Modelagem matemática aplicada ao comportamento espaço-temporal de Praelongorthezia praelonga (Douglas, 1891) (Hemiptera: Sternorryncha: Ortheziidae) em sistemas de plantio decitros.

Abstract

Insect pests and diseases continue to be the main challenge for agricultural production and the citrus economy in Brazil. The way the pest behaves from place to place is the basis of its infestation and distribution in space and time, creating anextensive list of damage to plantations. The Praelongorthezia praelongamealybug (Douglas, 1891) (Hemiptera: Sternorryncha: Ortheziidae) has been a threat to commercial citrus plantations in the country and in view of the occurrence of this pest in a highly relevant crop, effective implementation of actions that allow the reduction of the pest population through actions that enable its adequate integrated control. In this study, mathematical modeling of two methodological approaches was used with the general objective of studying the behavior of the P.praelongacochineal under two distinct optics, that of geostatistics and the analysis of main coordinates of neighboring matrices (PCNM). The studies were carried out in two commercial planting systems, one agroforestry and the other conventional, both located in the citrus pole of Capitão Poço in the state of Pará. On the one hand, the modeling of aligned semivariograms analyzed and evaluated the spatial distribution of P. praelongain both planting systems using the obtained kriging maps, on the other hand, the PCNM variables that were related to the occurrence of the pest evaluated the influence of the habitat on the space-time propagation speed caused by the same in both areas. Geostatistics has verified theexistence of spatial dependence that ranged from 3.3 to 7.3 m in the agroforestry system and 3.8 to 5.8 m in the conventional system. The results based on PCNM variables showed that despite the geographical proximity, the systems showed differences in thepropagation behavior of the pest both in space and in time. It was observed in both methodologies that the greatest infestations occurred in the conventional system when compared to the agroforestry system. Finally, it was shown that geostatistical modeling is an accurate method to reveal information relevant to the control of P. praelongaand that the modeling of PCNM variables is an explanatory method that efficiently investigates the spatial and temporal relationships between P. praelongaand the type of system of planting in she finds herself.