Predicting variation in plant N-uptake in three fields using soil organic matter, texture and Near Infrared Reflectance (NIR) spectroscopy
In: Precision Agriculture ‘05Search for other papers by J. Wetterlind in
Current site
Google Scholar
PubMed
Search for other papers by B. Stenberg in
Current site
Google Scholar
PubMed
Search for other papers by A. Jonsson in
Current site
Google Scholar
PubMed
Purchase instant access (PDF download and unlimited online access):
Adjustment of nitrogen fertiliser according to the within-field variation in N- mineralization during the growing season can reduce costs and decrease nutrient losses to the environment. The ability to predict that variation, particularly at low cost, is therefore of interest. In this study, SOM, texture and Near Infrared Reflectance (NIR)-spectroscopy were evaluated as tools for prediction of plant N-uptake using cross-validated PLS (Partial Least Squares) models. In 2 out of 3 fields, models based on NIR-spectra alone or in combination with SOM and texture gave predictions with r2 from 0.5 to 0.9 and average errors of cross-validation between 10 and 30 kg N ha-1. Including NIR did not improve the results in calibrations based on SOM and texture in the third field with very low r2 (<0.1). The results indicate that NIR and SOM plus clay are able to predict N-uptake with similar r2. Increasing the number of samples would be expected to improve the NIR-calibrations and reduce the need for separate clay and SOM measurements in addition to NIR.