Newsletter No 4 December 2012

Past as key to the Future



As long as humans live on the Earth, they have altered the global landscape, through hunting, gathering and agriculture. Kees Klein Goldewijk describes in his thesis a database of historic land use: HYDE. Both the dynamics of global population and the enormous expansion of agricultural land are estimated for the past 12 000 years, roughly the period since the emergence of (sedentary) agriculture.

Changes in land use, primarily through conversion of undisturbed ecosystems to other forms of land use (such as deforestation for agriculture or grazing, logging activities, or infrastructure such as roads) result in greenhouse gas emissions and contribute also considerably to the cumulative carbon dioxide (CO2) increase in the atmosphere. These carbon fluxes are still not well quantified in the global carbon budget, and uncertainties quantifying the impact of land use changes on the global carbon cycle lead to uncertainties in projections of atmospheric CO2 and climate, and consequently, affect policy makers in establishing reasonable emission mitigation strategies. To assess the effects of land cover changes on the climate system, models are required which are capable of simulating interactions between the involved components of the Earth system (land, atmosphere, ocean, and carbon cycle). Since driving forces for global environmental change differ among regions, a geographically (spatially) explicit modeling approach is called for, so that it can be incorporated in global and regional (climate and/or biophysical) change models in order to enhance our understanding of the underlying processes and thus improving future projections.

There are two different approaches concerning global historical land-use reconstructions; either by (dynamic) modeling or through combining statistical inventories with hind-casting techniques. The first group is represented by so-called Dynamic Global Vegetation Models (DGVMs), who are process based and look at the exchange of carbon, water, and vegetation dynamics to compute long historical transient time series of land use. Since most of these DGVM’s use biomes (envelopes of plant functional types), only natural vegetation patterns are modeled who lack human presence.

The second group of historical land-use datasets uses a mix of statistical data and hind-casting techniques, and they exist at different levels of detail, temporarily as well as spatially. Depending on the spatial and temporal scale, studies either rely completely on historical statistical data, or where data are limited or non-existing, hind-casting techniques and/or modeling techniques are applied. However, such global long term land use inventories are very rare. Therefore, the main objective of this thesis is to present a globally spatially explicit database of historical populations and land use for the Holocene, useful for the scientific community especially for integrated modeling of global (climate) change. The current HYDE version 3.1 is an updated and internally consistent combination of historical population estimates and improved allocation algorithms with time-dependent weighting maps for cropland and grassland, while the period covered now is extended from 10 000 BCE to 2000 CE.

Results indicate that total global population increased from 2 to 6145 million people over that time span, resulting in a global population density increase of < 0.1 cap/km2 to almost 46 cap/km2 and a urban built-up area evolving from almost zero to 0.5 million km2 (still only <0.5% of the total global land surface, but having tremendous impact through the demand of food, services, building materials, etc. Cropland occupied roughly less than 1% of the global ice-free land area for a long time period until 1000 CE, quite similar like the area used for pasture. In the centuries that followed the share of global cropland increased to 2% in 1700 CE (ca. 3 million km2), and 11% in 2000 CE (15 million km2), while the share of pasture area grew from 2% in 1700 CE to 24% in 2000 CE (34 million km2).

space 02

In order to demonstrate the potential use of this database two applications are described. One is a novel representation of land cover based on the idea that classical biomes (ecosystems defined by specific climate and soil characteristics) insufficiently represent the dominant role of human presence as determinant of land cover. As an alternative, anthropogenic biomes were introduced, or "anthromes". Reconstructions show that three centuries ago people already influenced almost every place on Earth, leaving virtually no undisturbed wilderness.

A second application concerns the impact of human-induced land-use change on terrestrial carbon storage in the pre-industrial Holocene. The simulated cumulative anthropogenic emissions support the hypothesis that human activities have led to changes in atmospheric CO2 concentrations at a global level. The thesis indicates that this hypothesis is very sensitive to the underlying assumptions of the land use reconstruction, in particular the magnitude of the historical land use per capita.