United States military troops in the field are exposed to the environment and are thus at high risk for transmission of arboviruses, and degradation of mission from continual harassment from insects. Passive vector control, such as application of residual insecticides to US military materials common in the field such as tents and camouflage netting, has been shown to be effective and can contribute to a successful integrated vector management (IVM) plan in the field to reduce this risk. However, other common US military field materials have not been evaluated with residual pesticides. In this study we conducted the first known investigation of the efficacy and longevity of a residual pesticide containing λ-cyhalothrin applied to HESCO® blast protection wall geotextile. We exposed treated material to a temperate Florida environment and found that this treatment can be effective against sand flies, filth-breeding flies, and mosquitoes for at least 6 wk. This study provides evidence that residual treatment of this US military material may be leveraged as an IVM component to enhance the US Department of Defense pest management system.
The core of the US Department of Defense (DoD) pest management system consists of individual personal protective measures, such as wearing permethrin-treated uniforms and applying US Environmental Protection Agency–approved repellents to exposed skin, to reduce contact between personnel deployed in the field and disease-vector insects such as mosquitoes or sand flies (Frances and Cooper 2007, Kitchen et al. 2009). Unit-level or installation-level operational pest control measures such as larviciding and adulticiding may also contribute to protecting personnel in the field. However, limited resources result in many locations, in particular smaller and more remote outposts, not being visited frequently enough, if at all, by military entomologists or contract pest management to effectively reduce target insect populations. Unfortunately, small, remote underserved field locations also tend to have the highest proportion of personnel exposed to the environment throughout the day and night, and thus the highest risk for transmission of arboviruses, other arthropod-borne pathogens, and degradation of mission from continual harassment from insects.
One solution to enhance the DoD pest management system is to incorporate techniques of passive insect control with residual pesticides on materials that routinely accompany US troops to the field. Residual pesticides with active ingredients like bifenthrin or λ-cyhalothrin may be applied to military materials organic to field units and ubiquitous across US military field locations, such as camouflage netting (Britch et al. 2010, 2011) or tents (Frances 2007). Direct field observations (Britch, personal observation) indicate that mosquitoes and filth flies rest on materials like camouflage netting or shade cloth in areas where humans tend to congregate outdoors in US military field camps. When appropriate surfaces are treated with residual pesticides containing a nonrepellent active ingredient such as λ-cyhalothrin, target insects readily land on the surface and may accumulate a sublethal or lethal dose while resting that will reduce or eliminate, respectively, their capability to contact humans (Davidson 1955; Kelly et al. 1997; Bauer et al. 2006a, 2006b; Diabate et al. 2006; Orshan et al. 2006; Xue 2008; Peck et al. 2013; Richards et al. 2017).
A prominent military material that has not yet been tested for suitability for residual pesticide application is the nonwoven polypropylene geotextile (Müller and Saathoff 2015) component of the HESCO® MIL™ defensive barrier (HESCO, Lillington, NC; Szabó et al. 2011, HESCO Bastion Ltd. 2018). The HESCO MIL cell consists of a welded heavy steel mesh cage, open at the top and bottom and lined with a heavy feltlike material (geotextile) capable of retaining sand, earth, or rocks. Arrays of linked HESCO MIL cells have been used extensively since the early 2000s to create US military camp perimeters (Fig. 1) and other defensive positions throughout Iraq and Afghanistan. Several attributes of HESCO MIL perimeters make HESCO geotextile a potentially highly appropriate substrate for residual pesticides: When HESCOs are filled, the spaces between cells may provide abundant habitat for rodents; sand flies feeding on these rodents can be targeted as they contact the material entering and leaving these spaces. Preliminary laboratory trials (Aubuchon, unpublished data) indicated that Phlebotomus papatasi Scopoli sand flies preferentially rested on HESCO geotextile compared to other materials common in US military camps, such as plywood and concrete. Also, earth-fill of HESCO perimeters frequently results in a moatlike borrow pit adjacent to the perimeter that may pool stagnant water and create immature habitat for populations of mosquitoes, biting midges, and filth flies. These insects may then rest on the perimeter, and thus be targeted by residual pesticide, as they move towards hosts or other attractants in a military camp. In combination with related integrated vector management (IVM) actions such as personal protection measures, habitat reduction, or space sprays, pesticide-treated barriers of HESCO cells could passively reduce populations of biting vector and nuisance insects.
