Keywords
crop rotation, croprota, integrated modelling, uncertainty
Start Date
1-7-2012 12:00 AM
Abstract
Modelling systems are widely used for the assessment of agriculturalmanagement measures, e.g. to reduce nitrate loads as well as soil erosion or toavoid soil organic matter decline at different scales. However, modelling above theplot or farm scale resolution is challenged considerably by the limited availabilityand high uncertainty of bio-physical as well as management data, such as on croprotations. Generally, information on applied crop rotations is scarce and highlyaggregated in most cases (e.g. regional statistics). In this paper, we applied thecrop rotation model CropRota [Schönhart et al. 2011] to derive crop rotations for 12defined agricultural areas (240 – 3,200 km2) in Saxony (Germany) based on regionalstatistics. We compared model results to observed land use data as well asexpert-based crop rotations to proof the robustness and uncertainties related toCropRota as a tool to support integrated modelling studies.We found that the use of 10 -15 crop rotations is sufficient to i) realize a quality inCropRota output data comparable to an expert knowledge system and ii) minimizethe variations in output data because of different scenarios of intensity in crop cultivation.The number of crop rotations can be reduced with a better adaption of theused crop rotation table to regional circumstances and for regions with a lowerheterogeneity in crop cultivation.
The choice of crop rotations as an important model input – a case study from Saxony
Modelling systems are widely used for the assessment of agriculturalmanagement measures, e.g. to reduce nitrate loads as well as soil erosion or toavoid soil organic matter decline at different scales. However, modelling above theplot or farm scale resolution is challenged considerably by the limited availabilityand high uncertainty of bio-physical as well as management data, such as on croprotations. Generally, information on applied crop rotations is scarce and highlyaggregated in most cases (e.g. regional statistics). In this paper, we applied thecrop rotation model CropRota [Schönhart et al. 2011] to derive crop rotations for 12defined agricultural areas (240 – 3,200 km2) in Saxony (Germany) based on regionalstatistics. We compared model results to observed land use data as well asexpert-based crop rotations to proof the robustness and uncertainties related toCropRota as a tool to support integrated modelling studies.We found that the use of 10 -15 crop rotations is sufficient to i) realize a quality inCropRota output data comparable to an expert knowledge system and ii) minimizethe variations in output data because of different scenarios of intensity in crop cultivation.The number of crop rotations can be reduced with a better adaption of theused crop rotation table to regional circumstances and for regions with a lowerheterogeneity in crop cultivation.
