Menesatti, P.; Antonucci, F.; Pallottino, F.; Roccuzzo, G.; Allegra, M.; Stagno, F.; Intrigliolo, F.
|Abstract in English:|
Nutritional status in citrus plants is normally determined by chemical analysis. These methodologies, carried out according to standardized procedures, are destructive methods used to estimate the plant nutritional status. Moreover they detect tree symptomless detrimental condition or confirm the nature of visible deficiency and toxicity symptoms. This study propose a rapid, non-destructive, cost-effective technique for the prediction of orange leaves nutritional status utilizing a Vis-Nir portable spectrophotometer and its comparison with chemical standard analyses. Trees nutritional status was evaluated by foliar analysis performed on 50 leaves of the index trees. Chemical determinations on leaves regarded the following elements: N, P, K, Ca, Mg, Fe, Zn, Mn. For spectral acquisition, a ‘pen probe’ was used to measure the spectral reflectance response on each single leaf (spot area ? 10 mm2). Mean reflectance values of all leaves for each treatment were compared to each reference chemical value by chemiometric multivariate methods (PLS, Partial Least Square). The best model (r= 0.995) and test (r=0.991) was obtained for the K (potassium). The results show also a high efficiency in the estimation of N (nitrogen) leaf content. Both, model and test of the PLS prediction shown high value of r (respectively, 0.945 and 0.909).