PHYSIOLOGICAL ECOLOGY
Overwintering Mortality of Indianmeal Moth (Lepidoptera: Pyralidae) in Southern Minnesota M. A. CARRILLO,1, 2 R. D. MOON,1 W. F. WILCKE,3 R. V. MOREY,3 N. KALIYAN,3 1 AND W. D. HUTCHISON
Environ. Entomol. 35(4): 843Ð855 (2006)
ABSTRACT Indianmeal moth, Plodia interpunctella (Hu¨ bner) (Lepidoptera: Pyralidae), has been classiÞed as one of the most cold-tolerant pests of stored grain. In this study, the supercooling point (SCP) of Þeld-collected, cold-acclimated Þfth instars was determined as an index of cold hardiness. In addition, mortality of laboratory-reared and Þeld-collected, cold-acclimated Þfth instars exposed to ⫺10⬚C was measured to understand the ability of P. interpunctella to survive winter conditions. Finally, the overwintering mortality of this species in southern Minnesota was measured in grain bins Þlled with shelled corn. The SCP of Þeld-collected, cold-acclimated Þfth instars was approximately ⫺24⬚C before their release in grain bins. Mortality of laboratory-reared Þfth instars exposed to ⫺10⬚C reached 100% after only 12 h, whereas the same percentage was achieved after 312 h for Þeld-collected, cold-acclimated individuals. Overwintering mortality of P. interpunctella under Þeld conditions reached 100% in all locations inside the grain bins; however, depth within the grain mass, location inside the grain bin, and the duration of exposure to outdoor conditions signiÞcantly affected the rate of mortality. A mathematical model for the overwintering mortality of P. interpunctella is proposed as a Þrst step to forecast early season infestations under Þeld conditions. KEY WORDS Plodia interpunctella, stored products, cold hardiness, overwintering, modeling
The Indianmeal moth, Plodia interpunctella (Hu¨ bner) (Lepidoptera: Pyralidae), is a cosmopolitan pest that attacks several commodities especially cereals, dried fruits, vegetables, and nuts (Tzanakakis 1959, Na and Ryoo 2000). Barak and Harein (1981) ranked this species as the third most common insect pest of storedcorn in Minnesota. In addition, Harein and Subramanyam (1990) asserted that P. interpunctella is the only pest of harvested soybean in Minnesota. In grain bins, P. interpunctella generally inhabits the headspace and the top few inches of the grain mass (Harein and Subramanyam 1990). However, infestations are also seen in aeration ducts and areas below perforated bin ßoors. The main damage is caused by the larvae whose frass and silken webs can obstruct aeration systems and ultimately decrease the quality of the grain (Abdel-Rahman et al. 1968, Hamid 1990). In the food processing industry, P. interpunctella can inhabit several commodities and can be a direct contaminant of the Þnal food product. For these reasons, this insect is an important pest in different commodities stored in the home, grain bins, and grain elevators. The survival of insects during winter in northern temperate latitudes depends on speciÞc environ1 Department of Entomology, University of Minnesota, St. Paul, MN 55108. 2 Corresponding author: 1677 80th Street, Monmouth, IL 61462 (e-mail:
[email protected]). 3 Department of Biosystems and Agricultural Engineering, University of Minnesota, St. Paul, MN 55108.
mental conditions experienced at the overwintering sites and the inherent overwintering strategy of a given species (Danks 1978, Bale 1991, Leather et al. 1993). In general, insects receive protection from the negative effects of extreme cold temperatures by selecting a favorable microhabitat and by cold hardening (Danks 1978). Cold hardiness has been deÞned as the ability of an organism to withstand exposure to low temperatures (i.e., temperatures below the lower developmental threshold) (Zachariassen 1985, Lee 1991). In general, organisms endure direct exposure to subzero temperatures by either freezing (i.e., freeze-tolerant species) or through supercooling of their body ßuids (i.e., freeze-intolerant species) (Salt 1961). Both strategies of survival at low temperatures require the production of cryoprotectants (e.g., alcohols and sugars) to either protect the cells during the freezing process or to lower the supercooling point (SCP; i.e., the temperature at which spontaneous freezing of body ßuids occurs) (Zachariassen 1985). In addition, the production of thermal hysteresis proteins (i.e., pure proteins or glycoproteins) may be necessary to either avoid recrystallization in freezetolerant insects or to stabilize the supercooled state in freeze-intolerant insects (Zachariassen 1985, Duman et al. 1991). Plodia interpunctella has been classiÞed as a freezeintolerant organisms and one of the most cold-tolerant species among stored-product insects (Fields 1992). Aeration with cold, outdoor air has been recognized as
0046-225X/06/0843Ð0855$04.00/0 䉷 2006 Entomological Society of America
844
ENVIRONMENTAL ENTOMOLOGY
Vol. 35, no. 4
Fig. 1. Mean ⫾ SE SCPs of laboratory-reared (E) and Þeld-collected (F) Þfth-instar P. interpunctella retrieved on different dates in 2004. Rearing condition (laboratory versus Þeld), date of collection, and their corresponding interaction signiÞcantly affected mean SCP (P ⬍ 0.0001).