Our objective in this study was to investigate the longevity and efficacy of a λ-cyhalothrin–based residual pesticide applied to HESCO geotextile and subjected to natural weathering. In a related series of previous studies (Britch et al. 2010, 2011; Britch, Linthicum, and Aldridge, unpublished data), we similarly investigated residual pesticides applied to US military camouflage netting across a range of militarily relevant environments, starting with the warm-temperate environment in north-central Florida to establish baseline capabilities, followed by trials in hot-arid desert and hot-humid tropical environments. For the greatest benefit to potential military operational use of residual pesticides on military materials, it is necessary to understand the range of efficacy against different target insects across a variety of habitats where the material could be deployed. In the present study with pesticide-treated HESCO geotextile, we conducted a baseline investigation of efficacy in a temperate habitat in north Florida, with subsequent trials in a variety of habitats to be described separately.
MATERIALS AND METHODS
On February 8 and 9, 2012, we built a series of 5 HESCO test enclosures, H-1 through H-5, 100 ft (approximately 31 m) apart at the southeast corner (29° 51′ 58.91″ N, 82° 2′ 26.40″ W) of an open field adjacent to longleaf pine (Pinus paulustris P. Miller) canopy and approximately 250 ft (75 m) west of a small, approximately 1-acre (approximately 0.43-ha) pond (Fig. 2). The HESCO enclosure design was loosely based on the camouflage netting enclosures used in Britch et al. (2011) to investigate the efficacy of residual pesticide on camouflage netting. Each HESCO enclosure consisted of 10 HESCO MIL defensive barrier units, 5 lower and 5 upper, arranged as shown in Fig. 3 to create a small space protected on 3 sides and open towards the front. Each MIL unit had an approximately 3-ft × 3-ft (0.9-m × 0.9-m) footprint and was 4 ft (1.23 m) tall, so each assembled enclosure stood 8 ft (2.46 m) high and occupied a footprint 9 ft (2.7 m) wide and 6 ft (1.8 m) deep. The units were left unfilled so to stabilize enclosures we fastened the upper cells to the lower cells with steel hog rings and anchored each upper corner to a large steel pin in the ground with braided steel cable. The HESCO enclosures were situated with 2 treated enclosures along the south perimeter of the pines with openings facing north, 1 untreated enclosure in the corner with its opening facing northwest, and 2 treated enclosures along the east perimeter with openings facing west (Fig. 2).
We applied Demand CS (Syngenta, Greensboro, NC), a residual pesticide containing λ-cyhalothrin, to all exposed sides of the HESCO geotextile in H-1, H-2, H-4, and H-5 enclosures at the maximum label rate in water using an SR 450 backpack sprayer (STIHL, Inc., Virginia Beach, VA) on February 9, 2012 (Fig. 3). Meteorological conditions at the time of spray were approximately 15.6°C (60°F), 80% RH, and 0.9 m/sec (2 mph) wind speed (Fig. 4). We captured meteorological data, including wind speed, rainfall, temperature, and humidity, throughout the field experiment with a 6250 Vantage Vue Wireless Weather Station and recorder (Davis Instruments, Hayward, CA) positioned approximately 300 ft (approximately 100 m) to the west of enclosure H-1 (Fig. 2).
