Development of near infrared (NIR) spectroscopic methods to predict carbon, nitrogen, silicon, phosphorus and potassium levels in mill by-products

Mill byproducts have been utilized on cane farms as a nutrient source and soil ameliorant for many years; however, the nutrient content is not usually measured on a regular basis. Concerns exist over the long term application of mill mud and ash without appropriate product monitoring, but these can be at least partially addressed by developing simple, rapid measurement techniques that can be employed at the sugar mill. This paper presents near IR (NIR) spectroscopic methods for the prediction of nutrient elements: carbon (C), nitrogen (N), silicon (Si), phosphorus (P) and potassium (K) levels in mill mud and mixed mill mud/ash byproducts. Eighty eight mill mud and mud/ash samples were obtained from three sugar mills in the Mackay region across two crushing seasons. Each sample was analyzed using std. chem. methods for C, N, Si, P and K and also had an NIR scan recorded on the wet sample prior to prepg. the sample for subsequent lab. anal. Partial Least Squares (PLS) regression models were constructed for each constituent. High R2 values were obtained for correlations between NIR predicted results and lab. results for all components, with values ranging from 0.890 to 0.962, together with very good std. errors of cross validation (SECV). This early work has confirmed that NIR calibrations can be developed for the measurement of nutrient elements in mill byproducts. The technique once implemented will provide benefits for both millers and growers and addnl. will address pressing environmental requirements.