Gomes, V.H.F., Mayle, F.E., Gosling, W.D., Vieira, I.C.G., Salomão, R.P. & ter Steege, H. (2020) Modelling the distribution of Amazonian tree species in response to long-term climate change during the Mid-Late Holocene. Journal of Biogeography DOI: 10.1111/jbi.13833
Valencia Castillo, B.G. (2014) From glacial to modern conditions: Vegetation and climate change under human influence in the Central Andes. PhD Thesis, Department of Environment, Earth & Ecosystems, The Open University.
Conservation, restoration and management strategies are employed to maintain Earth’s biological diversity and physical environment to a near “natural” state. However, the concept of “natural” is generally inexact and may include degraded landscapes. In absence of long-term empirical data of natural baselines, impacted assemblages (human altered baselines) could be falsely assumed to be natural and set as conservation or restoration goals. Therefore, the identification of long-term ecological baselines becomes a pressing requirement especially in threatened biodiversity hotspots such as the tropical Andes that were under human pressure for several millennial.
This thesis aims to identify ecological baselines for tropical Andean ecosystems based on multi-proxy palaeoecological reconstructions from three Andean lakes. Trends of vegetation change are used to identify when landscapes became anthropogenic in the Andes. Because vegetation assemblages at c. 10 ka experienced negligible anthropogenic impacts and had modern-like climate condition, this time was considered the most recent period likely to provide insight into natural ecological baseline conditions.
Changes in vegetation assemblages were evaluated over time departing from 10 ka around Miski and Huamanmarca, two sites that remained virtually impervious to human impacts. Baselines in Miski and Huamanmarca drifted continuously over time and showed that baselines are dynamic entities. The vegetation assemblages derived from Miski and Huamanmarca suggest that that human impact was not homogeneous throughout the Andean landscape.
Once baselines were defined it was possible to evaluate if the spatial distribution of Andean woodlands represented by Polylepis was a product of human impacts. A MaxEnt model generated based on 22 modern environmental variables and 13 palaeoecological vegetation reconstructions showed that Polylepis woodlands were naturally fragmented before humans arrived in South America (14 ka). However, the influence of humans during the mid and late Holocene enhanced the patchiness of the forest generating a hyper-fragmented landscape.