Coffee berry borer (Coleoptera: Scolytidae) population dynamics across Hawaii Island's diverse coffee-growing landscape: optimizing location-specific pesticide applications.

A major challenge to area-wide management of coffee berry borer (Hypothenemus hampei Ferrari) (Coleoptera: Scolytidae) is understanding how a heterogeneous coffee-growing landscape affects coffee berry borer population dynamics across temporal and spatial scales. We examined coffee phenology, weather, coffee berry borer flight activity, infestation, coffee berry borer position within the fruit, and management across 14 commercial coffee farms from 2016 to 2018 on Hawaii Island to characterize variation among districts and elevations. Here we aim to determine whether the timing of pesticide applications might be optimized based on specific locations. We observed larger populations of coffee berry borer at low-elevation farms and in the Kona district compared to mid- and high-elevation farms and the Ka'u district. Temperature, relative humidity, and rainfall all differed significantly across districts and elevations. We also observed a trend of higher fruit production at low-elevation farms compared to high-elevation farms, and differences in the timing of fruit development. Infestation increased with higher pest pressure and air temperatures and reduced fruit availability early and late in the season. Lastly, the timing and number of management interventions varied among districts and elevations. Combining information on trap catch, infestation, coffee berry borer position, and plant phenology, we present an optimized pesticide spray schedule for each location and find that the number of sprays could be reduced by 33-75% in comparison to the existing integrated pest management recommendations while maintaining effective control. Implementing a coordinated area-wide approach refined by small-scale optimization will lead to improved management of coffee berry borer on individual farms and a reduction in pest pressure across the coffee-growing landscape.

Less is more: Fewer attract-and-kill sites improve the male annihilation technique against Bactrocera dorsalis (Diptera: Tephritidae).

The Male Annihilation Technique (also termed the Male Attraction Technique; "MAT") is often used to eradicate pestiferous tephritid fruit flies, such as Bactrocera dorsalis (Hendel). MAT involves the application of male-specific attractants combined with an insecticide in spots or stations across an area to reduce the male population to such a low level that suppression or eradication is achieved. Currently, implementations of MAT in California and Florida targeting B. dorsalis utilize the male attractant methyl eugenol (ME) accompanied with a toxicant, such as spinosad, mixed into a waxy, inert emulsion STATIC ME (termed here "SPLAT-MAT-ME"). While highly effective against ME-responding species, such applications are expensive owing largely to the high cost of the carrier matrix and labor for application. Until recently the accepted protocol called for the application of approximately 230 SPLAT-MAT-ME spots per km2; however, findings from Hawaii suggest a lower density may be more effective. The present study adopted the methods of that earlier work and estimated kill rates of released B. dorsalis under varying spot densities in areas of California and Florida that have had recent incursions of this invasive species. Specifically, we directly compared trap captures of sterilized marked B. dorsalis males released in different plots under three experimental SPLAT-MAT-ME densities (50, 110, and 230 per km2) in Huntington Beach, CA; Anaheim, CA; and Sarasota-Bradenton, FL. The plots with a density of 110 sites per km2 had a significantly higher recapture proportion than plots with 50 or 230 sites per km2. This result suggests that large amounts of male attractant may reduce the ability of males to locate the source of the odor, thus lowering kill rates and the effectiveness of eradication efforts. Eradication programs would directly benefit from reduced costs and improved eradication effectiveness by reducing the application density of SPLAT-MAT-ME.

Review of surveillance systems for tephritid fruit fly threats in Australia, New Zealand, and the United States.

