Huanglongbing (HLB) is the main citrus disease worldwide. In the absence of curative or genetic control measures, HLB management is performed with use of healthy seedlings, eradication of symptomatic trees and chemical control of the vector, Diaphorina citri. Despite the management, incidence continues to increase in São Paulo state, Brazil. The difficulty of managing HLB is frequently related to the movement of infective vectors from inoculum sources to orchards where management is carried out (primary spread). The objective of this study was to model HLB dispersion gradient of epidemics in São Paulo, from orchards edges. For this, power law and exponential models were fitted to data of 21 rectangular orchards (~25 ha and ~13.800 trees each), in São Paulo state. The sweet orange orchards were planted, with healthy seedlings, between 2010 and 2012 and managed with at least four HLB symptoms inspection and trees eradication per year and from one to three chemical sprays per month for D. citri control. The gradient models were tested by non-linear regression to the proportion of trees eradicated due HLB symptoms, every 20 m, in relation to the edges of the orchard (concentrically). HLB incidence ranged, on the 21 orchards, from 0.7 to 23.4% of eradicated trees. Both models fitted to data, but the power law presented better fitting than the exponential law for 20 of the 21 orchards. The results confirm the primary spread of HLB in orchards under strict management.

This research was financially supported by the Project 2016/01796-1(FAPESP)