pinard, F, Joetzjer, E, Kindt, R, Kehlenbeck, K
• Format: Article
• Language: English
• Subjects: Agricultural and farming systems; Agroforestry; Agronomy. Soil science and plant productions; Analysis; Animal and plant ecology; Animal, plant and microbial ecology; Applied ecology; Biodiversity; Biological and medical sciences; Biomedical and Life Sciences; Case studies; Climate Change/Climate Change Impacts; Cluster analysis; Coffee; Coffee industry; Conservation Biology/Ecology; Conservation, protection and management of environment and wildlife; Ecology; farm size; Farms; Fences; forest trees; forests; Fundamental and applied biological sciences. Psychology; gardens; Gardens & gardening; General agroecology. Agricultural and farming systems. Agricultural development. Rural area planning. Landscaping; General agronomy. Plant production; General aspects; Generalities. Agricultural and farming systems. Agricultural development; Heterogeneity; Indigenous species; Life Sciences; live fences; Mountains; Native species; Natural vegetation; Original Paper; Plant diversity; Plant species; Protection and preservation; Seedlings; Small farms; Species diversity; Species richness; Synecology; Trees; Vegetation; volunteers
• ISSN: 0960-3115; 1572-9710
• DOI: 10.1007/s10531-013-0615-0

Coffee agroforestry systems (CAFS) are often considered to be species-rich, potentially contributing to the conservation of indigenous trees. To investigate the conservation capacity of a Kenyan CAFS, all tree species on 62 smallholder coffee farms (covering 39 ha in total) in the Aberdare Mountains of Central Kenya were recorded. In total, 6,642 trees of 59 species were enumerated, with a mean density of 256 trees per ha and a mean species richness of 11.2 species per farm. Indigenous species represented 63 % of the richness but only 31 % of the abundance. For individual farms, as expected, farm size had a positive correlation with tree species richness, but more interestingly there was a negative correlation with tree density. Cluster analysis based on densities of the 18 most important species (defined by an importance value index) revealed two groups of farms: one cluster represented small farms (mean size = 0.4 ha) with high tree species diversity and individual density, particularly of indigenous trees; the other cluster represented large (mean size = 1 ha) and less diverse farms with low tree densities, particularly regarding indigenous species. Tree individuals were unevenly distributed within farms, being more frequent in living fences (38 % of all individuals), the garden zone (20 %) and in coffee plots (18 %). The relative occurrence of indigenous species was also uneven, being greater in living fences and the garden zone. Most adult trees (83 %) were planted, but only 46 % of seedlings were, revealing the active removal of volunteer seedlings by farmers as trees mature. Surveyed coffee farms harboured 20 % of the 135 tree species of the potential natural vegetation for the region, but only 3.6 % of the on-farm tree individuals belonged to the most valuable types of dominant and forest vegetation. Thus, although a source of significant tree cover and heterogeneity at landscape level, the value of these CAFS as circa situm reservoirs of forest tree species is questionable.