We evaluated the efficacy and longevity of the residual treatment on the HESCO geotextile by comparing mortality with treated and untreated fabric in laboratory bioassays using colony-reared mosquitoes (Culex quinquefasciatus Say), house flies (Musca domestica (L.)), stable flies (Stomoxys calcitrans (L.)), and sand flies (P. papatasi) across a time series of fabric samples cut from the enclosures. We cut a 8-in. × 5-in. (20-cm × 12.5-cm) geotextile fabric sample from the same relative position on an outside-facing side of each of the 5 HESCO enclosures approximately once per week for the 1st month, once every 2 wk for the 2nd month, and once a month thereafter similar to the protocol described in Britch et al. (2011). Each fabric sample was placed separately in a labeled resealable plastic bag and stored at room temperature in the laboratory until use in bioassays.
We conducted bioassays following the protocol and laboratory conditions described in Britch et al. (2011) such that each 8-in. × 5-in. fabric sample from each HESCO enclosure was cut into 5 1-in. × 5-in. strips, in turn placed into 5 glass bioassay tubes for a total of 25 tubes for each time period sampled. The same 5 strips from each of the HESCO enclosures were used for each of the 4 test insect species, and we tested each species in separate bioassay runs. We drew specimens of the 4 test species from the following susceptible colonies reared in the United States Department of Agriculture–Agricultural Research Service–Center for Medical, Agricultural, and Veterinary Entomology insectaries in Gainesville, FL: Cx. quinquefasciatus, 1995 Gainesville strain (with 2003 Ocala strain admixture); M. domestica, 1940s Orlando strain; S. calcitrans, ∼1980 Gainesville strain; P. papatasi, ∼1980 Israeli strain ex. Walter Reed starting 2009 in Gainesville (with 2014 boost from same Walter Reed colony). Mortality in bioassay tubes was tallied at 24 h and 48 h, and we calculated the mean mortality and standard error of the mean (SEM) for each time-period sample across the 4 treated HESCO enclosures corrected with the Abbott formula (Abbott 1925) based on mortality in the 1 untreated HESCO enclosure. We considered the residual treatment highly efficacious at or above an arbitrary benchmark of 90% mortality at 24 h in bioassays, and we set an arbitrary benchmark for minimum efficacy of the treatment at ≥80% but <90% mortality.
We attempted to support estimates of longevity and efficacy from bioassays with the Centers for Disease Control and Prevention (CDC) light trap collections of natural field populations of mosquitoes in the 5 enclosures that coincided with most of the fabric sample cuttings. Our hypothesis was that collections in the untreated enclosure would be higher than the average collection across the 4 treated enclosures. To conduct field collections, a Model 2836BQ CDC mini light trap (Bioquip, Rancho Dominguez, CA) baited with light and CO2 (approximately 2 kg of dry ice) was suspended in the center of the protected space in each of the 5 enclosures as indicated in Fig. 3. We also set 2 additional CDC light traps suspended from 6-ft (1.8-m) shepherd hook poles, one in the immediate vicinity of the HESCO enclosures (“South Trap”) and one to the north in the open area (“North Trap”), to measure mosquito population densities in the general area (Fig. 2). All CDC traps were set in the early afternoon between noon and 1400 h and retrieved the next day between 1000 h and 1300 h. Fabric cuttings were taken either on the day traps were deployed or on the trap retrieval day.
Results from laboratory bioassays on the time series of treated HESCO fabric samples using colony-reared Cx. quinquefasciatus, M. domestica, S. calcitrans, and P. papatasi are shown in 4 charts in Fig. 5. Each chart shows a separate plot for 24-h and 48-h mortality. Data presented in Fig. 5 are mean percent bioassay mortalities (with SEM) across HESCO enclosures H-1, H-2, H-4, and H-5 Abbott-corrected using bioassay mortality observed on untreated control HESCO fabric samples from H-3. Mean 24-h and 48-h bioassay mortalities across all time periods on untreated control HESCO fabric samples used in Abbott corrections were, respectively, 1.5% and 2.1% (Cx. quinquefasciatus), 2.6% and 5.0% (M. domestica), 1.8% and 7.0% (S. calcitrans), and 5.1% and 10.7% (P. papatasi).
