Biosystems Engineering (2005) 92 (4), 419–428 doi:10.1016/j.biosystemseng.2005.08.010 PA—Precision Agriculture Management Zones based on Correlation between Soil Compaction, Yield and Crop Data E. Vrindts; A.M. Mouazen; M. Reyniers; K. Maertens; M.R. Maleki; H. Ramon; J. De Baerdemaeker Laboratory of Agro-Machinery and Processing, Katholieke Universiteit Leuven, 3001 Leuven, Belgium; e-mail of corresponding author: abdul.mouazen@biw.kuleuven.be (Received 17 November 2004; accepted in revised form 19 August 2005; published online 10 October 2005) In 2003, soil and winter wheat crop information were gathered on an Arenic Cambisol field dominated by sandy silt loam and sandy loam soils according to the Belgium soil classification. Crop information consisted of spectral reflection measurements taken in May and yield mapping in August. Soil information, obtained after harvest, included moisture content and dry bulk density. The soil and crop information of this field were analysed to assess the effect of the measured soil dry bulk density on the crop and to test if the measured soil bulk density could contribute to better field management. Where dry soil bulk density was above 16Mg/m 3 , the yield was limited, otherwise, no relation was observed between the yield and dry soil bulk density. Different methods of defining management zones based on soil and crop information were compared. Fuzzy clustering was used to make soil management zones based on the soil data only, but these clusters did not explain much of the yield variability. Another set of management zones was defined based on soil information and crop. These clusters provided a better description of the yield variation. A low-yielding area could be delineated related to high soil bulk density. Outside this area, yield was more affected by soil moisture variability and other factors. r 2005 Silsoe Research Institute. All rights reserved Published by Elsevier Ltd 1. Introduction The most popular approach to manage field varia- bility is the use of management zones. Each zone gets the appropriate level of inputs (seed rate, soil tillage, fertiliser rate, crop protection). A ‘management zone’ in a field expresses a sub-region of a field that has a relatively homogeneous combination of yield-limiting factors, for which a single rate of a specific crop input is appropriate. To be successful, true cause and effect relationships must exist between site characteristics and crop yield. The management zones are usually defined on the basis of soil and yield information, possibly over several years (Fraisse et al., 2001; Fleming et al., 2000). Soil information (topography, soil type,yetc.) is valu- able for management and can be used to create ‘stable’ management zones that remain the same per field. However, crops respond to more than soil type (for example available water or local incidence of weeds, pests and disease), and yield patterns can vary from year to year (Blackmore et al., 2003; Vrindts et al., 2003; Taylor et al., 2001). Current information on crop status, for example by remote sensing, can be considered a valuable tool which enables the management zones to be adjusted to the current growing season (Godwin et al., 2003; Reyniers, 2003). Ideally, the spatial variability of significant parameters inducing yield variability can be summarised in a small number of zones in the field, so that the farmer can effectively manage each zone. Within these zones, the significant parameters have a similar effect on the crop, leading to a similar crop response, and thus requiring the same input rates. Clustering soil and crop data can be used as a basis for the definition of management zones (Taylor et al., 2003), because the data are grouped into clusters based on the similar interaction between the soil and crop data. The clusters can also provide a starting point to analyse the causes for yield variability (Reyniers, 2003). ARTICLE IN PRESS 1537-5110/$30.00 419 r 2005 Silsoe Research Institute. All rights reserved Published by Elsevier Ltd