Experimental Insects.The field colonies of T. castaneum, R. dominica and the rice weevil, Sitophilus oryzae (Coleoptera: Curculionidae), were used on this study. For all species, four to eight-week-old adults were used. Cultures of strains of T. castaneum (collected from eastern KS), R. dominica (from Pottawatomie County, KS), and S. oryzae (from eastern KS) have been maintained in the laboratory since 2012, 2019, and 2012 respectively at the USDA Center for Grain Animal Health Research in Manhattan, KS. Tribolium castaneum was reared on a mixture of 95% unbleached, organic flour and 5% brewer’s yeast, while R. dominica and S. oryzae were reared on tempered organic whole wheat. The colonies were maintained at 25–27.5°C, 65% RH, and 14:10 or 16:8 (L:D) h photoperiod.TreatmentsIn each grain bin, a total of 60, 7.57-L (2 gal.) capacity buckets (hereafter, miniature silos or silos) were each filled with 500 g of clean wheat (20% cracked grain, containing 10.8% grain moisture) and assigned to the floor of a grain bin. Each of the miniature silos had holes drilled every ~7 cm around the circumference of the base and a 12.7 × 12.7 cm square cut out of the lid. The entire outer surface of the miniature silos was roughened up with sandpaper to provide an easy climbing substrate for insects. Twenty miniature silos employed Carifend® insecticide-incorporated netting (0.34% alpha-cypermethrin at 163.2 mg/m2 active ingredient (a.i.), 40 deniers, 100 holes/cm2; BASF, Ludwigshafen, Germany) covering the holes and gaps attached inside with a hot glue gun, while twenty of the miniature silos used control netting (physical identical with Carifend® net without insecticide; Casa Collection, Mesh White, 1721-9668; Jo-Ann’s, Hudson, OH, USA), and the final twenty miniature silos lacked netting completely.A total of three different 110-MT grain bins were used in Manhattan, KS. Each grain bin was divided into quarters and in each, we randomly placed 15 miniature silos with 5 silos of each treatment represented in each quadrant. On a monthly basis from June–September 2022 and the experiment was completely replicated from June–October 2023, while unmanaged stored product insect populations were supplemented by additional releases of insects. In each quarter of the grain bin, a release point for insects was randomly chosen, and 25 each of T. castaneum, R. dominica, and S. oryzae were released on a monthly basis. Thus, a total of 300 T. castaneum, R. dominica, and S. oryzae were released each month in each grain bin. Dataloggers (UX100-011A Hobo Temp/RH logger, Onset, Bourne, MA, USA) were also placed in inside the grain of a miniature silos in each quadrant of the grain bin and were set up to record the temperature and RH every 10 min from June–September. Dataloggers were placed on the top of grain inside miniature silos (hereafter, inside grain). Two additional data loggers were also placed on the north- and south-facing walls of the grain bins at a height of 150 cm (hereafter, inside bin).Sampling ProcedureMonthly samples of 100 g of grain were taken from July–October in 2022 and 2023 and coincided with additional releases of insects as described above. During each monthly sample, 100 g were taken from four miniature silos belonging to each treatment (e.g., BASF Carifend® LLIN, positive control, and negative control), which were scooped in plastic containers (5 × 11 cm D: H) to a pre-measured fill line demarcating 100 g. Samples were added to pre-labeled Ziplock bags and immediately brought back to the laboratory. Samples were sieved using two sieves (#10 sieve, 2.0 × 2.0 mm mesh, W.S. Tyler, Mentor, OH; and #20 sieve, 0.841 × 0.841 mm mesh; W.S. Tyler, Cleveland, OH) and the types of insect species dispersing into the grain, the number of individuals belonging to each species, and their life stages were recorded. The health condition of each adult was classified as alive, affected, or dead (according to Morrison et al. 2018). Grain was held in individual containers (5 × 11 cm D: H) for an additional 6 wk at 27.5 °C, 65% RH, and 16:8 L:D in an environmental chamber (Percival, Perry, IA, USA) to determine progeny production. Grain quality measures were also assessed at initial collection, including the number of insect-damaged kernels (IDK), weight of damaged and undamaged grain, and mold rating (using the procedure developed in Van Winkle et al. 2022).Prior to the study, half of the miniature silos