Using our benchmark of ≥90% bioassay mortality, the residual pesticide was highly effective against Cx. quinquefasciatus, S. calcitrans, and P. papatasi for at least 41 days posttreatment in the varying cool- to warm-temperate habitat. Limitations of the P. papatasi colony resulted in data from only 5 posttreatment samples (days 41, 103, 124, 211, and 246; Fig. 5). However, given that 24-h mortality at day 103 for P. papatasi was approximately 35%, and day 103 mortality for S. calcitrans at 24 h was 20%, we hypothesize that the waning of efficacy for P. papatasi mirrored that of S. calcitrans and dropped at a similar rate between 55 and 97 days posttreatment. Efficacy dropped to zero by day 103 for Cx. quinquefasciatus, day 211 for P. papatasi, and day 246 for S. calcitrans.
In contrast, the residual treatment did not meet our minimum efficacy benchmark (≥80% but <90%) against M. domestica in bioassays, even in samples from the day of treatment (day 0) which induced only approximately 70% mortality in bioassays at 24 h (Fig. 5). The effect of the residual against M. domestica dropped sharply thereafter to approximately 35% mortality (half of maximum efficacy) at 19 days posttreatment, and to zero efficacy by day 91. However, after the 24-h mortality curve had dropped to zero by day 91, M. domestica mortality showed efficacy above the zero line in samples starting with day 97 up to and including day 211 (Fig. 5).
For all 4 tested species, 24-h and 48-h bioassay mortality values were generally not significantly different (P > 0.05, t-test for each sample period; data not shown) throughout the study period, with few exceptions (0.05 ≥ P > 0.001) indicated in Fig. 5. However, the 24-h and 48-h mortality curves can be seen to diverge after day 6 (Cx. quinquefasciatus; Fig. 5A) and day 19 (S. calcitrans; Fig. 5C) until bioassays indicated near-zero efficacy. In M. domestica, nonsignificant separation between 24-h and 48-h mortality was apparent in all samples before efficacy dropped to zero by day 91.
Mean weekly air temperatures generally rose steadily throughout the field trials from approximately 60°F (approximately 16°C) in early February to a peak of about 82°F (approximately 28°C) in July (Fig. 4). The last posttreatment sample showing maximum mortality in Cx. quinquefasciatus, S. calcitrans, and P. papatasi, day 41, coincided with the 1st mean weekly temperature reaching approximately 70°F (approximately 21°C) that year, although RH had been high but variable (70–85%) throughout the year up to day 41. Rainfall did not exceed 1 in. (approximately 25 mm) total per week up to and including day 41.
Field collection data
Field collections of mosquitoes were synchronized with HESCO geotextile sampling periods day 19 through day 103 and for an additional collection day 113 when fabric samples were not taken. Field collections in HESCO enclosures and the 2 outside traps throughout the sampling period consisted primarily of 5 species: Aedes mitchellae (Dyar), Anopheles crucians Wiedemann, Ae. vexans (Meigen), Cx. salinarius Coquillett, and Cx. erraticus (Dyar and Knab) (data not shown). Additional species were observed in collections starting at day 41 (Coquillettidia perturbans (Walker)), day 55 (Ae. infirmatus Dyar and Knab, Cx. nigripalpus Theobald, Psorophora columbiae (Dyar and Knab), and Ae. canadensis canadensis (Theobald)), day 91 (Culiseta melanura (Coquillett), Cx. quinquefasciatus, and Ps. ciliata (Fabricius)), and day 103 (Cx. coronator Dyar and Knab). Samples of mosquito species did not exceed 5 individuals in any of the HESCO enclosures up to and including the collections from day 41, with the exception of An. crucians, with the majority of samples under 10 individuals and a maximum of 23 individuals. Although later collections of natural populations increased substantially, these early low collection numbers provided insufficient data to evaluate effects of the treated HESCO enclosures on collections of natural populations before the treatment efficacy waned in laboratory mosquito bioassays beginning with the day 41 fabric cuttings.
