Data from: Land use legacy effects on woody vegetation in agricultural landscapes of southwestern Ethiopia

Aim: Past land use legacy effects – extinction debts and immigration credits – might be particularly pronounced in regions characterized by complex and dynamic landscape change. The aim of this study was to evaluate how current woody plant species distribution, composition and richness related to historical and present land uses. Location: A smallholder farming landscape in southwestern Ethiopia. Methods: We surveyed woody plants in 72 randomly selected 1 ha sites in farmland, and grouped them into forest specialist, generalist and pioneer species. First, we investigated woody plant composition and distribution using non-metric multidimensional scaling. Second, we modelled species richness in response to historical and current distance from the forest edge. Third, we examined diameter class distributions of trees in recently converted versus permanent farmland. Results: Historical distance was a primary driver of woody plant composition and distribution. Generalist and pioneer species richness increased with historical distance. Forest specialists, however, did not respond to historical distance. Only few old individuals of forest specialist species remained in both recently converted and permanent farmlands. Main conclusions: Our findings suggest that any possible extinction debt for forest specialist species in farmland at the landscape scale was rapidly paid off, possibly because farmers cleared large remnant trees. In contrast, we found substantial evidence of immigration credits in farmland for generalist and pioneer species. This suggests that long-established farmland may have unrecognised conservation values, though apparently not for forest specialist species. We suggest that conservation policies in southwestern Ethiopia should recognize not only forests, but also the complementary value of the agricultural mosaic – similarly to the case of European cultural landscapes. A possible future priority could be to better reintegrate forest species in the farmland mosaic. Data for land use legacy effects on woody vegetationThree zip-archived, comma delimited csv files with (1) DBH data, (2) species richness data, and (3) species composition data. For (1), columns include site ID, short name of tree species (composed of first three letters of genus name and first three letters of tree species name), tree DBH (in cm), land use history (i.e. permanent or converted farmland), and tree species type (i.e. forest specialist, generalist, or pioneer species). For (2), columns include kebele ID (GBD = Borcho Deka, GKH = Kela Harari, SDM = Difo Mani, SGB = Gido Bere, TBW = Bere Weranigo, and TKF = Kuda Kofi), site ID, land use type, total species richness, total species abundance, forest specialist species richness, forest specialist species abundance, pioneer species richness, pioneer species abundance, generalist species richness, generalist species abundance, proportion of vegetation cover within the site (in %), wetness index, historical (1973) distance (in m), current (2015) distance (in m), land use history (permanent or converted farmland), and altitude (in m above sea level). (3) is a site-by-species abundance matrix with site IDs and species represented by their short name (composed of first three letters of genus name and first three letters of species name).