Development of mid-infrared spectroscopic methods for the determination of soil chemical quality in tillage fields

Healthy soils are the foundation for a prosperous future by providing nutrition, clean water, and the basis for all biodiversity on land. But soil health is threatened by soil degradation and climate change. To improve soil quality, a profound knowledge of its status and solutions for sustainable management are needed. In the last decades, precision agriculture (PA) has developed as a means of managing soils for optimum yield. Capturing this variability requires a high number of samples, which may not be feasible with conventional sampling methods and costly laboratory analyses. This thesis examined ways to analyse soil quality with mid-infrared spectroscopy (MIR) and zone-based sampling in field headlands to improve the understanding of these areas in tillage systems and capture the variability in soil chemical quality caused by management. To find answers to these questions the following tasks were examined: 1) The application of MIR as a rapid method of soil analysis for LR in tillage systems was examined and a partial least squares regression model (PLSR) (R² 0.87, RMSE 1.56 t/ha) resulted in values which can be used to substitute laboratory extractions for field level LR predictions. 2) The performance of a handheld MIR spectrometer for in-field sampling and soil analysis in a case study of lime requirement and organic matter was examined. A correction for the moisture of field fresh spectra showed improved results in the prediction of both LR and %OM. 3) An assessment of the variability of soil chemical quality parameters among different headland zones was taken, and elevated values of major nutrients (Phosphorus (P), Potassium (K)) were recorded around the main headland tramline. Analysis with a generalized linear mixed model resulted in statistically significant differences in P, K, lime requirement (LR) and pH between zones.

Funder

Walsh Scholarship

Publisher

NUI Galway

Rights

Attribution-NonCommercial-NoDerivs 3.0 Ireland

cbn

Collections

University of Galway Theses (PhD Theses)