Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Abstract. Proper assessment of the nutritional status of a plant is crucial to supply the necessary amount of nutrients. Tissue testing, despite being the common assessment method, has drawbacks for being laborious, time-consuming, and destructive. Several studies tackled deploying non-destructive sensors to provide information on nutrient contents based on the leaf chemical contents as the reference data. However, the current practice by potato growers for tissue testing is based on petiole chemical analysis rather than leaf analysis. Thus, this study aimed to estimate nutrients in potato intact fresh leaves based on the petiole chemical testing. A total of 45 datapoints were collected from the outdoor farming of Russet Burbank variety. The spectral measurements were performed on leaves using a portable Vis-NIR spectrophotometer and the chemical analysis were done on petioles following the official methods of the Association of Official Analytical Chemists. Datasets were created between the chemical contents of the petioles and the leaves spectrum based on a linear relationship. Lasso multiple linear regression was selected for its features to identify the most informative, and the least redundant wavelengths. Performance of the generated model was evaluated using the Ratio of standard error of prediction to standard deviation (RPD). The Lasso models showed a fair distribution of nitrogen, phosphorus, potassium, calcium, magnesium, and zinc with coefficient of determination values above 0.5 and acceptable to excellent RPD values. The results of this study show that there is a potential to estimate the nutrients in potato plants in reference to petiole chemical content.