A data-driven reconstruction of historic land cover/use change of Europe for the period 1900 to 2010

Fuchs, R.



The population in Europe almost has doubled within just a little more than 100 years. The related need for food, fibre, water, and shelter led to a tremendous reorganization of the European landscape and its use. These land cover/use changes have far-reaching consequences for many ecosystem processes that directly or indirectly drive the climate on continental and global scale. Different types of land changes lead to different changes in carbon pools. Examples are rapid carbon pool changes due to deforestation or a delayed carbon pool change from long-term uptake of carbon in re-/afforested areas. This time lag of greenhouse gas fluxes requires the consideration of present and past land use change dynamics. To assess the fluxes of present and past land use change dynamics data or model-based reconstructions of historic land cover/use are needed. Historic land cover/use data as input for historic land reconstructions are fragmented, hard to obtain (copyright, secrecy statuses, accessibility, language barriers), difficult to harmonize and to compare. This lack of available data limits historic land change assessments, especially on large scales. Many continental to global historic land cover/use reconstructions provide little detail of change dynamics, have a rather coarse spatial resolution and reconstruct only a few land cover/use classes. Furthermore, most of them consider only the net area difference between two time steps (net changes) instead of accounting for all area gains and losses (gross changes), which leads to serious underestimation of the amount of area subject to change.

This research aimed to reconstruct historic European land cover/use and its changes for the period from 1900 to 2010 addressing some of the shortcomings of previous studies. The main objective of this thesis was to explore new reconstruction methods that improve the spatial and temporal detail and reduce the uncertainty in the estimates at continental level by better using available data sources. The use of available historic data sets as input data for the reconstruction was evaluated. The main objective was achieved by providing a full representation of gross land changes at continental scale in order to capture all major land change processes and their dynamics for Europe throughout the last century. The thesis also explored the implications of those change dynamics on environmental and biogeochemical research, such as climate change research.

In chapter 2 the combination of different data sources, more detailed modelling techniques and the integration of land conversion types was investigated to create accurate, high resolution historic land change data for Europe suited for the needs of greenhouse gas and climate assessments. A method was presented to process historic net land changes consistently on a 1 km spatial resolution for five IPCC land categories (settlement, cropland, grassland, forest and other land) back to the year

1950 for the EU27 plus Switzerland. Existing harmonized land cover/use change data from census data and from remote sensing were intensively used to feed into the reconstruction.

Chapter 3 analysed how historic statistics of encyclopaedias and old topographic maps can improve the accuracy and representation of land cover/use and its changes in historic reconstructions. This study made use of historic statistics and old topographic maps to demonstrate the added value for model-based reconstructions of historic land cover/use for Central Europe back to 1900. The added value was evaluated by performing a reconstruction with and without the historic information. The study showed that a data driven reconstruction for historic land cover/use improved the modelling accuracy in comparison to a traditional model-based reconstruction approach that more strongly relies on assumptions and proxy variables for the spatial allocation and land change trends.

Chapter 4 explored to what extent historic land cover/use reconstructions underestimate land cover/use changes in Europe for the 1900–2010 period by accounting for net changes only. Available historic land-change data were empirically analysed for differences in quantities between gross and net changes. The empirical results of gross change quantities were applied in a spatially explicit reconstruction of historic land change to reconstruct gross changes for Europe back to 1900. Besides, a land-change reconstruction that only accounted for net changes for comparison was created. The two model outputs were compared with five commonly used global reconstructions for the same period and area. The gross change reconstruction led in total to twice the area change of net changes. All global reconstructions used for comparison estimated fewer changes than the gross change reconstruction.

Chapter 5 investigated to what extent historic gross land changes lead to differences in continental carbon flux estimations compared to net land changes. Historic changes of carbon in soils and vegetation in Europe for the period 1950 to 2010 were assessed, while accounting for legacy effects and gross change dynamics with decadal time steps at 1 km spatial resolution. A net land change assessment was performed for comparison to analyse the implications using gross land change data. For areas that were in both reconstructions subject to land changes (35% of total area) the differences in carbon fluxes were about 68%, and highest over forested areas. Overall for Europe the difference between accounting for either gross or net land changes led to 7% difference (up to 11% per decade) in carbon fluxes and systematically higher fluxes for gross land change data as compared to net land change data.

The research conducted in this thesis contributes to the improvement on historic land cover/use reconstructions and gives a harmonized, consistent ‘bigger picture’ of Europe’s land history with high spatial resolution.