Mineral nutrients play an important role in the biochemistry of grapevine and its growth. Grapevines are known to store significant quantities of certain nutrients to overcome their short-term scarcities in the soil. Hence viticulturists have developed a lot of interest in studying the relationship between the biochemistry of the leaf/petiole and its spectral reflectance to understand the fruit ripening rate, water status, nutrient levels, and disease.
The dataset obtained by measuring the spectral reflectance, defined as the ratio of backscattered radiance from a surface and the incident radiance on that surface, directly over the leaves during the bloom period of growth data in the wavelength region of 330 – 2500 nanometers. This will yield a high dimensional reflectance data with an ill-conditioned covariance matrix. Four regularization and one functional regression method is compared to improve the estimation accuracy and enhance the model interpretability by selecting continuous, unbiased, sparse and useful variables (wavelengths).