Many countries conduct fruit fly surveillance but, while there are guidelines, practices vary widely. This review of some countries in the Pacific region demonstrates the diversity of fruit fly surveillance practices. All utilize 3 parapheromones-trimedlure, cuelure, and methyl eugenol-to trap adult male fruit flies. Some target species are not attracted to these compounds so other attractants such as food-based lures are used in certain areas or circumstances. Lure loading and replacement cycles depend on the target species and the local climate. Malathion and dichlorvos (DDVP) are commonly used toxicants, but not in all countries, and other toxicants are being developed to replace these older-generation pesticides. Jackson and Lynfield are commonly used trap designs but newer designs such as cone and Biotrap are being adopted. Local factors such as chemical registrations and climate affect the choice of trap, lure, dispenser, toxicant, and bait concentration. These choices affect the efficacy of traps, in turn influencing optimal trap deployment in space and time. Most states now follow similar practices around trap inspection, servicing, and data handling, but these processes will be disrupted by emerging automated trap technologies. Ultimately, different practices can be attributed to the unique fruit fly risk profiles faced by each state, particularly the suite of fruit flies already present and those that threaten from nearby. Despite the diversity of approaches, international trade in fruit continues with the assurance that fruit fly surveillance practices evolve and improve according to each country's risk profile and incursion experience.

Previously introduced braconid parasitoids target recent olive fruit fly (Bactrocera oleae) invaders in Hawai'i.

The olive fruit fly Bactrocera oleae (Diptera: Tephritidae) was detected on Maui and Hawai'i Islands in 2019, affecting yields and quality of the state's emerging olive oil industry. Given previous parasitoid releases to control other invasive frugivorous tephritids in Hawai'i, we were interested in determining whether these parasitoids were naturally targeting recent olive fly invaders in field, if local olive cultivar differences affected parasitization rates, and if there was a seasonal pattern of parasitization that could inform future management decisions. To address these questions, we collected data from olive growing in Hawai'i during 2021 and 2022. During the fruiting season we collected monthly samples and reared out B. oleae in the lab. We detected two previously introduced braconid wasps: first Diachasmimorpha tryoni during 2021 and 2022 and later Fopius arisanus during the 2022 collection. Cultivar effects were limited to a single site in our study, where more D. tryoni were reared from 'Arbequina' olives. Seasonality of olive fruit fly and parasitoid activity was earlier in lower elevation sites, as expected based on tree phenology and temperature-dependent insect development. This represents the first report of D. tryoni parasitism activity against B. oleae and may reflect elevational effects combined with the ecological complexity in interactions between multiple invasive arthropod pests, their invasive and cultivated plant hosts, and introduced braconid parasitoids.

Simulation to investigate site-based monitoring of pest insect species for trade.

Pest insect surveillance using lures is widely used to support market access requirements for traded articles that are hosts or carriers of quarantine pests. Modeling has been used extensively to guide the design of surveillance to support pest free area claims but is less commonly applied to provide confidence in pest freedom or low pest prevalence within sites registered for trade. Site-based surveillance typically needs to detect pests that are already present in the site or that may be entering the site from surrounding areas. We assessed the ability of site-based surveillance strategies to detect pests originating from within or outside the registered site using a probabilistic trapping network simulation model with random-walk insect movement and biologically realistic parameters. For a given release size, time-dependent detection probability was primarily determined by trap density and lure attractiveness, whereas mean step size (daily dispersal) had limited effect. Results were robust to site shape and size. For pests already within the site, detection was most sensitive using regularly spaced traps. Perimeter traps performed best for detecting pests moving into the site, although the importance of trap arrangement decreased with time from release, and random trap placement performed relatively well compared to regularly spaced traps. High detection probabilities were achievable within 7 days using realistic values for lure attractiveness and trap density. These findings, together with the modeling approach, can guide the development of internationally agreed principles for designing site-based surveillance of lure-attractant pests that is calibrated against the risk of non-detection.

Transect-based trapping for area-wide delimitation of insects.

