Understanding the impact of human activities on the environment since the introduction of settled agriculture in the Neolithic requires an understanding of both natural changes in land cover and human induced changes in land use. We are currently working on reconstructing both past land cover and land use during the Holocene using mainly pollen data, which represents by far the largest and most readily available source of information for the pre-industrial period. Two significant problems in using this kind of data is to understand the relationship between the pollen proxy and actual vegetation (both natural and human influenced), as well as inferring vegetation changes across the land surface that lies between pollen sites. We have been developing a series of modeling and calibration approaches to address these problems in order to reconstruct past changes in land cover and land use at continental scales throughout the Holocene (see image). A direct application of this work is to help evaluate global models of human land use (see other projects), as well as provide information on the terrestrial biosphere for Earth system modeling.
Image: coming ; [Holocene_landcover_image.png]
Image-caption: Here we reconstruct the modern forest cover of Europe (shown left, based on remote sensing data), using a network of modern pollen sites. Upper right: traditional approaches use uncalibrated measures such as the percentage arboreal pollen, but this causes significant errors in some locations such as northern Scandinavia. Lower right: using a modern-analogue method we are able to calibrate the pollen record to provide a more accurate reconstruction of forest cover using pollen data that can also be applied to the fossil record for the whole Holocene.