THINKING "OUTSIDE" THE BOX

Will exterior-only termite treatments work? A University of Kentucky study suggests it’s a possibility.

No topic is generating more interest than the non-repellent liquid termiticides. Premise® (imidacloprid), Termidor® (fipronil), and Phantom® (chlorfenapyr, which is not yet commercially available) are causing many termite control professionals — including some bait users — to rethink their control options. That’s pretty amazing, considering all the attention paid to baits in recent years. Companies that had all but abandoned their termite rigs are moving a greater percentage of their service back to liquids. The incentive for change is attributable, in part, to the bold marketing programs of manufacturers, promising compensation if companies do not achieve at least five years of effectiveness with their products.

Unfortunately, these newer, non-repellent materials still must be applied in the way termiticides have for decades. Consequently, customers’ floors and walls often have to be drilled, furnishings and stored items moved, a treating hose must be dragged through the house and large amounts of pesticide are applied indoors. In other words, all of the non-intrusive, consumer-friendly advantages of baits are negated. In a recent householder survey, 93 percent of respondents expressed concern about the application of termite control chemicals inside their home — 78 percent indicated they were very concerned (see "Termites & Public Attitudes," http://www.pctonline.com/articles/article.asp?MagID=1&ID=975&IssueID=122). Regardless of how effective the non-repellent liquid termiticides may be, many customers still will not want them applied in their homes.

But what if structural protection can be afforded by applying these products primarily around the exterior? Anecdotal reports from the field suggest this possibility, and we saw similar indications following an exterior treatment of a severely infested tobacco barn last summer (see "The New Termite Killers," http://www.pctonline.com/articles/article.asp?MagID=1&ID=1220&IssueID=126). Here we report on several additional exterior-only treatments performed on both barns and houses. We think you’ll find the results intriguing, but be forewarned — there are quite a few details. Considering the potential implications, we thought they were important.

THE STUDY. Our study was designed to answer two related questions: First, whether subterranean termites in buildings could be eliminated by exterior-only, liquid termiticide applications; and second, could such treatments reduce or eliminate foraging activity around structures. Evaluation sites included 10 structures in central Kentucky, seven tobacco barns and three houses. All were heavily infested with the Eastern subterranean termite, Reticulitermes flavipes. Tobacco barns are especially useful for this type of work, because (1) there are no occupants concerned with whether the experimental treatment is going to be effective; (2) the barns are often heavily infested, both inside and out, and (3) they have bare walls and dirt floors, aiding observation of termites and treatment effects. All of our study sites, including the three houses, had multiple areas of activity indicated by swarms, active mud tubes, infested structural wood, and termites feeding in ground debris.

MONITORING METHODS. To be successful, exterior treatments should ideally eliminate termites in areas away from outside-treated zones (e.g., termites living in interior walls, or entering through untreated expansion joints, voids, plumbing penetrations, etc). To help observe such effects, long-term monitoring stations were established which served as untreated ‘windows’ into the colony’s network of activities. The approach used was similar to that for assessing the effects of baits on termite populations (Grace et al. 1996, Su and Scheffrahn 1996).

Monitoring stations were installed on both the interior and exterior of structures. Below-ground stations were established in areas where termites were feeding in cellulose debris, or on wooden survey stakes. They were installed by digging a six-inch hole and inserting a plastic bucket (6 inches in diameter by 6 inches high) with the bottom cut out. A pre-weighed monitoring block was placed at the bottom, consisting of several wood pieces bound together and separated by toothpick-diameter dowel rods. If the ground was too hard to excavate, monitoring blocks were set directly over active termite foraging tunnels emerging from the soil and covered with an inverted, plastic bucket. In some structures, monitoring stations were also established above ground on walls, flooring, etc. Small plastic boxes were secured over active mud tubes and provisioned with wood or corrugated cardboard as a food source.

We also used an acoustic emission device to monitor activity in infested wood members, before and after treatment. This device has been shown to accurately detect the sounds made by termite mandibles tearing wood fibers (see "Active or Inactive?" http://www.pctonline.com/articles/Article.asp?MagID=1&ID=1362&IssueID=138). The unit records termite feeding activity within the wood as counts or clicks per minute. In general, the greater the number of counts recorded per minute, the more intense the activity. Active mud tubes and ground debris were also inspected, periodically, for presence or absence of termites.

The number of monitored, active locations per structure (six or more, excluding infested debris and tunnels) was substantial for these types of evaluations. Our goal was to obtain the clearest possible picture of how termites, feeding in discreet areas, were affected by exterior treatments. Distances between monitoring stations and the exterior-treated zone ranged from two feet — to more than 30 feet — encompassing the typical feeding locations of termites in most structures. Foraging territories of individual colonies were delineated by introducing dye-impregnated paper into one or more active monitoring stations, and noting in which traps the pigmented termites reappeared (Atkinson 2000).