Typical delimitation trapping survey designs for area-wide (nonlocalized) insect populations are regularly spaced grids, and alternative shapes have not been evaluated. We hypothesized that transect-based designs could give similar detection rates with significantly shorter servicing distances. We used the TrapGrid model to investigate novel "trap-sect" designs incorporating crossed, spoked, and parallel lines of traps, comparing them to a regular grid, in single survey and multiple-site scenarios. We calculated minimum servicing distances and simulated mean probabilities of detecting a pest population, judging overall performance of trap network designs using both metrics. For single sites, trap-sect designs reduced service distances by 65-89%, and most had similar detection probabilities as the regular grid. Kernel-smoothed intensity plots indicated that the best performing trap-sect designs distributed traps more fully across the area. With multiple sites (3 side by side), results depended on insect dispersal ability. All designs performed similarly in terms of detection for highly mobile insects, suggesting that designs minimizing service distances would be best for such pests. For less mobile pests the best trap-sect designs had 4-6 parallel lines, or 8 spokes, which reduced servicing distances by 33-50%. Comparisons of hypothetical trap-sect arrays to real program trap locations for 2 pests demonstrated that the novel designs reduced both trap numbers and service distances, with little differences in mean nearest trap distance to random pest locations. Trap-sect designs in delimitation surveys could reduce costs and increase program flexibility without harming the ability to detect populations.

Harmonic radar tracking of individual melon flies, Zeugodacus cucurbitae, in Hawaii: Determining movement parameters in cage and field settings.

Tephritid fruit flies, such as the melon fly, Zeugodacus cucurbitae, are major horticultural pests worldwide and pose invasion risks due primarily to international trade. Determining movement parameters for fruit flies is critical to effective surveillance and control strategies, from setting quarantine boundaries after incursions to development of agent-based models for management. While mark-release-recapture, flight mills, and visual observations have been used to study tephritid movement, none of these techniques give a full picture of fruit fly movement in nature. Tracking tagged flies offers an alternative method which has the potential to observe individual fly movements in the field, mirroring studies conducted by ecologists on larger animals. In this study, harmonic radar (HR) tags were fabricated using superelastic nitinol wire which is light (tags weighed less than 1 mg), flexible, and does not tangle. Flight tests with wild melon flies showed no obvious adverse effects of HR tag attachment. Subsequent experiments successfully tracked HR tagged flies in large field cages, a papaya field, and open parkland. Unexpectedly, a majority of tagged flies showed strong flight directional biases with these biases varying between flies, similar to what has been observed in the migratory butterfly Pieris brassicae. In field cage experiments, 30 of the 36 flies observed (83%) showed directionally biased flights while similar biases were observed in roughly half the flies tracked in a papaya field. Turning angles from both cage and field experiments were non-random and indicate a strong bias toward continued "forward" movement. At least some of the observed direction bias can be explained by wind direction with a correlation observed between collective melon fly flight directions in field cage, papaya field, and open field experiments. However, individual mean flight directions coincided with the observed wind direction for only 9 out of the 25 flies in the cage experiment and half of the flies in the papaya field, suggesting wind is unlikely to be the only factor affecting flight direction. Individual flight distances (meters per flight) differed between the field cage, papaya field, and open field experiments with longer mean step-distances observed in the open field. Data on flight directionality and step-distances determined in this study might assist in the development of more effective control and better parametrize models of pest tephritid fruit fly movement.

Simulation-based evaluation of two insect trapping grids for delimitation surveys.

In the United States of America, delimitation trapping surveys with square grids have been used for decades for exotic insects without rigorous evaluation. We used simulations to investigate the effectiveness of two representative designs: an 8-km grid for Acrolepiopsis assectella (leek moth) and a 14.5-km grid for Ceratitis capitata (Mediterranean fruit fly, "Medfly"). We investigated grid compositions and design factors, measuring performance as the mean probability of pest capture over all traps, p(capture), and designed improved grids for both species. For the standard designs, p(capture) was 0.86 for leek moth and 0.71 for Medfly, with the latter performing better due to greater lure and trap attractiveness. For both designs, 86 percent or more of mean p(capture) came from core area captures. Egress testing indicated that both grids were oversized. An improved grid for leek moths would use 177 traps in a 4.8-km diameter circle, which had mean p(capture) = 0.73 and reduced the cost by 80 percent. The best Medfly grid was a 4.8-km diameter circle with 232 traps, which gave mean p(capture) of 0.66 and reduced the cost by 86 percent. Simulation may be used to improve trapping survey plans, often saving significantly on costs while maintaining survey performance.