a valuable tactic for managing stored-grain pests (Salt 1936, Fields 1992, Hagstrum and Flinn 1994, Flinn et al. 1997, Mason and Strait 1998, Fields and White 2002). However, challenges such as abilities of insects to adapt behaviorally (e.g., searching for physical protection) and physiologically (e.g., possible production of cryoprotectants) to the exposure of low temperatures need to be overcome. Despite these challenges, the use of low temperatures to control P. interpunctella still seems to be feasible in the northern United States because ambient temperatures are well below 0⬚C during winter months (Mason and Strait 1998, Kaliyan 2004). Understanding the cold hardiness of the targeted pest can also contribute to the efÞcacy of winter aeration as a pest management tactic. Thus far, some information regarding the cold hardiness of P. interpunctella is available (Mansbridge 1936, Salt 1936, Solomon and Adamson 1955, Tzanakakis 1959, Lee et al. 1992, White et al. 1994, Lewthwaite et al. 1998, Dohino et al. 1999, Naeemullah et al. 1999, Carrillo and Cannon 2005, Carrillo et al. 2005a). However, most of these reports are from laboratory studies, and information about the mortality of this species when exposed to winter temperatures under Þeld conditions is limited. By deÞnition, freeze-intolerant organisms die when temperatures are at or below their SCP. However, considerable mortality has been observed for P. interpunctella at temperatures above the SCP, most likely as a result of chill injury (i.e., possible membrane collapse as a result of excessive thermoelastic stress; Lee 1991) (White et al. 1994, Naeemullah et al. 1999, Carrillo et al. 2005a). Therefore, an aeration plan scheduled to deliver winter air that is between the SCP and 0⬚C into grain-storage facilities could be of potential use to control this pest in temperate regions. The objective of this study was to assess winter mortality of P. interpunctella in southern Minnesota in
76.2-T (3,000-bu) grain bins Þlled with shelled corn. In addition, rates of larval mortality at constant ⫺10⬚C and the seasonal variation of SCPs were described to understand the cold hardiness of P. interpunctella. Finally, a mathematical model for mortality of P. interpunctella at low temperatures is proposed as a Þrst step to forecast possible early season infestation under Þeld conditions. Materials and Methods Experimental Insects and Standard Rearing Conditions. Laboratory-reared Þfth-instar P. interpunctella were obtained from a culture initiated with individuals collected in the Þeld in April 2002 and maintained under laboratory conditions at 28⬚C, 65Ð70% RH, and a photoperiod of 14:10 (L:D) h. Larvae were reared on a diet of wheat bran, chick feed, corn meal, glycerol, honey, and water at a volumetric ratio of 30:20:10:6.5: 1.5:1, respectively (Carrillo and Cannon 2005). Field-collected Þfth instars were obtained as needed from outdoor cultures initiated with laboratory-reared P. interpunctella adults in early autumn 2003 and 2004 at the University of Minnesota Outreach, Research, and Education (UMORE) Park, Rosemount, MN. In 2003, an outdoor culture was initiated with adults released on 28 August and 14 September. In 2004, another culture was initiated with adults released on 4 August. Adults were released into 98-liter plastic trash cans (Trashmaster, Sardis, MS) that were 75% Þlled with shelled corn and supplemented with 0.90 kg of the larval diet. Diet was added as an immediate food source for Þrst instars that were expected to hatch a few days after adults were released. Corrugated cardboard spools were placed on top of the diet to serve as protection sites for overwintering larvae and to facilitate their collection. The cans were closed with a Þne “no-see-um” mesh fabric (Outdoor Wilderness Fabric, Nampa, ID) to conÞne
August 2006
CARRILLO ET AL.: OVERWINTERING MORTALITY OF P. interpunctella
845
Fig. 2. Corrected mean ⫾ SE percentage mortality of Þfth-instar P. interpunctella exposed to ⫺10⬚C for different lengths of time. (A) laboratory-reared and (B) Þeld-collected, cold-acclimated. Each mean represents 30 larvae in three independent replications. Means followed by different lowercase letters are signiÞcantly different (P ⬍ 0.0001). Mortality for laboratoryreared and Þeld-collected individuals was corrected for ⬇3 and 13% natural mortality, respectively.
adults. A velcro-closed opening in the middle of the fabric facilitated larval collection. SCP Determination. The SCP was used as an index of cold acclimation for Þeld-collected Þfth-instar P. interpunctella (Carrillo and Cannon 2005). SCPs were measured using surface-contact thermometry as described by Carrillo et al. (2004). Fifth instars were attached to 24-gauge copper-constantan thermocouples with the help of a thin layer of high-vacuum grease (Dow Corning, Midland, MI). After attachment, two insect-thermocouple arrangements were placed at the centers of 19 by 19 by 19-cm polystyrene containers with a starting temperature of ⬇0⬚C. The containers were closed with rubber stoppers and transferred to a ⫺80⬚C freezer (Revco ScientiÞc, Asheville, NC) to cool insects at a rate of ⬇1⬚C/min (Carrillo et al. 2004). A total of Þve containers were used per cooling period. Body temperatures were recorded at 1-s intervals with a multichannel data logger (Personal Daq/56 data acquisition system; IOtech, Cleveland, OH), transferred to a personal computer, downloaded to a spreadsheet, and graphed as a scatter plot. The SCP was determined as the lowest temperature reached before freezing (Lee 1991). Freezing was visualized by a small rise in temperature during an otherwise decreasing trend, indicating the release of latent heat of fusion (Lee 1991). In 2003, SCPs of Þeld-collected Þfth instars (n ⫽ 37) were recorded on the day of collection on 13 November. In 2004, SCPs of Þfth instars (n ⫽ 19 Ð39) were recorded throughout the acclimation period, and a direct comparison between laboratory-reared and Þeld-collected individuals was conducted by measuring the SCP of both groups during the same cooling
period. SCPs were measured on the day of collection on 13 September, 1 October, 15 October, 4 November, 23 November, and 11 December. SCPs in 2004 were transformed to the power of 1.75, as recommended by the Box-Cox procedure (MacAnova 2002) and analyzed using analysis of variance (ANOVA; PROC GLM, SAS Institute 1995). Rearing condition (i.e., laboratory or Þeld), sampling date, and the interaction of rearing condition and sampling date were included as predictors in the ANOVA model. Mortality of P. interpunctella Exposed to ⴚ10ⴗC Under Laboratory Conditions. Laboratory-reared and Þeld-collected, cold-acclimated Þfth instars were studied to assess the effect of cold acclimation on mortality when larvae were exposed to a constant ⫺10⬚C. Corrugated cardboard spools containing Þeldcollected, cold-acclimated larvae were obtained from the previously established outdoor culture on 18 November 2003 and stored in an upright freezer at ⬇0⬚C for 13Ð26 d. At the time of use, larvae were removed from the 0⬚C freezer, placed into a programmable growth chamber (Percival ScientiÞc, Perry, IA) at 0⬚C, and warmed to room temperature (⬇25⬚C) at a rate of 0.3⬚C/min. Subsequently, larvae were carefully removed from the cardboard spools using soft forceps. Laboratory-reared or Þeld-collected, cold-acclimated larvae were placed in groups of 10 into 95 by 15-mm plastic petri dishes. Petri dishes were placed into a programmable growth chamber at room temperature and insects were cooled to ⫺10⬚C at a rate of 0.3⬚C/ min. This treatment temperature was selected because it was higher than the mean SCP of Þeld-collected larvae (n ⫽ 40) from the same cohort at the time of the experiment. In addition, the treatment
846
ENVIRONMENTAL ENTOMOLOGY
Vol. 35, no. 4
Fig. 3. Corrected mean ⫾ SE percentage mortality (solid lines) of overwintering Þfth-instar P. interpunctella at two depths (⬇5 and 40 cm), two locations (C ⫽ center, E ⫽ edge), and seven sampling dates inside 76.2-T (3,000-bu) grain bins Þlled with shelled corn during the winter of 2003Ð2004. Dotted line (A) represents daily mean air temperature; dashed lines (AÐD) represent daily mean temperature at each studied depth and location; solid triangles represent aeration events. Each mean at the center and edge of the bin represents three and four independent replications, respectively. Depth, location, sampling date, and the interaction of depth and sampling date signiÞcantly affected larval mortality (P ⬍ 0.05). Mortality was corrected for ⬇5% natural mortality. Mean ⫾ SE SCP for Þfth instars before their release in grain bins was ⫺23.46 ⫾ 0.63⬚C. Minimum hourly winter temperature experienced at any measured location within the grain bin was ⫺21.79⬚C.
