PAN-European Soil Erosion Risk Assessment (PESERA)

Start year
Share on:
End year

Soil erosion poses severe limitations to sustainable agricultural land use, as it reduces on-farm soil productivity and causes the accumulation of sediments and agri-chemicals in waterways.

PESERA project

Pesera.jpgThe Pan-European Soil Erosion Risk Assessment Project is a fifth framework Research Technology and Development Project funded by the European Union. The PESERA Project resorts under the thematic programme Quality of Life, key action ‘support for common agricultural policy’. ISRIC is responsible for climatic and land use scenario analyses for the PESERA project. ISRIC will apply the PESERA model for a number of climatic and land use scenarios in different parts of Europe.

The main goal of PESERA is to develop, calibrate and validate a physically based and spatially distributed model to quantify soil erosion at a regional scale. Three  major objectives have been defined from this:

  1. To develop and calibrate a process-based and spatially distributed model to quantify soil erosion by water and assess its risk across Europe.
  2. To validate the developed PESERA model at low and high resolution and across different agro-ecological zones, and to compare the robustness and flexibility of the PESERA model output with other models or methods used for erosion risk assessment.
  3. To ensure the relevance of the PESERA model for policy makers.  

The project concentrated on promoting a robust and flexible model by demonstrating its performance at different resolutions and across agro-ecological zones, and by ensuring its relevance to policy makers through impact assessment and scenario analysis. Soil erosion indicators developed from a physically based model will not only provide information on the state of soil erosion at any given time, but also assist in understanding the links between different factors causing erosion. A strong expert and end-user network was established across Europe.

The PESERA project is planned in three phases with contributions from the project partners and is implemented by seven European partners.

PESERA project partners

PARTNER Logo Major responsibility
Laboratory for experimental Geomorphology

Katholieke Universiteit



Calibration and validation at the high resolution-Plant growth model and soil cover mudule
School of Geography

University of Leeds

Modelling Strategy and application at the European scale
Unité de Science du Sol

Institut National de la Recherche Agronomique


Validation at low resolutions, regional scale

European Soil Bureau, Environmental Institute Joint Research Centre-European Commision

Application at the European scale Website development
Agricultural University of Athens

End-user groups
Estacion Experimental de Zonas Aridas, Almeria Consejo Superior de Investigaciones Cientificas

Spatial and temporal resolution linkages
International Soil Reference and Information Centre

Scenario analysis


PESERA model and scenario analyses

The PESERA model is a process model for Europe-wide forecasting at a coarse scale. It estimates erosion rates in individual storms using a sediment transport equation that has explicit terms for topography, overland runoff and soil erodibility. Other soil characteristics and soil/land cover are implicitly incorporated as soil runoff threshold.

The PESERA model produces a quantitative forecast of soil erosion and crop growth, and therefore it has the potential to respond explicitly and rationally to changes in climate or land use, offering great promise for scenario  and analysis and impact assessment. ISRIC applies PESERA in different parts of Europe for a number of climatic and land use scenarios.


Project results

The general trend is that erosion overall decreases. Decrease is spectacular in the South window. In the North window erosion decreases in spring and summer, and in autumn and winter a mixed pattern of localised increases and decreases were observed. Little difference in erosion risk were found between maize on all arable land and dominant land use for the climate change scenario.

Animated gif

Influenced by the Gulf stream, the climate in western and central Europe has temperate characteristics, with mild summers and mild winters and a rather even rainfall distribution over the year. Going eastward there is a tendency towards a continental climate with warmer summers and colder winters. Southern Europe has a more Mediterranean climate characterised by warm, dry summers and cool, wet winters with high rainfall intensity. Besides the influence of the Gulf stream, the climate in Europe is determined essentially by the interactions among three pressure centres: the Icelandic Low, the Azores High, and Continental Highs, which predominate in winter, and lows, which generally are confined to summer months (IPCC, 1997). The European climate is highly variable from one place to another due to topographic features that have a significant influence on regional climate (plains, mountains, water bodies).

HADRM3 – A2B SRES climate scenario (2071-2080)

Mean annual temperature change HADRM3-A2b (Source LINK, 2003) and mean temperature change 2071, May, southern Spain and Portugal.