Mating Competitiveness of Bactrocera dorsalis (Diptera: Tephritidae) Males From a Genetic Sexing Strain: Effects of Overflooding Ratio and Released Females.

The oriental fruit fly, Bactrocera dorsalis (Hendel), is a global pest that infests a range of fruit and vegetables. Males are attracted to methyl eugenol, and control is often achieved by the Male Annihilation Technique, where methyl eugenol + insecticide dispensers are deployed to eliminate males, preclude matings, and reduce population growth. The Sterile Insect Technique (SIT) has also been used to control B. dorsalis. The SIT involves the release of mass-reared, sterilized males to achieve matings with wild females, who then produce inviable eggs. Two key elements of SIT include the overflooding ratio achieved (sterile: wild males) and the strain type utilized, namely bisexual or genetically sexed (allowing male-only releases). Here, we describe the effects of these two factors on the mating competitiveness of a males from a genetic sexing strain of B. dorsalis, termed DTWP. Mating success was scored for DTWP versus wild males in field cages at ratios of 1:2, 1:1, 2:1, and 10:1 both when DTWP females were or were not concurrently released with DTWP males. Close correspondence was found between observed numbers of matings of particular male-female combinations and expected numbers based on the numbers of flies released of each sex and each strain. As a result, the proportion of total matings achieved by the DTWP across the eight treatments showed a corresponding increase with overflooding ratio. At a given ratio, DTWP males had a higher relative mating success when DTWP females were absent rather than present, although the reason for this was unclear.

Evaluating Bactrocera dorsalis (Hendel) (Diptera: Tephritidae) Response to Methyl Eugenol: Comparison of Three Common Bioassay Methods.

Insect responses to chemical attractants are often measured using olfactory bioassays prior to testing in field experiments. The attraction of sexually mature male Bactrocera dorsalis to methyl eugenol (ME) and the loss of attraction by ME pre-fed males have been demonstrated in laboratory bioassays and field trapping studies. It has been suggested that ME nonresponsiveness can be exploited to improve the effectiveness of B. dorsalis management programs by protecting sterile males from ME-based control measures. Currently, work is underway to identify alternatives that reduce or eliminate ME response. To support the development of compounds and evaluation of their effect on B. dorsalis attraction to ME, we compared the effectiveness of three common bioassay methods that have been used to measure lure response in Bactrocera flies under controlled conditions (choice assays using Y-tube [Y], small-cage arena [SC], and rotating carousel field-cage [RC]) to determine which bioassay method is efficient and reliable. A series of bioassays comparing ME-exposed and ME-naïve wild-type and genetic sexing strain males showed that the RC and SC were effective at both observing attraction to ME and detecting a significant reduction in ME response from ME-exposed males. However, the male attraction to ME and a significant decrease in response to ME after ME feeding was not observed in our Y-tube assays. These suggest that RC and SC are preferable options to evaluate ME non-responsiveness in B. dorsalis, and that Y-tube tests are difficult to administer correctly.

Influence of seasonal and climatic variables on coffee berry borer (Hypothenemus hampei Ferrari) flight activity in Hawaii.

Coffee berry borer (CBB, Hypothenemus hampei Ferrari) is the most serious insect pest of coffee worldwide, yet little is known about the effect that weather variables have on CBB flight activity. We sampled flying female CBB adults bi-weekly over a three-year period using red funnel traps baited with an alcohol lure at 14 commercial coffee farms on Hawaii Island to characterize seasonal phenology and the relationship between flight activity and five weather variables. We captured almost 5 million scolytid beetles during the sampling period, with 81-93% of the trap catch comprised of CBB. Of the captured non-target beetles, the majority were tropical nut borer, black twig borer and a species of Cryphalus. Two major flight events were consistent across all three years: an initial emergence from January-April that coincided with early fruit development and a second flight during the harvest season from September-December. A generalized additive mixed model (GAMM) revealed that mean daily air temperature had a highly significant positive correlation with CBB flight; most flight events occurred between 20-26°C. Mean daily solar radiation also had a significant positive relationship with flight. Flight was positively correlated with maximum daily relative humidity at values below ~94%, and cumulative rainfall up to 100 mm; flight was also positively correlated with maximum daily wind speeds up to ~2.5 m/s, after which activity declined. Our findings provide important insight into CBB flight patterns across a highly variable landscape and can serve as a starting point for the development of flight prediction models.