temperature was also higher than the mean SCPs previously reported for unacclimated larvae from the same colony (Carrillo and Cannon 2005, Carrillo et al. 2005a). A 24-gauge copper-constantan thermocouple was placed inside the chamber to monitor temperature ßuctuations; temperature readings were taken every 5 min. Laboratory-reared larvae were removed after 4, 8, 12, 16, 24, 28, 32, and 36 h of exposure. Field-collected, cold-acclimated larvae were removed after 24, 48, 72, 96, 120, 144, 168, 192, 216, 264, and 312 h of exposure. After the designated exposure times, insects were warmed to room temperature at a rate of 0.3⬚C/min, and mortality was assessed as lack of adult emergence after 1 mo of further incubation at standard rearing conditions. A control group, for both laboratory-reared and Þeld-collected, cold-acclimated larvae, was held at standard rearing conditions to correct for natural mortality (Abbott 1925). Experiments were independently replicated three times. Mortality data were analyzed using ANOVA (PROC GLM, SAS Institute 1995) performed on the arcsinesquare-root transformed proportion of mortality. Time of exposure was the only predictor included in the ANOVA model because experiments for labora-
tory-reared and Þeld-collected, cold-acclimated larvae were conducted at different cooling periods. Mortality of P. interpunctella Under Field Conditions. Field-collected, cold-acclimated Þfth instars were released into 76.2-T (3,000-bu) grain bins located at UMORE Park to measure the mortality of P. interpunctella under Þeld conditions. In 2003, four grain bins Þlled with shelled corn were used to establish the experiment on 20 November. In 2004, the experiment was established on 11 December in three grain bins Þlled with shelled corn. To place larvae inside grain bins, small cages were made of stainless steel mesh (0.023-cm wire diameter and 50 wires per 2.54 cm) rectangles 9.5 cm long by 5.0 cm wide. Cages were made by stapling the two ends of the stainless steel mesh rectangle to form a cylinder that was 2.54 cm in diameter by 5.0 cm high. The two cylinder openings were closed using 2.54cm-diameter plastic lids (J & R products, Craigville, IN). Cages were Þlled with a mean ⫾ SE of 8.63 ⫾ 0.14 g of organic shelled corn (Great River Milling, Winona, MN) to allow contact of the larvae with the grain and to simulate the conditions inside the grain
August 2006
CARRILLO ET AL.: OVERWINTERING MORTALITY OF P. interpunctella
847
Fig. 4. Mean ⫾ SE percentage mortality (solid lines) of overwintering Þfth-instar P. interpunctella at two depths (⬇5 and 40 cm), two locations (C ⫽ center, E ⫽ edge), and three sampling dates inside 76.2-T (3,000-bu) grain bins Þlled with shelled corn during the winter of 2004Ð2005. Dotted line (A) represents daily mean air temperature; dashed lines (AÐD) represent daily mean temperature at each studied depth and location; solid triangles represent aeration events. Each mean represents three independent replications, except for the last two sampling dates in A and C that represent two independent replications. Depth, sampling date, and the interaction of depth and sampling date signiÞcantly affected larval mortality (P ⬍ 0.05). Mean ⫾ SE supercooling point for Þfth instars before their release in grain bins was ⫺23.76 ⫾ 0.54⬚C. Minimum hourly winter temperature experienced at any given measured location the grain bin was ⫺21.79⬚C.