In this investigation we demonstrated that λ-cyhalothrin residual pesticide applied to HESCO blast protection wall geotextile in a temperate environment and subjected to natural weathering may be highly effective against mosquitoes, stable flies, and sand flies for up to 41 days posttreatment. For mosquitoes, stable flies, and sand flies up to day 41 posttreatment, 24-h and 48-h mortalities were essentially the same, suggesting a strong knockdown capability of the treatment. Bioassay data from days 19 through 91 for Cx. quinquefasciatus, days 28 through 124 for S. calcitrans, and days 103 through 124 for P. papatasi show divergence between the 24-h and 48-h mortality curves (Fig. 5). Though generally nonsignificant, these diverging patterns could suggest a reduced capability for knockdown as overall efficacy wanes. The separation of 24-h and 48-h mortality curves for M. domestica essentially throughout the experimental period, again despite being generally nonsignificant, could indicate a weak knockdown capability against this species relative to the other 3 tested species, and, coupled with efficacy below the minimum threshold, an overall relatively low capacity to control this species.
We arbitrarily defined highly efficacious at 90% and minimum efficacy at 80% mortality in bioassays. However, in an operational setting reduction of target insects even lower than 80% by exposure to residuals will positively contribute to reduction of risk of exposure to vector-borne pathogens or nuisance insects and should be considered in building an IVM program. For example, if filth-breeding flies pose a threat to personnel in the field, residual treatments of HESCO and other materials should be conducted even though treatments may never reach even 80% efficacy and only last for 1–2 wk. This is because personnel in the field should coincidentally be executing related IVM measures, such as reducing exposed food and human waste or improving physical exclusions of flies from enclosed inhabited spaces, which will work together with residual treatments to reduce the overall impact from the flies.
One interesting pattern we observed in the M. domestica bioassays was the small increase in efficacy starting at day 103 (Fig. 5). This apparent restoration of efficacy could have been due to an unintended residual effect from a pesticide spray trial that took place on day 103 at the HESCO enclosures for a separate experiment. However, this unexpected increase in efficacy was not observed in Cx. quinquefasciatus, S. calcitrans, or P. papatasi bioassays, possibly because the relative effect of the nearby pesticide trial was far below what remained of the existing residual pesticide treatment and thus did not contribute a noticeable additive effect in bioassays. Another explanation for variation in efficacy in later M. domestica samples, and, for that matter, variation in all samples represented by the SEM error bars in all 4 charts in Fig. 4 and the increase in efficacy in P. papatasi at day 246, is the common error during application of the residual with the backpack sprayer. As the operator waves the backpack sprayer nozzle back and forth across the target surface, there may be unintended slivers of untreated fabric between treated swaths. Also, even though fabric samples are cut from the same relative side of each HESCO enclosure, there are positional effects from natural differences in insolation, moisture, and other weathering that may affect the rate of decay of the residual treatment such that a later sample may actually have more bioactivity compared to earlier samples.
Field collections indicate populations were very low at least up to day 41 posttreatment, so it is very difficult to determine whether treated enclosures reduced natural populations. Apparent reductions in treated compared to untreated enclosures (data not shown) were likely artifacts of low populations; for instance, the trap in the untreated enclosure may have collected 2 mosquitoes, whereas the traps in the other 4 enclosures collected 1 each. Conversely but with a similar argument, low collection numbers could suggest no difference or an apparent increase in treated enclosures.
When mosquito populations did begin to increase by day 55 posttreatment, laboratory bioassays indicate the efficacy had sharply dropped, and little difference exists between mosquito collections in treated compared to the untreated enclosures. Initiation of the field trials in late winter was not ideal but was driven by site, resource, and personnel availability. At this site in years prior to 2012, temperatures and mosquito populations had been observed to increase rapidly in the February–March time frame (Aldridge, personal observation). However, this was not the case in 2012 and populations of the 5 prominent mosquito species observed in day 19 to day 41 collections (see Results) were low during the periods of peak efficacy of the residual treatment as measured by laboratory bioassays, preventing us from validating bioassay findings with field collections. Also, the ratio of 1 untreated to 4 treated enclosures may have been insufficient to observe differences in efficacy in the field due to possible positional effects of the 5 treated units.
