Zoning the occurrence of Thyrinteina arnobia in eucalypts and its relationship with the spectral behaviour of canopies and soil attributes

Abstract

This study aimed to outline areas at risk of the occurrence of eucalypt defoliator caterpillars and their relationship with variations in the spectral behaviour of canopies and soil attributes. The study was conducted in three eucalypt plantation areas of the Bracell Bahia company, located in the state of Bahia, Brazil. Initially, the spatiotemporal distribution of climatic variables and water balance was evaluated. Subsequently, using geographic information systems and fuzzy logic, occurrence risk zoning was developed for defoliator caterpillars associated with different classes of eucalypt crop aptitude. After defining the areas at risk of pest occurrence, specific plots in different aptitude classes were selected to assess, intra-plot, characteristics that can increase or reduce the risk of defoliator caterpillar attack and damage intensity, as well as the plant’s ability to recover after a controlled outbreak. Information on the spectral behaviour of the canopies and the availability of nutrients in the soil was used in this step. Zone modelling enabled the delineation of areas and periods at a higher risk of pest occurrence for the different aptitude conditions of the eucalypt crop. The intra-plot methodology proved that the class of eucalypt crop aptitude with the greatest potential for recovering vegetative vigour 60 days after caterpillar damage was the apt class. This result can be attributed to K limitation and the high spatial resolution of the PlanetScope orbital sensor.

Keywords:

• precision forestry
• remote sensing
• fuzzy logic
• defoliator caterpillars
• nutrients