mass. Each cage then received 10 cold-acclimated larvae collected at random from the outdoor cultures. Larvae were carefully removed from the cardboard spools and placed in each cage using soft forceps. This procedure was carried out in the Þeld under the protection of a small unheated shed with temperatures ranging between 0 and 5⬚C. Caged larvae were placed at random inside the grain bins at two horizontal locations (i.e., center and southern edge) and at each of two depths (i.e., ⬇5 and 40 cm below the top grain surface) per location. Cages at the southern edge of the bin were ⬇10 Ð15 cm away from the bin wall. In the winter of 2003Ð2004, each of two grain bins (n ⫽ 2) randomly received a fan-assisted temperature treatment ranging from either ⫺7 to ⫺4 or ⫺1 to 2⬚C as recommended for the state of Minnesota (Cloud and Morey 1991). Treatments were applied using aeration fans (i.e., 1.5Ð2.0 HP). The initial temperature of all four grain bins was equilibrated at ⬇0 to 5⬚C before the beginning of the experiment by turning on the fans when the air temperature was around these values. In addition, all bins were aerated for 5 h the day after the experiment was set up (21 November 2003). Bins designed to receive the aeration treatment between
⫺7 and ⫺4⬚C were aerated on 25 November (6 h) and on 19 December (3 h) 2003. Bins designed to receive the aeration treatment between ⫺1 and 2⬚C were aerated on 25 November 2003 (4 h) and on 12 January 2004 (4 h). In the winter of 2004 Ð2005, no particular aeration treatment was used. However, the initial temperature of all three bins was equilibrated at ⬇0 Ð5⬚C before the beginning of the experiment by turning on the fans when the air temperature was around these values. In addition, bins were aerated as determined by the UMORE Park personnel on 12 December 2004 (17 h) and 25 January 2005 (288 h). Temperatures at the four aforementioned depth locations inside the grain bins were recorded hourly, whereas air temperatures were recorded every 4 h using HOBO data loggers (Onset Computer, Cape Cod, MA). To determine percentage mortality, two cages per depth location were removed at random on separate sampling dates. In the winter of 2003Ð2004, cages were removed on 2 and 17 December 2003 and on 2 January, 17 January, 4 February, 22 February, and 14 March 2004. In the winter of 2004 Ð2005, cages were removed on 31 December 2004 and on 22 January and 6 February 2005. Cages were brought to the laboratory,
848
ENVIRONMENTAL ENTOMOLOGY
Table 1. ANOVA of the effect of aeration regimen (ⴚ7 to ⴚ4 or ⴚ1 to 2°C), depth (⬇5 versus 40 cm), location (center versus southern edge), and sampling date on the mortality of overwintering fifth-instar P. interpunctella during the winter of 2003–2004 at Rosemount, MN Source of variation
df
MS
F
P
Aeration Error I Depth Location Sampling date Aeration ⫻ depth Aeration ⫻ location Aeration ⫻ sampling date Depth ⫻ location Depth ⫻ sampling date Location ⫻ sampling date Error II
1 2 1 1 6 1 1 6 1 6 6 65
0.02624 0.04067 0.61993 0.21626 5.00033 0.08728 0.00117 0.02028 0.09746 0.07880 0.05793 0.02985
0.65 1.36 20.76 7.24 167.49 2.92 0.04 0.68 3.26 2.64 1.94
0.5061 0.2633 ⬍0.0001 0.0090 ⬍0.0001 0.0921 0.8435 0.6666 0.0754 0.0236 0.0874
Three- and four-way interactions were not included in the model (P ⬎ 0.05).
placed into a programmable growth chamber at 0⬚C, and warmed to room temperature at a rate of 0.3⬚C/ min. Larvae were removed from the cages and transferred to individual 95 by 15-mm plastic petri dishes supplied with larval diet and placed under standard rearing conditions. Mortality was deÞned as lack of larval movement when prodded with a pair of soft forceps 24 h after transfer. For each year, a control group of Þeld-collected, cold-acclimated Þfth instars (2003: 42 larvae; 2004: 60 larvae) was separated at the beginning of the experiment and held at standard rearing conditions to correct for natural mortality (Abbott 1925). Mortality data were analyzed using ANOVA (PROC GLM, SAS Institute 1995) performed on the arcsine-square-root transformed proportion of mortality. For the winter of 2003Ð2004, data were analyzed according to a split-plot design. The whole plot treatment was the aeration regimen, and the subplot treatment was the speciÞc site of the cages within the grain mass. For the winter of 2004 Ð2005, data were analyzed using ANOVA with location, depth, and sampling date as predictors in the model. In both winters, when signiÞcant differences were found, means were separated using TukeyÕs studentized range test (honestly signiÞcant difference [HSD]) at ␣ ⫽ 0.05. The ANOVA models for both winters included the main Table 2. ANOVA of the effect of depth (⬇5 versus 40 cm), location (center versus southern edge), and sampling date on the mortality of overwintering fifth-instar P. interpunctella during the winter of 2004 –2005 at Rosemount, MN Source of variation
df
MS
F
P
Depth Location Sampling date Depth ⴛ location Depth ⴛ sampling date Location ⴛ sampling date Error
1 1 2 1 2 2 23
0.32582 0.04015 1.32659 0.00391 0.35564 0.04403 0.03017
10.80 1.33 43.97 0.13 11.79 1.46
0.0032 0.2605 ⬍0.0001 0.7221 0.0003 0.2531
Three-way interaction was not included in the model (P ⬎ 0.05).
Vol. 35, no. 4
effects and corresponding interactions. Three- and four-way interactions were removed from the model when P ⬎ 0.05. Overwintering Mortality Model. A mortality model similar to that reported by Carrillo et al. (2005b) for the bean leaf beetle, Cerotoma trifurcata (Fo¨ rster) (Coleoptera: Chrysomelidae), was developed for cold-acclimated Þfth-instar P. interpunctella under Þeld conditions. Mortality data collected from the previous experiment were used to develop the model. Logistic regression (PROC Logistic, SAS Institute 1995) was used to model the probability of death as a function of the cumulative cooling degree hours (⬚C ⫺ h; CCDH) below a selected threshold temperature. The CCDH at each particular depth (i.e., ⬇5 and 40 cm below the top grain surface) and location (i.e., center and southern edge) within the grain bins was the only predictor used in the regression model and was calculated as
冘 n
CCDH ⫽
(Tr ⫺ Tmi) ⫻ ti
[1]
i⫽1
where Tr is the threshold temperature (⬚C), Tm is the measured temperature (oC), t is the time interval (i.e., 1 h), and i is the observation number, from 1 to n. To select the most appropriate Tr, separate logistic regressions of the proportion of mortality of overwintering P. interpunctella versus CCDHs were calculated for Tr ranging from ⫺6 to 6⬚C. This range of Tr was selected because temperatures lower than ⫺6⬚C would greatly ignore the effect of high subzero temperatures on larval mortality, and temperatures warmer than 6⬚C did not commonly occur in our Þeld experiments. From each regression, the deviance (also known as G2, Cook and Weisberg 1999) was plotted versus its respective Tr to determine the temperature(s) that yielded the lowest error in the model. When multiple Tr yielded a similar deviance, the Tr closest to 0⬚C was selected. In the logistic regression, the independent variable was the CCDH to which individuals were exposed. The dependent variable PCCDH was the proportion of individuals dying at each particular CCDH. The PCCDH at CCDH was linearized following the form ln
冉
冊
PCCDH ⫽ m ⫻ CCDH ⫹ b 1 ⫺ PCCDH
[2]
where m and b were coefÞcients representing the slope and the intercept of the line, respectively. Equation 2 was back-transformed as 1 SCCDH⫽ 1⫹e⫺(m ⫻ CCDH⫹b)
[3]
where SCCDH was the estimated proportion of overwintering cold-acclimated Þfth-instar P. interpunctella dying at each particular CCDH. The performance of the logistic model developed with a CCDH that carried the smallest deviance for the winter of 2003Ð2004 was visualized by comparing predicted and observed mortality of P. interpunctella
August 2006
CARRILLO ET AL.: OVERWINTERING MORTALITY OF P. interpunctella
849
Fig. 5. Mean ⫾ SE percentage mortality (solid lines) of Þfth-instar P. interpunctella as a result of a “hot spot” at two depths (⬇5 and 40 cm) at the center of 76.2-T (3,000-bu) grain bins Þlled with shelled corn during the winters of 2003Ð2004 (A and C) and 2004Ð2005 (B and D). Dotted lines represent daily mean air temperature; dashed lines represent daily mean temperature at each speciÞc depth; solid triangles represent aeration events. Each mean represents one replication. Mortality for the winter of 2003Ð2004 was corrected for ⬇5% natural mortality.
