Analysis Of Anthropogenic Influence On Pollen Based Climate Research

In paleoclimatology the relationships between vegetation and climate are used to infer past climate from plant fossils. However, if a transfer function is tested on modern pollen and climate observations, there is a mismatch between observed and inferred climate. Several causes for such a mismatch have been identified by literature, but this research focusses on humans. Humans influence community assembly of vegetation, but the effect of this influence on climatic studies is not well investigated. The aim of this research was to assess anthropogenic influence on climatic mismatch. Data was collected from previous research on 29 Amazonian sites. This was run through a transfer function and mismatch was calculated for the mean annual temperature, mean diurnal range and annual precipitation. Linear regression tests were used to test the data on a correlation between human influence (land use and distances to features) and mismatch. Mismatch decreased at first and then increased in the opposite direction. There is evidence that precipitation is affected by humans. Of all considered measures the distance to the nearest hardened road had the strongest relationship with induced mismatch. This file contains all used data and R scripts. With the contents the results can be recreated. Instructions on use are enclosed