Relationships between Soil-Based Management Zones and Canopy Sensing for Corn Nitrogen Management

Published in Agron. J. 104:119–129 (2012) Posted online 1 Dec 2011 doi:10.2134/agronj2011.0044 Copyright © 2012 by the American Society of Agronomy, 5585 Guilford Road, Madison, WI 53711. All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. N management in cereal crops has been the subject of considerable research and debate for several decades. Inefficient N management practices have contributed to low (~30–40%) N use efficiency (NUE) estimates for cereal crops such as corn (Raun and Johnson, 1999; Cassman et al., 2002). Contributing factors to low NUE abound but can ultimately be summarized in three main points, as stated by Shanahan et al. (2008): (i) poor synchrony between soil N supply and crop demand; (ii) uniform application rates of fertilizer N to spatially variable landscapes; and (iii) failure to account for temporally variable influences on crop N needs. Poor synchronization between soil N supply and crop demand is the result of N application before crop establishment and failure to account for N mineralization, leaving inorganic N in the soil subject to denitrification, leaching, or volatilization. Previous studies have found that in-season N application resulted in higher NUE than preplant applied N (Welch et al., 1971; Randall et al., 2003a,b). Studies have also shown that optimal N rates vary spatially across a field (Mamo et al., 2003; Scharf et al., 2005; Shahandeh et al., 2005) and using tools to account for this variability could potentially increase NUE (Hong et al., 2007). Therefore, innovative N management strategies are needed to address the factors that cause relatively low NUE. Soil-based methods to increase NUE have included the concept of management zones (MZ). Management zones are defined as subregions of a field with relatively homogeneous attributes in landscape and soil conditions, resulting in similar yield-limiting factors or yield potential (Doerge, 1999). Implied is that these MZ would have similar input-use efficiencies. Delineating MZ has included mathematical assessment of quantitative data sets to determine groups or clusters of data that are most similar. Methods for performing this data clustering to create MZ have varied considerably, with no universal algorithm being widely accepted. As reviewed by Fridgen et al. (2004), methods include supervised clustering, unsupervised clustering, c-means (k-means), fuzzy c-means (fuzzy k-means), and others. Management Zone Analyst (MZA, USDA-ARS Cropping Systems and Water Quality Research Unit, Columbia, MO) is a software program developed using Microsoft Visual Basic (Microsoft Corp., Redmond, WA) that uses a fuzzy c-means algorithm. The advantage of MZA over other software programs is that it provides concurrent output for a range of cluster numbers, so the user can define the optimum number of MZ (Fridgen et al., 2004). A variety of data types have been used to delineate MZ within fields. These have included, but are not limited to: soil survey maps (Franzen et al., 2002); modified soil survey maps (Carr et al., 1991); topography (Kravchenko and Bullock, 2000); remote sensing and farmer experience (Fleming et al., 2000); ECa (Fraisse et al., 2001; Kitchen et al., 2005); ECa, grain yield, or slope and texture (Ferguson et al., 2003); yield ABSTRACT