Submission form for phytosanitary treatments submission number: 2009-tppt-107

Name: Mr. Nacanieli Waqa

Position and organization: Biosecurity Technician

Mailing address: Secretariat of the Pacific Community, Private Mail Bag, Suva, Fiji Islands

Phone: +679 337 0733, Fax: +679 337 0021

E-mail: Nacanieliw@spc.int

(1) the extra time that conditioning in the temperature range 30–42°C increases the estimated lethal time required to achieve 99% mortality (LT₉₉) at a target temperature of 46°C

(2) the decrease in LT₉₉ due to lethality in the range 42–46°C which results in insect mortality as measured at 46°C. The mortality of fruit fly eggs associated with any temperature ramp from 30 to 46°C can be described by this model.

Artocarpus altilis (breadfruit)

Analyses of observed mortality due to treatment at individual temperatures from 42 to 48°C used the model: log(_log(1_p))_a_bt, (1) where p was the expected observed mortality, and t was the treatment time in minutes (Preisler and Robertson, 1989). This gave approximate linearity in time. The estimated LT99, which relates to mortality after there has been allowance for control mortality, was calculated to give an expected mortality of c_(1_c)0.99, where c was the control mortality as determined in the 25°C immersion treatments. Eq. (1) was fitted using a robust version of the generalized linear model analysis available in S-Plus (Chambers and Hastie, 1991) that assumed that any variance was proportional to that of a binomial distribution. The robust version reduces the weight given to points lying away from the main body of the data. This method handles the variability in the response associated with occasional large outliers (Maindonald, 1992). Confidence intervals were calculated so that non-overlap was equivalent to a statistically significant difference in a Student’s t-test at the 1% level. Although the log–log analysis of Eq. (1) can be used successfully to determine LT99 times we have utilized a kinetic expression for mortality to model the response of B. tryoni eggs over the entire treatment history. Rather than use percentage mortality we shall use its complement, percentage survivors, S(T, t)_100 1_mortality 1_control mortality which incorporates Abbott’s method (Abbott, 1925) to correct for control mortality. The treatment time, t, is the independent variable and the equation used for S at each temperature is S(T, t)_100e_(kL t)2 (2) This form is obtained by simplifying a form, S_100e_(c_kt)1:a, suggested by earlier workers (King et al., 1979) and used by several authors to model lethal response of fruit flies (Jang, 1986; Laidlaw and Hayes, 1990). The constant ‘c’ is normally small and was omitted, ‘1:a’ ranges from 1.5 to 2.5 and was set near the mid-point of the range so that only the ‘rate’ constant, kL, was retained as a temperature dependent parameter. To simplify notation we drop the explicit reference to the independent variables and write S(T, t)_S.

To assess mortality response to constant (static) temperature treatments in the lethal range (> 42°C), B. tryoni eggs were immersed in static temperature water baths at 42, 43, 44, 45, 46, 47 and 48°C for various times (1.5–380 min, depending on temperature). The mortality observed for each temperature, T, at intervals ti gave a direct value for mortality as a function of static temperature treatment time. After heat treatment the eggs were hydro-cooled as described above. Three to six replicates were completed at each immersion temperature. Static temperature conditioning heat treatments In order to determine the effects of exposure to temperatures in the non-lethal range (30–42°C) on the mortality response, B. tryoni eggs were first immersed in a water bath at 32, 34, 36, 38, 40 or 42°C for times ranging from 15 min to 12 h (depending on the temperature bath employed). The tubes containing eggs were removed from the pretreatment bath, drained of hot water and within 2 min simultaneously immersed in a second water bath operating at 46°C. Samples were removed after various exposures (0–70 min) until the lethal time for 99% mortality (LT₉₉) could be established. The samples were then hydro-cooled. Comparison of LT₉₉ for conditioned and nonconditioned eggs provided a quantitative measure of the conditioning as a function of temperature and time of the treatment. The temperatures and exposure durations of the pretreatment were chosen so that conditioning and not mortality would result. This was verified in each test with a second control, which experienced the pretreatment component only. Three to seven replicates were completed at each pretreatment time and temperature combination. Fig. 1 illustrates the procedure schematically for a case where the first group of samples was immersed directly in the 46°C target bath, the second group of samples was pretreated at 38°C for 30 min and then immersed in the target 46°C bath, and a final group of samples was pretreated at 38°C for 60 min before being immersed in the target 46°C bath. Samples were removed at intervals from the 46°C bath, the number of non-hatched eggs determined for each sample and the LT₉₉ times evaluated

to prevent egg desiccation.

Although the static temperature treatment results provide an assessment of mortality due to exposure to particular temperatures in the range 30–48°C, it is the effect of the continuously changing temperature during ramped heating that comes closer to addressing the insect mortality response as it occurs in whole fruit. To mimic this, B. tryoni eggs were immersed in a water bath which changed temperature linearly with time from 30 to 46°C in 1, 3, 6, or 9 h (heating rates of 16, 5.3, 2.7 and 1.8°C h^-1 respectively). The eggs so treated were then maintained at 46°C until the LT₉₉ value could be established. A comparison of the LT₉₉ for eggs undergoing ramped heating and the LT₉₉s for eggs immersed in 46°C without prior treatment then provided a direct measure of the conditioning due to temperatures and times encountered for a particular ramp. The effect of ramped heating from 25 to 48°C in 1 h (23°C h^-1) was also determined. Longer ramp times to a target of 48°C were of limited value because of the high levels of kill at 48°C and temperatures near this target. Three replicate tests were conducted at each ramp. Fig. 2 contains a schematic of the procedure for the smooth linear ramp used in experiments as well as a step-wise ramp which will be used later to illustrate the connection between static and ramped treatments.

Egg hatch normally begins about 40 h after laying. Treated and control eggs were held at 26±1°C for 3 days to ensure all hatching was complete and then examined for mortality using a binocular microscope 10-40 x magnification). Egg mortality was based on the presence or absence of an exit rupture in the chorion.