In the A2b scenario, temperature is rising by 3.5 to 4.5°C in Belgium and northern France, and 5-10 °C (May and June) -and in some parts even 10-15°C- in the southern Spain and Portugal. Broadly speaking, winters are getting wetter and summers are getting drier in the A2b scenario. Rainfall changes shows a strong seasonal pattern and the variation between years in the decade studied is considerable. But most of all, rainfall changes are characterised by a consistent spatial pattern. In the figure below annual rainfall over Europe (2073) is expressed as increase/decrease relative (%) to the base-line climate (MARS database).

The Hadley Centre for Climate Prediction and Research has developed climate scenarios that are based on these emission scenarios. The output of the Hadley Centre’s models for climate change prediction was used to feed the PESERA 1km grid model for erosion prediction. The model selected was the regional climate change model, HADRM3. The HADRM3 files were transferred to grid and then interpolated from a 50x50 to a 1x1 km grid.


Selected windows

Two windows were chosen to run PESERA-RDI_GRID on. The first window covers Belgium and small parts of southern UK, northern France, West-Germany, The Netherlands, and Luxembourg. The second window covers the southern part of Spain and Portugal.


PESERA-RDI_GRID coarse scale model

The PESERA-RDI_GRID coarse scale model is a physically-based model that estimates potential monthly erosion at a 1km grid resolution. PESERA predicts runoff with a daily time-step that is predicted from daily rainfall volumes, soil storage capacity, and vegetation interception. The model estimates ground cover, surface crusting, runoff and sediment transport, to give an estimate of water and sediment delivered to stream channels. The emphasis of the PESERA-RDI_GRID model is the prediction of hill-slope erosion, and the delivery of erosion products to the base of each hill-slope. Channel delivery processes and channel routing are explicitly not considered. More details on the model are provided on the PESERA webpages (PESERA model concepts, Kirkby, 2001), Irvine and Kirkby (2003) and Gobin and Govers (2003). See figure PESERA-RDI_GRID for an illustration of data flow.

The PESERA -RDI_GRID coarse scale model was run for the dominant Arable Crop (Actual PESERA runs), and all arable land under maize as a worst case scenario. See figure for dominant land use of the window areas. The data layers on actual and potential land use were produced by Gobin (2003).


Results South of Spain and Portugal window

The projected increase in winter rains in the HADRM3-A2B scenario for 2071-2080, does not lead to an increase in predicted erosion, on the contrary, erosion rates decline spectacularly. Only summer (July-August) shows increases in erosion rates. This is mainly a change of erosion pattern, or more an increase in the area for which erosion (risk) is predicted. Some areas have a an increase (relative to zero) of 100%, but at the absolute scale scenario erosion rates are low (0.5-1 t.ha-1).


Results Belgium and northern France window

The enhanced difference in erosion risk between summer and winter for the scenario runs follows the pattern of the expected change in rainfall for this window, with an increase in autumn and winter and a clear decrease in summer months. However, more detailed spatial variations are not explained by differences in rainfall and only to some extent by topography and/or land use. The difference between actual erosion under dominant crop or maize and the different years in scenario A2b show a strong relative decrease of erosion in spring and summer for the scenario in almost all years and a mixed picture in autumn and winter.



  • Gobin, A. and Govers, G.e., 2003. Pan--European Soil Erosion Risk Assessment. Third Annual Report. EU Fifth Framework Programme. Project no. QLK5-CT-1999-01323. Available at: 
  • Gobin, A., 2003. Agricultural Parameters for the PESERA model at the European scale. PESERA working paper, Laboratory for experimental Geomorphology, Katholieke Universiteit Leuven, Leuven.
  • IPCC, 1997. The Regional Impacts of Climate Change: An Assessment of Vulnerability. A Special Report of IPCC Working group II.
  • IPCC, Watson, R.T., Zinyowera, M.C. and Moss, R.H.E. (eds.). Cambridge University Press, Cambridge, pp. 527 pp. IPCC, 2000. Special Report on Emissions Scenarios, UNEP-WMO.
  • Irvine, B. and Kirkby, M., 2003. Mapping of European Soil Erosion: The PESERA- RDI model, 7th International Conference on GeoComputation, 8 - 10 September 2003, University of Southampton, United Kingdom.
  • LINK, 2003. The Climate Impacts LINK Project: Pioneering Climate Data Provision. Climate Data for the International Climate Change Research Community since 1991.
  • Mantel, S., Van Lynden, G.J. and Huting, J., 2003. PESERA, Work package 6: Scenario analysis. Final report: April 2003 - September 2003, ISRIC - World Soil Information, Wageningen.
Global issues