larvae for the winter of 2004 Ð2005. Finally, information from the two winters studied (i.e., 2003Ð2004 and 2004 Ð2005) was used to develop a Þnal model to predict the overwintering mortality of P. interpunctella in southern Minnesota. Results and Discussion SCPs. The mean SCP of Þeld-collected larvae at the time of their release in grain bins was ⫺23.46 ⫾ 0.63 and ⫺23.76 ⫾ 0.54⬚C (SE) for 2003 and 2004, respectively. In 2004, overall signiÞcant differences were found when comparing SCPs for laboratory-reared and Þeld-collected larvae (F ⫽ 31.93; df ⫽ 11,266; P ⬍ 0.0001; Fig. 1). The SCP was signiÞcantly affected by rearing condition (i.e., laboratory versus Þeld; F ⫽ 163.41; df ⫽ 1,266; P ⬍ 0.0001) and sampling date (F ⫽ 23.87; df ⫽ 5,266; P ⬍ 0.0001). In addition, the interaction of rearing condition and sampling date was also signiÞcant (F ⫽ 21.78; df ⫽ 5,266; P ⬍ 0.0001). The mean SCP for laboratory-reared larvae was relatively stable during the 3-mo period (Fig. 1). In contrast, the SCP of larvae reared in the Þeld declined signiÞcantly by ⬇7⬚C between 1 and 15 October 2004 (Fig. 1). Our
results corroborate those previously reported by Carrillo and Cannon (2005) for the same species where a decline of ⬇10⬚C was observed in the SCPs of individuals measured in the summer compared with those measured in the fall. However, mechanisms underlying this phenomenon are still unknown for this species. Naeemullah et al. (1999) suggested that P. interpunctella larvae may increase its supercooling ability by clearing their guts and thereby eliminating sites for ice nucleation. According to Zachariassen et al. (2004), gut cleaning alone can depress the SCP ⬇10⬚C in freeze-intolerant insects. However, the possibility of production of low-molecular weight compounds (e.g., alcohols and sugars) or antifreeze proteins for this species cannot be discarded as a factor inducing or contributing to the depression in SCP during winter (Sømme 1982). The mean SCP for Þeld-collected larvae used to assess mortality at ⫺10⬚C (i.e., ⫺24 ⫾ 0.65⬚C) was similar to values reported for larvae before their release in grain bins. This low mean SCP indicated that larvae were already acclimated by the time of collection, probably as a result of the direct exposure to
850
ENVIRONMENTAL ENTOMOLOGY
Vol. 35, no. 4
Fig. 6. Deviance versus threshold temperature used to calculate the cumulative cooling degree hours. E, winter of 2003Ð2004; F, combined winters of 2003Ð2004 and 2004Ð2005.
environmental cues (e.g., photoperiod and temperature). Mortality of P. interpunctella Exposed to ⴚ10ⴗC Under Laboratory Conditions. Percentage mortality increased signiÞcantly with time of exposure for both laboratory-reared (F ⫽ 46.07; df ⫽ 9,20; P ⬍ 0.0001) and Þeld-collected, cold-acclimated larvae (F ⫽ 7.32; df ⫽ 11,24; P ⬍ 0.0001; Fig. 2). Laboratory-reared larvae reached 100% mortality after only 12 h of exposure, whereas Þeld-collected, cold-acclimated larvae reached 100% mortality after 312 h of exposure. According to Fields (1992), cold-acclimated organisms are in general 2Ð10 times more cold hardy than unacclimated counterparts. Our results indicated that Þeld-collected P. interpunctella larvae were ⬇26 times more tolerant to cold than laboratory-reared individuals. Additionally, our results corroborated cold acclimation for Þeld-collected larvae, as determined by their low mean SCP (i.e., ⫺24 ⫾ 0.65⬚C). Similar results have been reported for this species in both laboratory-reared and cold-acclimated larvae. For example, Naeemullah et al. (1999) found that laboratory-reared larvae exposed to ⫺10⬚C reached 100% mortality after only 8 h of exposure. In contrast, White et al. (1994) found 90 Ð100% mortality of coldacclimated larvae after 336 h of exposure to ⫺10⬚C. However, SCPs reported by White et al. (1994) for Table 3. Parameter estimates of logistic regression models for the proportion of overwintering mortality of P. interpunctella versus the CCDH Model
Intercept ⫾ SE
Slope ⫾ SE
df Deviance
2003Ð2004a ⫺4.5795 ⫾ 0.3386 0.000627 ⫾ 0.000046 34 Combinedb ⫺3.6264 ⫾ 0.2466 0.000977 ⫾ 0.000063 55
58.34 210.73
Percentage mortality can be calculated as: SCCDH ⫽ 100 ⫻ (1/1 ⫹ e⫺关(slope ⫻ CCDH) ⫹ Intercept兴 a CCDH was calculated using a threshold temperature of 5⬚C. b Represents the winters of 2003Ð2004 and 2004 Ð2005. CCDH was calculated using a threshold temperature of 2⬚C.