in a bin were randomly assigned to a possible phosphine fumigation management treatment if above the threshold, whereas half remained unfumigated regardless of insect pressure. For the fumigation management treatment, fumigation was triggered when the number of IDK and/or the number of insects found in the equivalent of 100 g was at the Federal Grain Inspection Service (FGIS) specified tolerances of two live insects, or conservatively 16 IDKs, which is half the tolerance for when a lot is considered “sample grade” or unfit for human consumption by USDA .Phosphine Fumigation ProcessOnce insect infestation levels triggered a fumigation event, the affected miniature silos were moved to a dedicated 110-MT fumigation-only grain bin. The fumigation treatments were performed in 55-gallon (~208 L) barrels which were filled with 9 miniature silos (at maximum) containing sample wheat. Miniature silos were carefully and individually hand-swaddled with a clear plastic bag to ensure no loss of grain in transit between experimental location and fumigation bin. Aluminum phosphide pellets (Deitia Degesch AG, Laudenbach, Germany) were used. For each barrel, two pellets (0.6g per pellet) were placed onto a disposable plate holding a damp paper towel. A plate was set inside each barrel and the barrels were covered with lids that contained rubber gaskets and a clamping ring which was used to hold the lid tightly to the top of the barrel. The lids had several ports which were used for gas sampling, input-air, and exhaust-air during ventilation. The phosphine concentration inside each sealed barrel reached ~1000–1200 ppm after 8 hours and was maintained for ~ 4 days. The average concentration * time (CT) product values were > 100,000 ppm*hr, which is considered a very strong fumigation and able to kill phosphine resistant species (Brabec et al, 2021). The phosphine concentration and temperature were monitored in each barrel with wifi phosphine sensors (Centaur Analytics, Ventura, CA). The wifi sensors collected data every 1-3 hours over the 4-day fumigation period. A control sensor was included that was located directly adjacent to the barrels to ensure there was no leakage. In addition, phosphine concentration measurements were taken ~daily with a hand-held meter (X-am 5000 multi-gas detector, Dräger, Lübeck, Germany). The Drager hand-held meter was used as a calibration reference for the wifi phosphine sensors. After the 4-day fumigation treatment, each barrel was carefully vented by flushing with the fresh air for ~20 min to purge the phosphine gas from each barrel and directed out through a small chimney pipe. Then, the lid could safely be removed and the experimental buckets unloaded.Insect Drop TestA testing arena was constructed using a 34.6 × 21 × 12.4 cm (L×W×H) plastic container (Sterilite®, Sterilite Corporation, Townsend, MA) with a 9 cm hole drilled in the center of the container. The hole was covered on the bottom of the container with Carifend® LLIN. The netting was affixed to the bottom of container using adhesive caulking (DAP Kwik Seal, DAP Products Inc., Baltimore, MD) and a plastic funnel was attached below the netting. The top 3-5 cm of the container was coated with Floun® (polytetrafluoroethylene, Sigma-Aldrich Co., St. Louis, MO) and the top of the lid was fitted with 3 cm hole coved with a fine mesh screen to prevent insect escape and allow for airflow. The container was placed above a 0.57 L jar filled with 12-13 cm of insect diet. The entire apparatus was placed inside a 7.57 L bucket as a secondary container.Insects used in this study were obtained from pesticide susceptible lab strains maintained at the USDA Center for Grain Animal Health Research in Manhattan, KS. For all species, 2-3-week-old adults and larvae were used. Adults of 10 stored product species were used. Larval species used were T. castaneum, T. confusum, T. inclusum, T. variabile, and O. surinamensis.Fifty adults or larvae of each species were placed inside one of three testing containers for ~24 h at ambient laboratory temperatures. After 24 h, the number of insects that passed through the netting and found in the diet below the container were recorded. Insects that were partially through the netting or clinging on the underside of the net, were considered as passed through. Each drop test was repeated twice on separate dayd and with separate insect colonies for a total of six independent replicates.