Performing Department
(N/A)
Non Technical Summary
This is a proposal for a standard, multi-state, multi-institutional project, international partnership requesting $799,976, that aims to improve border biosecurity for two important stored product quarantine pests, khapra beetle (KB) and larger grain borer (LAGB). Historical establishment of KB in the USA led the government to spend $125 million to eradicate the pest in 1953. The LAGB has been found crossing the southern border from its normal distribution in south-central America. There have been increasing interceptions of both pests in the last 20 years. However, there are deficiencies with current traps, and a need for understanding where to target surveillance and mitigation. With the expected phase-out of methyl bromide for fumigation, even for quarantine, there is a need for updated USA disinfestation recommendations. Our objectives include (1) optimizing sulfuryl fluoride application and other emerging technologies, (2) upgrading and tailoring enhanced bio-surveillance systems for real-time monitoring, and finally (3) predicting theinvasion risk using geographic thermal profiles and ecological variation of KB and LAGB. We will communicate the importance of novel tools to stakeholders, while improving quarantine procedures. Our proposal includes personnel representing Kansas,Massachusetts, and Greece, including partnerships with APHIS, and international institutions to speed biosecurity solutions. This will protect the USA's role as the world's breadbasket at a time when there is a global food crisis, supporting the USDA Program Area PHPPP and Strategic Goal 1 and 4, and helping illuminate the effects of climate change on these two important quarantine pests and increase the supply of nutritious food.
Animal Health Component
0%
Research Effort Categories
Basic
40%
Applied
60%
Developmental
0%
Goals / Objectives
This proposal aims to improve border biosecurity for two important stored product quarantine pests, khapra beetle (KB) and larger grain borer (LAGB). Historical establishment of KB in the USA led the government to spend $125 million to eradicate the pest in 1953. The LAGB has been found crossing the southern border from its normal distribution in south-central America. There have been increasing interceptions of both pests in the last 20 years. However, there are deficiencies with current traps, and a need for understanding where to target surveillance and mitigation.With the expected phase-out of methyl bromide for fumigation, even for quarantine, there is a need for updated USA disinfestation recommendations. Our detailed objectives include:Obj. 1A Evaluate susceptibility of LAGB to sulfuryl fluoride (SF), phosphine (PH3) and other potential emerging fumigation technologies. Obj. 1B Obtain baseline data on controlled fumigation depending on container gas-tightness, spatial and temporal distribution of fumigant gas concentration in empty shipping containers. Obj. 1C Validate field fumigation on LAGB and KB mortality from bioassays in empty, partially filled, and full shipping containers. Objective 2A. Upgrade and tailor enhanced biosurveillance systems for real-time monitoring of LAGB and KB.Obj. 2B. Using behaviorally-compatible traps, determine reliability in distinguishing LAGB and KB from other stored product insects. Obj. 2C. Validate automated, behaviorally-compatible traps in the field for reliability in detecting LAGB and KB at food facilities. Objective 3A. Predict the invasion risk of LAGB and KB using thermal profiles and ecological variation. Establish LAGB and KB life history thermal reactions norms and use them to predict population peaks with current and predicted weather patterns.Obj. 3B. Model movement of LAGB and KB across a variety of landscapes and human-mediated interactions.Obj. 3C Implement community ecology, movement, thermal profiles, and ecosystem variation in interception risk models.Obj. 4 Rigorously engage US biosecurity stakeholders to transfer knowledge about mitigation, surveillance, and prediction of KB and LAGB. This information will be transferred to end users through conventional outputs as well as via innovative, interactive digital tools.