diapausing cold-acclimated larvae were only ⫺13⬚C, ⬇10⬚C higher than those reported here. Additionally, the same authors found that 100% mortality of coldacclimated larvae occurred after only 24 h of exposure to ⫺15⬚C, and they attributed mortality to freezing. Therefore, the results reported here, and in White et al. (1994), suggest that grain aeration when air temperatures are ⬍-10⬚C can kill cold-acclimated P. interpunctella larvae in relatively short periods of time. Mortality of P. interpunctella Under Field Conditions. Temperatures within the grain bins varied with location and depth. Fluctuations at the surface and southern edge of the bin were greater than those located deeper and at the center (Figs. 3 and 4). Similar temperature differences and ßuctuations within a grain bin have been previously reported (Flinn et al. 1992, 1997). SigniÞcant differences in timing of larval mortality were observed in the winters of 2003Ð2004 (F ⫽ 35.17; df ⫽ 32,65; P ⬍ 0.0001) and 2004 Ð2005 (F ⫽ 14.50; df ⫽ 9,23; P ⬍ 0.0001). Mortality of P. interpunctella reached 100% in all sites (Figs. 3 and 4). In both winters, the low mean SCPs of larvae measured at the time of experimental set-up (i.e., ⫺23.46 and ⫺23.76⬚C for 2003Ð2004 and 2004 Ð2005, respectively), and the minimum temperatures recorded at the experimental sites (i.e., ⫺21.79⬚C; Figs. 3 and 4) suggest that most mortality of P. interpunctella occurred as a result of accumulation of chill injuries rather than freezing. In the winter of 2003Ð2004, no signiÞcant differences in levels of mortality were found between aeration regimens (Table 1). However, aeration regimens used as treatments in this study only allowed us to aerate the grain bins four times (Fig. 3). Aeration was not carried on more occasions because of sudden drops of air temperature below the treatment ranges soon after the experiments started (Fig. 3). Despite the lack of signiÞcant difference between aeration treatments in the winter of 2003Ð2004, mortality was signiÞcantly affected by the depth within the grain
August 2006
CARRILLO ET AL.: OVERWINTERING MORTALITY OF P. interpunctella
mass, location inside the grain bin, and the duration of exposure to outdoor conditions (i.e., sampling date; Table 1; Fig. 3). Across sampling dates, percentage mortality of caged larvae increased as the winter progressed (Fig. 3). With depth, caged larvae at the surface (i.e., ⬇5-cm depth below the top grain surface) died earlier than those deeper into the grain mass (i.e., ⬇40-cm depth below the top grain surface; Fig. 3). With location, caged larvae at the southern edge of the grain bin died earlier than those at the center of the grain bin. Our results indicate that overwintering P. interpunctella are likely to receive a higher degree of protection from cold temperatures when overwintering at the center of the grain bin and deeper into the grain mass (e.g., Fig. 3C). At this location and depth combination, temperature ßuctuation was minimal and 100% mortality was reached about 1 mo later than in the least protected area
851
(Fig. 3B), where temperatures were more similar to temperatures of outdoor air. In the winter of 2004 Ð2005, mortality occurred at a faster rate than in the winter of 2003Ð2004 (Figs. 3 and 4); however, a similar strong effect of depth, sampling date, and the interaction of depth and sampling date on timing of larval mortality was observed (Table 2; Fig. 4). Despite the observed similarities in both winters, mortality was not signiÞcantly affected by the location within the grain bin in the winter of 2004 Ð 2005 (Table 2). Given these results, the center of grain bins and the insulation of the grain mass are likely to confer greater protection for overwintering P. interpunctella in temperate regions. In addition, the lack of an aeration plan and the use of larger-size grain bins are also likely to contribute to the overwintering success of P. interpunctella in cold climates (Yaciuk et al. 1975, Flinn et
Fig. 7. Percentage mortality of overwintering Þfth-instar P. interpunctella versus CCDHs. Open and closed circles represent observed mortality from the winters of 2003Ð2004 and 2004Ð2005, respectively. Dashed and solid lines represent predicted mortality based on the logistic model for the winter of 2003Ð2004 and for the combined winters of 2003Ð2004 and 2004Ð2005, respectively. (A) CCDH was calculated with a threshold temperature of 5⬚C. (B) CCDH was calculated with a threshold temperature of 2⬚C. Threshold temperatures selected had the lowest deviance in the logistic model.
852
ENVIRONMENTAL ENTOMOLOGY
Vol. 35, no. 4
Fig. 8. Observed ⫾ 95% CL (F) and predicted (solid line) percentage mortality of overwintering Þfth-instar P. interpunctella at two depths (⬇5 and 40 cm), two locations (C ⫽ center, E ⫽ edge), and seven sampling dates inside 76.2-T (3,000-bu) grain bins Þlled with shelled corn during winter 2003Ð2004. Predicted mortality is based on the combined logistic model: SCCDH ⫽ 100 ⫻ (1/1 ⫹ e⫺[(0.000977 ⫻ CCDH) ⫺ 3.6264]), where S is the estimated percentage mortality at a particular CCDH ⬍ 2⬚C. Dotted lines represent hourly temperature at each studied depth and location.
August 2006
CARRILLO ET AL.: OVERWINTERING MORTALITY OF P. interpunctella
853
Fig. 9. Observed ⫾ 95% CL (F) and predicted (solid line) percentage mortality of overwintering Þfth-instar P. interpunctella at two depths (⬇5 and 40 cm), two locations (C ⫽ center, E ⫽ edge), and three sampling dates inside 76.2-T (3,000-bu) grain bins Þlled with shelled corn during winter 2004Ð2005. Predicted mortality is based on the combined logistic model: SCCDH ⫽ 100 ⫻ (1/1 ⫹ e⫺[(0.000977 ⫻ CCDH) ⫺ 3.6264]), where S is the estimated percentage mortality at a particular CCDH ⬍ 2⬚C. Dotted lines represent hourly temperature at each studied depth and location.