Simulation-Based Investigation of the Performance of Delimiting Trapping Surveys for Insect Pests.

Fully trapped survey designs are widely used to delimit adventive pests populations that can be detected using traps and lures. Delimitation includes verifying the presence of the pest and determining its spatial extent. The size and shape of the survey design and the density of traps can vary; however, resulting variation in detecting efficiency is often unknown. We used a trapping network simulation model with diffusion-based insect movement to investigate delimiting survey trapping design performance for fully trapped and some modified designs. Simulations included randomized outbreak locations in a core area and a duration of 30 d. We assessed impacts of insect dispersal ability, grid size and shape, and trap attractiveness and density on survey performance, measured as mean probability of capturing individual pests [p(capture)]. Most published grids are square, but circles performed equally well and are more efficient. Over different grid sizes, p(capture) increased for insects with greater dispersal ability but was generally unresponsive to size because most captures occurred in central areas. For low dispersing insects, the likelihood of egress was approximately zero with a 3.2-km square grid, whereas an 11.3-km grid was needed to contain highly vagile insects. Trap attractiveness affected p(capture) more strongly than density: lower densities of poorly attractive traps may underperform expectations. Variable density designs demonstrated potential for cost savings but highlighted that resource-intensive outer bands are critical to boundary determination. Results suggesting that many grids are oversized need empirical verification, whereas other principles, such as using circular shapes, are readily adoptable now.

Spinosad resistance in field populations of melon fly, Zeugodacus cucurbitae (Coquillett), in Hawaii.

BACKGROUND: Control of Zeugodacus cucurbitae, a serious agricultural pest worldwide, often includes or is dependent on the use of spinosad-based insecticides. This is especially the case in Hawaii, where GF-120, a protein bait containing spinosad as the active ingredient, has been in use as a key integrated pest management (IPM) tool against this Tephritid for the last two decades. Here, we report on resistance to spinosad [resistance ratios (RRs) and median lethal concentration (LC50 )] in Hawaii's populations of Z. cucurbitae. RESULTS: High resistance was found in populations from three farms on Oahu (RR = 102-303; LC50  = 191-567 mg L-1 ) and in a population from Maui (RR = 8.50; LC50  = 15.9 mg L-1 ). These will be problematic for control given that the most concentrated dilution ratio on the GF-120 label is 96 mg L-1 of spinosad (1 part GF-120 to 1.5 parts water). Background resistance in a naïve wild population from the Island of Hawaii (RR = 2.73; LC50  = 5.1 mg L-1 ) was relatively low compared with a spinosad-susceptible laboratory colony (LC50  = 1.87 mg L-1 ). Resistance in the three Oahu and one Maui populations declined over generations in the absence of spinosad but remained elevated in some cases. Moreover, melon flies collected from one of the Oahu farms 1 year after the cessation of spinosad use revealed high persistence of resistance. CONCLUSION: Compared with a 2008 survey of spinosad resistance, our findings indicate a 34-fold increase in resistance on one of the Oahu farms over 9 years. The evolution and persistence of high levels of resistance to spinosad in Z. cucurbitae in Hawaii highlights the need for alternative control tactics, particularly rotation of active ingredients. © 2021 The Authors. Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Mosquito sound communication: are male swarms loud enough to attract females?