We modified the 5 HESCO enclosures on day 55 by hanging an untreated piece of US military desert-pattern radar-scattering ultralightweight camouflage netting system (ULCANS; see Britch et al. 2011) across the opening to investigate whether the more fully enclosed space might either affect the CDC trap counts or better resolve the anticipated difference between the treated and untreated HESCOs by possibly encouraging mosquitoes to linger in the space and accumulate more residual pesticide. Again, however, the efficacy of the residual had waned by this time and, combined with the naturally increasing populations, it was not possible to determine the effect of adding the ULCANS netting across the enclosure openings.
Given that we used colony Cx. quinquefasciatus mosquitoes in bioassays, we were particularly interested in the effect of the residual on natural populations of this species. However, Cx. quinquefasciatus appeared in collections too late in the weathering period of the material to investigate this comparison. All but one of the species that were present up to day 41 in the field were collected at such low numbers (data not shown) that a meaningful comparison was not possible between the treated and untreated enclosures. The one exception, An. crucians, with slightly higher overall collection numbers up to day 41, showed greater numbers collected on average in the 4 treated enclosures on day 19, but substantially lower numbers on day 28, and back up to greater numbers on day 41. These observations for An. crucians suggest that either this species is not affected by this residual pesticide in the field and/or the collections are too sparse to resolve real differences across the 2 types of enclosures.
This study is the first known investigation of the efficacy and longevity of a residual pesticide applied to HESCO geotextile and provides evidence that residual treatment of this US military material may be leveraged as an IVM component to enhance the US DoD pest management system. We consider this investigation in the temperate habitat in north Florida a baseline assessment of this material with a residual pesticide that should be followed by similar studies across a variety of environments such as hot-arid (desert) and hot-humid (tropical). The HESCO system is used by the US military worldwide and studies in specific environments will provide information on expected capabilities of residuals on HESCOs against local insect threats. Future studies should also investigate effects of soil-filled HESCO MIL cells on efficacy and longevity of the residual treatment, effects of synergized IVM measures such as the presence of timed pesticide misting systems on HESCO perimeters, and effects of larger enclosure sizes such as small perimeters of, for example, 40 cells stacked 2 high. Future studies should be temporally situated in high-biting-pressure periods to investigate potential heterogeneous effects of the residual on an array of natural populations of medically important mosquito, sand fly, and filth-breeding fly species.
Similar to residual barrier applications on vegetation (Britch et al. 2009), US military ULCANS camouflage netting (Britch et al. 2011), and tents (Frances 2007), the use of residual pesticides on HESCO could be used to form protective perimeters of various sizes. Following the principles of IVM, combining barrier treatments across multiple substrates could additively increase the overall efficacy. For example, ULCANS is common in US military field settings, but tends to be used in focal spots within camps to protect personnel from being seen or to shade them—which means residual treatment of ULCANS will coincide with the presence of personnel susceptible to insect harm. On the other hand, HESCOs tend to be deployed as full perimeters around small as well as very large operating bases, as well as smaller interior perimeters around sleep, work, and food preparation/consumption areas, thereby providing substrate for a concentric series of treated perimeters that can greatly increase the presence of residual pesticide throughout the protected area.
For expert production of colony insect specimens we thank H. Brown, T. Carney, K. Kern, B. Smith, C. Swain, and J. Urban (mosquitoes), H. Furlong and R. TenBroeck (stable flies), W. Delaney, D. Johnson, C. Taylor, and R. White (house flies), and F. Golden (sand flies) at the United States Department of Agriculture–Agricultural Research Service (USDA-ARS)–Center for Medical, Agricultural, and Veterinary Entomology insectaries. This research was supported by the USDA-ARS and the US Department of Defense (DoD) Deployed War-Fighter Protection Program (DWFP). Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by USDA, DoD, the Florida Army National Guard, or the DWFP. The USDA is an equal opportunity provider and employer.