854
ENVIRONMENTAL ENTOMOLOGY
al. 1992, Hagstrum and Flinn 1994, Flinn et al. 1997). However, in our study, the small number of aeration events and possibly high initial moisture content of the grain resulted in the formation of a “hot spot” (i.e., temperatures of ⬇60⬚C as a result of fungal growth) at the center of one of the bins in each winter, which caused 100% mortality in caged larvae early in the study (Fig. 5). Similar temperatures have been reported for hot spots in grain bins (Sauer et al. 1992, Banks and Fields 1995). Despite the probability for hot spot development at bin centers, overwintering larvae in unconÞned conditions could be triggered to move away from the center and relocate in areas where temperatures are not extreme, thus increasing their chances for survival. Overwintering Mortality Model. A Tr of 5⬚C, used to calculate CCDH for the winter 2003Ð2004, yielded the lowest deviance (Fig. 6). Parameter estimates for the logistic model, from the winter of 2003Ð2004 (Table 3), were used to calculate the estimated proportion of mortality at a particular CCDH following equation 3. This model was found to provide a good Þt for the data used for parameter estimation (Fig. 7A). However, predicted mortality for the winter of 2004 Ð 2005 was generally lower than that observed in the Þeld (Fig. 7A). Possible explanations for this pattern of predictions may reßect the fact that temperatures encountered early in the winter of 2004 Ð2005 were lower than occurred in 2003Ð2004 (Figs. 8 and 9). Temperatures close to the SCP early in the season of 2004 Ð2005 (Fig. 9) may have caused more chill injuries, thus resulting in an earlier mortality at relatively few CCDHs. Therefore, the model was Þt again, using data from both winters combined (Table 3; Fig. 7B), using a new Tr of 2⬚C, which yielded the lowest deviance (Fig. 6). This combined model showed a good Þt to the overwintering mortality at each particular depth and location within the grain bin in both winters (Figs. 7B, 8, and 9). According to Bale (1991), the study of the overwintering mortality of a particular species under Þeld conditions is more informative and useful than the information obtained from laboratory experiments. However, laboratory data are also important to explain certain events occurring in the Þeld (Bale 1991). For example, from this study, it is understood that, despite the extremely low minimum temperature recorded inside grain bins (⫺21.79⬚C), the mean SCP of overwintering P. interpunctella (approximately ⫺24⬚C) was never reached, thus suggesting that mortality likely occurred as a result of chill injury and not as a result of freezing. As with many mathematical models, if the CCDH falls considerably outside the range used for parameter estimation, the threshold temperature used to calculate CCDH is different, and/or external factors (e.g., grain protectants, moisture, and hot spots) induce considerable mortality for overwintering P. interpunctella, the predictions made by these models will likely be inaccurate. In addition, the mathematical model can be subsequently reÞned or adjusted to Þt particular conditions by including data from different
Vol. 35, no. 4
environmental conditions (Bale 1991). For example, data from different years and for different storage facilities would be necessary to build a more robust model to better predict the overwintering mortality of this species in southern Minnesota. Nevertheless, this type of mortality model seems to have good potential as a pest management tool to forecast infestation levels of P. interpunctella the following season. This study quantiÞed the overwintering mortality of P. interpunctella under laboratory and Þeld conditions. Our results show that overwintering mortality of P. interpunctella in grain bins is primarily a function of temperature as has been found for other insect species (Lam and Pedigo 2000). Finally, our results further show that winter aeration in temperate regions could be the primary tactic to manage overwintering P. interpunctella in low-temperature environments such as granaries and other food storage facilities. Acknowledgments We thank D. Walgenbach, P. Larsen, and J. Rowe for providing grain bins for Þeld experiments, C. Kulhanek and C. Esnaola for technical assistance, C. Bingham and Y. Zhang for help with statistical analyses, E. Burkness for collecting air temperature data, and G. Heimpel, G. Felton, and an anonymous reviewer for constructive reviews of this manuscript. This work was funded by a University of Minnesota Grantin-Aid, The Anderson Research Grant Program, and a University of Minnesota Doctoral Dissertation Fellowship awarded to M. A. Carrillo.
References Cited Abbott, W. S. 1925. A method of computing the effectiveness of an insecticide. J. Econ. Entomol. 18: 265Ð267. Abdel-Rahman, H. A., A. C. Hodson, and C. M. Christensen. 1968. Development of Plodia interpunctella (Hb.) (Lepidoptera, Phycitidae) on different varieties of corn at two levels of moisture. J. Stored Prod. Res. 4: 127Ð133. Bale, J. S. 1991. Implications of cold hardiness for pest management, pp. 461Ð 498. In R. E. Lee Jr. and D. L. Denlinger (eds.), Insects at low temperature, 1st ed. Chapman & Hall, New York. Banks, J., and P. Fields. 1995. Physical methods for insect control in stored-grain ecosystems, pp. 353Ð 409. In D. S. Jayas, N.D.G. White, and W. E. Muir (eds.), Stored-grain ecosystems. Marcel Dekker, New York. Barak, A. V., and P. K. Harein. 1981. Insect infestation of farm-stored shelled corn and wheat in Minnesota. J. Econ. Entomol. 74: 197Ð202. Carrillo, M. A., and C. A. Cannon. 2005. Supercooling point variability in the Indianmeal moth, Plodia interpunctella (Hu¨ bner) (Lepidoptera: Pyralidae). J. Stored Prod. Res. 41: 556 Ð564. Carrillo, M. A., N. Kaliyan, C. A. Cannon, R. V. Morey, and W. F. Wilcke. 2004. A simple method to adjust cooling rates for supercooling point determination. CryoLetters 25: 155Ð160. Carrillo, M. A., C. A. Cannon, W. F. Wilcke, R. V. Morey, N. Kaliyan, and W. D. Hutchison. 2005a. Relationship between supercooling point and mortality at low temperatures in Indianmeal moth (Lepidoptera: Pyralidae). J. Econ. Entomol. 98: 618 Ð 625.
