**1.1 Citrus greening disease**

Citrus greening disease is a limiting factor in citrus production worldwide (Bové, 2006). The symptoms of HLB are similar to those of nutritional stress (Halbert & Manjunath, 2004). A survey conducted over an eight-year period in Réunion Island, for example, indicated that 65% of the trees were badly damaged and rendered unproductive within seven years of planting (Aubert et al., 1996). In Thailand, citrus trees generally decline within 5-8 years after planting due to HLB (Roistacher, 1996). In his compilation of global infection statistics, Toorawa (1998) estimated that 50 million trees were infected in South and Southeast Asia, and 3 million were infected in Africa. In India and Saudi Arabia, there has been a marked decline in the citrus industry as a result of HLB.

Development of an Individual-Based Simulation Model

**2. The model** 

for the Spread of Citrus Greening Disease by the Vector Insect *Diaphorina citri* 89

Until quite recently, almost all models of pathogens and hosts were developed within the framework of mean-field models, which assumed that the respective individuals were uniformly distributed and that the respective interactions occurred uniformly. In recent years, however, the increased calculation speed of computers has enabled the development of an Individual-Based Model (IBM), wherein each individual, of the pathogen and the vector, behave individually, acting according to a predefined set of rules. The model is able to examine disease-spread dynamics in a simulation field, by calculating the cumulative results of the individual behaviors. Because it can treat the vector and host individually, the IBM offers a new and powerful tool in the study of insect borne plant disease. We therefore employed this technique to develop an HLB disease-spread model based on the *D. citri* vector, with reference to the pine wilt disease-spread model developed by Takasu (2009). The C language used in writing the source code for the model was based on the C language

technology of simulated individuals, published electronically by Takasu (2008).

In this simulation, we targeted the region of the Mekong Delta, Vietnam, in order to establish the basic parameters. Some parameters were provided by previous reports and

> Status of some *H* trees on which virulent vectors fed change to *LP*.

(d) (c)

Nth transmission cycle (a) (b)

> The virulent individuals start dispersing.

After the latent period, the trees show the symptoms.

Disease spread

The virulent insect emerges from the *IP*.

some by our own observations in the target area. The model is summarized in Fig. 1.

After the latent period, the trees show the symptoms.

Healthy tree (*H*) → Latent tree (*LP*, 3 months) → Infected tree (*IP,* 45 months) → Dead (D)

The virulent insect emerges from the *IP*.

Repeat calculations Output of disease spread Output of disease spread

Fig. 1. (a) Change in tree status over time. (b) Summary of citrus greening disease-spread

Disease spread

Status of some *H* trees on which virulent vectors fed change to *LP*.

Transmission of HLB by the virulent vector'

(d) (c)

1st transmission cycle (a) (b)

The virulent individuals start dispersing.

○:Healthy tree (*H*) ●:Latent tree (*LP*) ●:Infected tree (*IP*) :Non-virulent vector :Virulent vector

From Kobori et al. (2010).

simulation by our model.

Vector migration in disease free orchards

(a)

(b)

**2.1 Framework of the model and estimation of the parameters** 

The pathogens are phloem-inhabiting bacteria in the generous *Candidatus* Liberibacter (Halbert et al., 2004). Although these bacteria have hitherto not been sufficiently cultured for the application of Koch's postulates, some experimental results have strongly suggested that they are the pathogen in HLB (Su et al., 1986; Buitendag & von Broembsen, 1993). There are two principal means of transmission for a healthy tree: graft transmission, whose frequency has been estimated in many previous studies (e.g., Lin & Lin, 1990; van Vuuren, 1993), but with widely varying values; and transmission through vector insects. In Asian countries, HLB is borne by the Asian citrus psyllid, *D. citri*.

Detection of the HLB pathogen has been achieved by several methods. DNA identification through the PCR method was used to detect the bacteria both in citrus plants and vector insects (Bové et al., 1993; Tian et al., 1996). Wang et al. (2006) conducted a study which developed a reproducible conventional PCR method with several primer sets, and two quantitative real-time PCR methods for detection and monitoring of the pathogen. The HLB pathogen also can be detected with an electron microscope, ELISA (Garnier & Bové, 1993).