Given the unsurpassed sound sensitivity of mosquitoes among arthropods and the sound source power required for long-range hearing, we investigated the distance over which female mosquitoes detect species-specific cues in the sound of station-keeping mating swarms. A common misunderstanding, that mosquitoes cannot hear at long range because their hearing organs are 'particle-velocity' receptors, has clouded the fact that particle velocity is an intrinsic component of sound whatever the distance to the sound source. We exposed free-flying Anopheles coluzzii females to pre-recorded sounds of male An. coluzzii and An. gambiae s.s. swarms over a range of natural sound levels. Sound levels tested were related to equivalent distances between the female and the swarm for a given number of males, enabling us to infer distances over which females might hear large male swarms. We show that females do not respond to swarm sound up to 48 dB sound pressure level (SPL) and that louder SPLs are not ecologically relevant for a swarm. Considering that swarms are the only mosquito sound source that would be loud enough to be heard at long range, we conclude that inter-mosquito acoustic communication is restricted to close-range pair interactions. We also showed that the sensitivity to sound in free-flying males is much enhanced compared to that of tethered ones.

Coffee Berry Borer (Hypothenemus hampei), a Global Pest of Coffee: Perspectives from Historical and Recent Invasions, and Future Priorities.

Coffee berry borer (Hypothenemus hampei (Ferrari), CBB) has invaded nearly every coffee-producing country in the world, and it is commonly recognized as the most damaging insect pest of coffee. While research has been conducted on this pest in individual coffee-growing regions, new insights may be gained by comparing and contrasting patterns of invasion and response across its global distribution. In this review, we explore the existing literature and focus on common themes in the invasion biology of CBB by examining (1) how it was introduced into each particular region and the response to its invasion, (2) flight activity and infestation patterns, (3) economic impacts, and (4) management strategies. We highlight research conducted over the last ten years in Hawaii as a case study for the development and implementation of an effective integrated pest management (IPM) program for CBB, and also discuss biosecurity issues contributing to incursion and establishment. Potential areas for future research in each of the five major components of CBB IPM (monitoring and sampling, cultural, biological, chemical, and physical controls) are also presented. Finally, we emphasize that outreach efforts are crucial to the successful implementation of CBB IPM programs. Future research programs should strive to include coffee growers as much as possible to ensure that management options are feasible and cost-effective.

Flight Burst Duration as an Indicator of Flight Ability and Physical Fitness in Two Species of Tephritid Fruit Flies.

We introduce a method to quantify flight ability and physical fitness of individual fruit flies which we term 'Flight Burst Duration' (FBD). This consisted of tethering individual insects by the dorsal thorax using a vacuum and measuring the length of time the insect beats its wings while suspended off a surface. Consecutive measurements with Bactrocera dorsalis Hendel (Dipera: Tephritidae) and Zeugodacus cucurbitae Coquillett (Diptera: Tephritidae) in the same day and across days indicated that a single measurement was sufficient, and that FBD was consistent and repeatable. Insects under stress from starvation displayed shorter FBD over time, and we suggest that the measure also relates to the physical condition or survival fitness of the individual. Though somewhat laborious and time-consuming, we propose that FBD can be useful for research studies requiring individual-level phenome data and for obtaining estimates quality and dispersive movement for insects.

Evaluation of Exclusion Netting for Coffee Berry Borer (Hypothenemus Hampei) Management.

Exclusion nets are increasingly being used to protect a variety of agricultural crops from insect pests as a sustainable alternative to chemical controls. We examined the efficacy of exclusion nets in controlling the world's most damaging insect pest of coffee, Hypothenemus hampei (coffee berry borer), on two small-scale coffee farms on Hawai'i Island. We recorded microclimate data, fruit infestation, population per fruit, sex ratio, mortality by Beauveria bassiana, coffee yield and quality in four paired exclusion and control (un-netted) plots on both farms. Mean and maximum daily temperature and relative humidity were similar between treatments, while mean and maximum daily solar radiation was reduced by ~50% in exclusion plots. Green and ripe fruit from exclusion plots had significantly lower infestation compared to un-netted control plots at both farms. We observed no significant difference between exclusion and control plots in the number of CBB per fruit or the female:male sex ratio. CBB mortality was significantly higher in control relative to exclusion plots in one of the two farms. Ripe fruits harvested from exclusion plots were on average significantly heavier and wider than those from control plots; however, there was no significant difference in the average yield per tree between treatments. Lastly, coffee quality was not significantly different between control and exclusion plots. Our results suggest that with complete sanitation prior to net installation in an environment where CBB is actively circulating, exclusion netting can successfully control CBB on small-scale coffee farms without reducing coffee yield or quality, and has the potential to lower production and labor costs by eliminating the need to spray pesticides.