August 2006
CARRILLO ET AL.: OVERWINTERING MORTALITY OF P. interpunctella
Carrillo, M. A., R. L. Koch, E. C. Burkness, K. Bennett, D. W. Ragsdale, and W. D. Hutchison. 2005b. Supercooling point of bean leaf beetle (Coleoptera: Chrysomelidae) in Minnesota and a revised predictive model for survival at low temperatures. Environ. Entomol. 34: 1395Ð1401. Cloud, H. A., and R. V. Morey. 1991. Management of stored grain with aeration. University of Minnesota, St. Paul, MN. Cook, R. D., and S. Weisberg. 1999. Applied regression including computing and graphics. Wiley, Indianapolis, IN. Danks, H. V. 1978. Modes of seasonal adaptation in the insects. Part I: winter survival. Can. Entomol. 110: 1167Ð 1205. Dohino, T., S. Masaki, I. Matsuoka, M. Tanno, and T. Takano. 1999. Low temperature as an alternative to fumigation for disinfesting stored products. Res. Bull. Plant Protection Jpn. 35: 5Ð14. Duman, J. G., L. Xu, L. G. Neven, D. Tursman, and D. W. Wu. 1991. Hemolymph proteins involved in insect subzero-temperature tolerance: ice nucleators and antifreeze proteins, pp. 94 Ð127. In R. E. Lee Jr. and D. L. Denlinger (eds.), Insects at low temperature, 1st ed. Chapman & Hall, New York. Fields, P. G. 1992. The control of stored-product insects and mites with extreme temperatures. J. Stored Prod. Res. 28: 89 Ð118. Fields, P. G., and N.D.G. White. 2002. Alternatives to methyl bromide treatments for stored-product and quarantine insects. Annu. Rev. Entomol. 47: 331Ð351. Flinn, P. W., D. W. Hagstrum, W. E. Muir, and K. Sadayappa. 1992. Spatial model for simulating changes in temperature and insect population dynamics in stored grain. Environ. Entomol. 21: 1351Ð1356. Flinn, P. W., D. W. Hagstrum, and W. E. Muir. 1997. Effects of time of aeration, bin size, and latitude on insect populations in stored wheat: a simulation study. J. Econ. Entomol. 90: 646 Ð 651. Hagstrum, D. W., and P. W. Flinn. 1994. Survival of Rhyzopertha dominica (Coleoptera: Bostrichidae) in stored wheat under fall and winter temperature conditions. Environ. Entomol. 23: 390 Ð395. Hamid, A. 1990. Static pressure drop as affected by Indian meal moth webbing in corn. MS thesis, Department of Agricultural Engineering, Kansas State University, Manhattan, KS. Harein, P. K., and Bh. Subramanyam. 1990. Controlling Indianmeal moths in shelled corn and soybeans. University of Minnesota, St. Paul, MN. Kaliyan, N. 2004. Using low temperatures to control stored grain insects. MS thesis, University of Minnesota, St. Paul, MN. Lam, W., and L. P. Pedigo. 2000. A predictive model for the survival of overwintering bean leaf beetles (Coleoptera: Chrysomelidae). Environ. Entomol. 29: 800 Ð 806. Leather, S. R., K.F.A. Walters, and J. S. Bale. 1993. The ecology of insect overwintering. Cambridge University Press, Cambridge, UK. Lee, R. E., Jr. 1991. Principles of insect low temperature tolerance, pp. 17Ð 46. In R. E. Lee Jr. and D. L. Denlinger (eds.), Insects at low temperature, 1st ed. Chapman & Hall, New York. Lee, R. E., Jr., J. M. Strong-Gunderson, M. R. Lee, and E. C. Davidson. 1992. Ice-nucleating active bacteria de-
855
crease the cold-hardiness of stored grain insects. J. Econ. Entomol. 85: 371Ð374. Lewthwaite, S. E., P. R. Dentener, S. M. Alexander, K. V. Bennett, D. J. Rogers, J. H. Maindonald, and P. G. Connolly. 1998. High temperature and cold storage treatments to control Indian meal moth, Plodia interpunctella (Hu¨ bner). J. Stored Prod. Res. 34: 141Ð150. MacAnova. 2002. MacAnova for Windows, version 4.12. University of Minnesota, School of Statistics, St. Paul, MN. Mansbridge, G. H. 1936. A note on the resistance to prolonged cold of some insect pests of stored products. Proc. R. Entomol. Soc. Lond. A. 11: 83Ð 86. Mason, L. J., and C. A. Strait. 1998. Stored product integrated pest management with extreme temperatures, pp. 141Ð177. In G. J. Hallman and D. L. Denlinger (eds.), Temperature sensitivity in insects and applications in integrated pest management. Westview, Boulder, CO. Na, J. H., and M. I. Ryoo. 2000. The inßuence of temperature on development of Plodia interpunctella (Lepidoptera: Pyralidae) on dried vegetable commodities. J. Stored Prod. Res. 36: 125Ð129. Naeemullah, M., K. Tanaka, H. Tsumuki, and M. Takeda. 1999. Relationship of cold tolerance to developmental determination in the Indian meal moth, Plodia interpunctella (Lepidoptera: Phycitidae). Appl. Entomol. Zool. 34: 267Ð276. Salt, R. W. 1936. Studies on the freezing process in insects. University of Minnesota, Agricultural Experiment Station, St. Paul, MN. Salt, R. W. 1961. Principles of insect cold-hardiness. Annu. Rev. Entomol. 6: 55Ð74. SAS Institute. 1995. UserÕs manual, version 6, 4th ed. SAS Institute, Cary, NC. Sauer, D. B., R. A. Meronuck, and C. M. Christensen. 1992. Microßora, pp. 313Ð340. In D. B. Sauer (ed.), Storage of cereal grains and their products, 4th ed. American Association of Cereal Chemists, St. Paul, MN. Solomon, M. E., and B. E. Adamson. 1955. The powers of survival of storage and domestic pests under winter conditions in Britain. Bull Entomol. Res. 46: 311Ð355. Sømme, L. 1982. Supercooling and winter survival in terrestrial arthropods. Comp. Biochem. Physiol. 73A: 519 Ð 543. Tzanakakis, M. E. 1959. An ecological study of the Indianmeal moth Plodia interpunctella (Hu¨ bner) with emphasis on diapause. Hilgardia 29: 205Ð246. White, N.D.G., P. G. Fields, C. Demianyk, D. S. Jayas, and W. E. Muir. 1994. Canadian research on methyl bromide alternatives in post-harvest grain and structural treatments. Proceedings of the Annual International Research Conference on Methyl Bromide Alternatives and Emission Reductions, 13Ð16 November 1994, Orlando, FL. Yaciuk, G., W. E. Muir, and R. N. Sinha. 1975. A simulation model of temperatures in stored grain. J. Agric. Engin. Res. 20: 245Ð258. Zachariassen, K. E. 1985. Physiology of cold tolerance in insects. Physiol. Rev. 65: 799 Ð 832. Zachariassen, K. E., E. Kristiansen, and S. A. Pedersen. 2004. Inorganic ions in cold hardiness. Cryobiology 48: 126 Ð 133. Received for publication 9 June 2005; accepted 12 April 2006.
