Grower and regulator conflict in management of the citrus disease Huanglongbing in Brazil: a modelling study.
Abstract
In managing plant diseases, there is often tension between a regulator seeking to destroy infected plants to prevent further infection on a national scale and growers seeking to retain infected plants to continue obtaining yield. A high-profile example is Huanglongbing (HLB) or citrus greening, a bacterial disease that threatens Brazil's citrus industry. To prevent the spread of HLB, especially from abandoned infected groves, the government regulated that if 28% of a plantation unit (grove) is found to be symptomatic, then the whole plantation unit must be destroyed. This decision used the best evidence available in 2008, which suggested that a 28% detectable prevalence corresponded with 100% actual prevalence, the disparity being due to asymptomatic infected trees and imperfect detection methods. Using a mathematical model with parameters estimated from field data, we evaluate the assumptions underlying the 28% threshold. We investigate the effects of spraying insecticide and removing diseased plants on the infectious pressure and potential loss of yield from an infected grove. We find that the relationship between detectable and actual prevalence is much wider than allowed for in the regulations. There is a high probability that groves with detectable levels of symptomatic plants substantially below 28% have a >90% prevalence of infected plants. Paradoxically, in a well-managed orchard, the threshold of 28% may not be reached at 100% prevalence. Infectious pressure from an infected grove is substantially reduced when growers control disease. Individual growers failing to manage disease, therefore, threaten the wider grower community. Control is likely to increase yield and prolong grove productivity, but in some groves may reduce yield. Policy implications. Current disease thresholds aimed at restricting the spread of the citrus disease, Huanglongbing (HLB), in Brazil allow heavily infected groves to remain in the landscape, but lower thresholds would disadvantage growers who are already controlling disease. There is probably no threshold that is optimal for individual growers and regulators but roguing and spraying is beneficial to both parties. Regulations should focus less on prevalence thresholds, and instead encourage early detection and co-ordinated spraying among growers to control Huanglongbing on a regional level.