Project Methods
1. Fumigation trials will be performed in empty, partially filled, and filled 20-ft shipping containers maintained the USDA-ARS facitlity 1 mile from KSU. Fumigations will be conducted over a six-month period when temperatures are consistently >16-20°C. Each trial will last about 7 d, thus multiple experimental trials can be completed during a six-month period. Means and standard errors (SE) will be calculated for the percentage of alive, affected, and dead adults initially and percent recovery will be calculated. The percent mortality will be plotted in 3D mapping software, along with gas distribution readings2. Experiments with remote, automated, real-time traps in the laboratory and semi-field to determine the behaviorally-compatible traps for LAGB and KB will have two trials. Treatments will consist of automated, remote traps, including a modified SmartProbe H (AIVision Food, Davis, CA, USA), a Sight Trap (Insects Limited, Westfield, IN, USA) compared with a 4-funnel Lindgren trap (Great Lakes IPM, Vestaburg, MI, USA; current standard trap for LAGB), and the APHIS Wall Trap (4165-01, Trece, Inc.; current standard for KB). Traps will be baited with either a lure containing LAGB aggregation pheromone (bullet lure, IL-953, Insects Limited) or the KB sex pheromone, trogodermal (KB/WB lures, 3155-25, Trece, Inc., Adair, OK, USA) plus a couple drops of the food kairomone oil (Storgard™ Oil, Trece, Inc.).3. We will then use only two commercial standards and optimal behaviorally-compatible traps for LAGB and KB. A semi-field test will be conducted in pilot-scale warehouses (5.85 × 2.81 × 2.0 m L:W:H) in Kansas and/or warehouse rooms (5.85 × 3.90 × 3 m, L:W:H) in Velastino, Greece. A warehouse will be randomly assigned to one of the trapping treatments, and then four traps will be placed in each of the corners of the room. Five hundred LAGB or KB will be released in the center of each space. Pictures from remote traps will be automatically taken daily and sent to the cloud for storage. Each trap will also be manually checked on a weekly basis for a month. There will be n = 10 replicates per trap and these assays will be done in three successive months during the active period of the year. Trap captures in the release-recapture assays will be analyzed with a GLM for each species based on Poisson or Quasipoisson distribution (depending on if overdispersion is a problem). Trap type will be the explanatory variable, and upon a significant result, Tukey HSD will be used for multiple comparisons.4. We will determine the reliability in distinguishing LAGB and KB from other stored product insects. In the first trial, using an AI-powered, machine-learning approach, LAGB and KB will be distinguished from morphologically similar species, including R. dominica and other dermestids (e.g., T. variabile). In the laboratory, images will be generated for a training dataset of the focal species of concern from multiple angles in replicate at high and low resolution and from close-up and wide field, singly, and in groups. This will be done, in part, using a 3D imaging system. Light will be diffused using a partially cut frosted plastic jar (15.2 × 7.6 cm D:H) making a total of n = 60 replicate individuals per lens, orientation, and species. In addition, images will be generated similarly with a wide angle lens (17-40 mm EF 1:4 L Series USM Lens, Canon, Tokyo, Japan). We will compare the performance of four widely used convolutional neural networks (CNN) classification models that vary in complexity of architecture to evaluate tradeoffs in accuracy and speed for stored product quarantine species identification. 5. We will validate automated, behaviorally-compatible traps in the field fordetecting LAGB and KB at food facilities. A field trial will first be conducted at three commercial food processing facilities in Greece using next generation traps validated in prior studies, and compared to negative controls (no stimuli) and positive controls (e.g., wall trap for KB, and 4-funnel Lindgren trap for LAGB). Three traps of each kind will be placed on a floor around the perimeter of a room and replicated on two floors in the food facility. Images from the traps will be captured daily, and traps will be checked once every two weeks for 2 yr. A repeated measures linear mixed model will be used to analyze the captures, and location as a random response variable. The precision of the images in matching identifications of LABG and KB will be determined as well as the false positive rate.6. We will determine LAGB and KB life history thermal reactions norms and use them to predict population peaks with current and predicted weather patterns. We will have two interlocking projects. An initial review of the available data for both LAGB and KB will provide an outline for what additional data are needed to be collected for modeling. Where gaps in thermal ranges occur at 5 temperatures between 20 to 35°C (depending on what is missing from the literature), pairs of males and females of each species will be matched one day post-eclosion from the pupal stage. Sex will be determined by size and color (larger size and lighter color for females) in KB and larger body, abdomen, and distance between clypeal tubercles for female LAGB. Pairs will be placed in 8-dram glass vials prepared with PTFE and a cotton ball at the top and a small amount of rearing diet in each vial. Pairs will be allowed to mate freely and transferred to new vials every two days for KB and every week for LAGB. Survivorship of both the males and females will be recorded at each transfer. All vials having had male-female mating pairs will be saved for progeny output and assessed between 26-30 d for KB and 37-45 d for LAGB.7. We will model movement of LAGB and KB across a variety of landscapes and human-mediated interactions. We will model the movement patterns across natural areas and account for the adaptive transitions to human-made storage environments. For both KB and LAGB we will consider the associations and movement capacity from natural habitats such as forested areas to human-made facilities or structures. A null model will first be developed to consider differences in movement among different locations.An extension of the model will also consider climate change scenarios and how population growth and changing harvesting schedules may impact overall movement patterns in these insects. Overall canopy cover information available from the U.S. Forest Service will be downloaded, as well as any historical occurrence data from museums. Factors associated with movement, such as wind speed and direction for LAGB will be added to our model. Connectivity factors such as trucking and train shipping routes will also be assessed. In addition, waterways such as ports along the coastal region of the USA will be evaluated for interception data. Predicted changes in climate will also be considered for these models.8. Implement community ecology, movement, thermal profiles, and ecosystem variation in interception risk models. Agricultural economics can also be implemented in this population stacking model as species abundance models. All of these models can have impacts on both a local scale or on a larger scale based the question of the producer, manager, or researcher. Current monitoring efforts on both LAGB and KB can be leveraged for population genetics work to implement markers in movement and population growth models. Machine learning models will be implemented using Python and Jupyter Notebook.