Postharvest Population Reservoirs of Coffee Berry Borer (Coleoptera: Curculionidae) on Hawai'i Island.

Coffee berry borer, Hypothenemus hampei Ferrari (Coleoptera: Curculionidae: Scolytinae), is the most damaging insect pest of coffee worldwide. Old coffee berries (raisins) are widely acknowledged as coffee berry borer reservoirs, yet few studies have attempted to quantify coffee berry borer populations in raisins remaining on farms postharvest. We collected ground and tree raisins at six coffee farms on Hawai'i Island to assess raisin density, infestation, coffee berry borer abundance, and adult mortality in three areas of each farm: trees, driplines (ground below the tree foliage), and center aisles (ground between tree rows). We also assessed infestation of the new season's crop by conducting whole-tree counts of infested green berries. Mean raisin density was significantly higher in the dripline compared to the center aisle and trees (131 vs 17 raisins per m2 and 12 raisins per tree, respectively). Raisin infestation was significantly higher in samples from trees (70%) relative to those from the dripline (22%) and center aisle (18%). Tree raisins had significantly higher coffee berry borer abundance compared to both areas of the ground (20 vs 3-5 coffee berry borer per raisin). Adult mortality was significantly higher on the ground (63-71%) compared to the trees (12%). We also observed a significant positive correlation between ground raisin density and infestation of the new season's crop. Across all farms, we estimated that 49.5% of the total coffee berry borer load was present in dripline raisins, 47.3% in tree raisins, and 3.2% in center aisle raisins. Our findings confirm the importance of whole-farm sanitation in coffee berry borer management by demonstrating the negative impact that poor postharvest control can have on the following season's crop.

Coffee berry borer (Hypothenemus hampei) (Coleoptera: Curculionidae) development across an elevational gradient on Hawai'i Island: Applying laboratory degree-day predictions to natural field populations.

Coffee berry borer (CBB, Hypothenemus hampei) (Coleoptera: Curculionidae: Scolytinae) is the most destructive pest of coffee worldwide. Information on CBB development times can be used to predict the initiation of new infestation cycles early in the coffee-growing season and thus inform the timing of insecticide applications. While laboratory estimates of CBB development under constant conditions exist, they have not been applied under the heterogeneous environmental conditions that characterize many coffee-growing regions. We measured CBB development times and abundance in commercial coffee farms across an elevational gradient on Hawai'i Island and applied thermal accumulation models from previous laboratory studies to test their fit to field data. Artificial lures were used to infest coffee berries at five farms ranging in elevation from 279-792 m, and weather variables were monitored at macro (farm-level) and micro (branch-level) scales. CBB development was followed in the field from the time of initial berry infestation by the founding female through the development of F1 mature adults. Mean development time from egg to adult across all sites was 38.5 ± 3.46 days, while the mean time required for the completion of a full life cycle (from time of infestation to presence of mature F1 females) was 50.9 ± 3.35 days. Development time increased with increasing elevation and decreasing temperature. Using macro-scale temperature data and two different estimates for the lower temperature threshold (14.9°C and 13.9°C), we estimated a mean requirement of 332 ± 14 degree-days and 386 ± 16 degree-days, respectively, from the time of berry infestation to the initiation of a new reproductive cycle in mature coffee berries. Similar estimates were obtained using micro-scale temperature data, indicating that macro-scale temperature monitoring is sufficient for life-cycle prediction. We also present a model relating elevation to number of CBB generations per month. Our findings suggest that CBB development times from laboratory studies are generally applicable to field conditions on Hawai'i Island and can be used as a decision support tool to improve IPM strategies for this worldwide pest of coffee.