Structures were inspected monthly, and the number of monitoring stations with termites was recorded. Wood consumption was measured by removing the monitoring blocks and washing, drying and weighing them, relative to their initial weights. Considerable attention was paid to minimizing termite disturbance, and maintaining a favorable environmental condition within monitoring stations. Monitoring blocks and locations around stations were periodically moistened with water, particularly during drier months. Prior to treatment, termites seldom abandoned stations once they had initiated feeding.

TREATMENT METHOD. Ten structures were treated; six with Termidor, four with Premise. Treatments were applied by local pest control companies working under our direct supervision. Only the outside perimeter of each structure was treated. A narrow (4-inch by 4-inch) trench was dug around the exterior foundation, into which 0.06 percent Termidor 80 WG or 0.05 percent Premise 75 was applied at 4 gallons per 10 linear feet per foot of depth. Where distance to the footing was more than 12 inches, the dilution was rodded into the base of the trench to a maximum depth of 2 to 2.5 feet. If an active, below-ground monitoring station was in the path of treatment, a two-foot buffer of untreated soil was maintained on either side of the station. (Any product will affect termites if you spray them directly.)

RESULTS. Here’s what happened, structure by structure; first with Termidor, followed by Premise. Please keep in mind that these studies are ongoing.

TERMIDOR: Barn #1. This tobacco barn in Paris, Ky., was originally treated on the exterior with Termidor in July 1999 (see "The New Termite Killers," http://www.pctonline.com/articles/article.asp?MagID=1&ID=1220&IssueID=126). Activity inside the barn was intense, with large numbers of termites feeding in eight independent monitoring stations, and in numerous pieces of wood debris on the dirt floor. Within 3 to 4 months of application, none of the monitoring sites were still active, and tremendous numbers of dead and dying termites were found inside several stations. In subsequent inspections, no termites were found in any of the monitors or elsewhere throughout the barn during all of 2000, leading us to conclude that the infestation has been eliminated.

Fig. 1. Diagram of barn #2 treated with Termidor on the exterior. Themonitors inside the dotted line were all inhabited by termites of the same colony.

Barn 2. This tobacco barn, located in Nicholasville, Ky., was as severely infested as the previous one. It had been treated unsuccessfully on more than one occasion by a local pest control company. The perimeter foundation had a concrete block stem wall and the wood siding often extended below grade. Perimeter dimensions were 270 linear feet. Six below-ground monitoring stations were established where termites were discovered in wood debris. Most were located 2 to 4 feet from the exterior treated trench, but one was more than 12 feet away, toward the center of the barn. Two above-ground monitoring stations were also established over active mud tubes on interior walls. A diagram of the barn and location of stations is shown in Figure 2. Marked (dyed) termites originating in a station on the west end of the barn, were subsequently found in all other monitoring stations, indicating a single interconnected foraging population or colony.

Fig. 2. Decline in termite activity in barn #2, following June28 Termidor application. Blue lines represent number of stations with termites; red line shows grams of wood consumed per station per week.

One hundred gallons of Termidor were applied to the exterior trench on June 28, 2000. Six days after treatment, active mud tubes on the interior surface of the northeast wall — including some that were 3 to 4 inches wide — became inactive. Less than three months after treatment, no live termites were found in any of the monitoring stations, and consumption of the wood monitoring blocks declined to zero (Figure 2). As with Barn #1, dead termites were found in both below and above-ground stations. One station, attached to the inside finished wall of a tobacco stripping room, also had large numbers of phorid flies (adults and pupae), presumably scavenging on dead termites, inside. We also found phorid flies associated with dead termites in Barn #1. Interestingly, one dead termite found in an above-ground station mounted on the northeast wall of the barn was dyed red — indicating it had also fed at the opposite end of the barn where the dyed paper was introduced.

No further signs of termite activity were found in any of the monitoring stations, or elsewhere throughout the barn, during September through November. (For reference, termites in Kentucky are normally quite active during this period.) Considering their prevalence prior to treatment, we believe the infestation has been eliminated.

Fig. 3. Diagram of house #2 treated with Termidor on the exterior. Note location of monitoring stations and acoustic emission detector (AED) readings.

Barns 3 and 4. These two barns in Nicholasville and Wilmore, Ky., also had multiple areas of activity. Both barns were rectangular with perimeter dimensions of 230 and 280 linear feet, respectively. The Nicholasville site had seven active, below-ground monitoring stations located on the exterior, while the Wilmire barn had six (3 exterior, 3 interior).

The Nicholasville barn was trenched and treated with 86 gallons of Termidor on July 26, 2000. Five infested exterior, below-ground monitoring stations located 2 to 4 feet from the treated trench, were free of termites within two months. Two other exterior stations, located 16 and 23 feet away from the treated trench near an infested tree stump, remained very active through the end of November (four months post-treatment). Although these monitoring locations were a considerable distance from the barn, the termites were of the same foraging population (colony) as those within stations closer to the structure. It will be interesting to see whether the remote, but interconnected, locations remain active this spring. In the meantime, termites in proximity to the structure appear to have been eliminated.

Results were similar at the Wilmore site. The barn was treated with 89 gallons of Termidor on September 7, 2000. Five infested underground stations (3 interior, 2 exterior), located 2 to 4 feet from the treated exterior trench, were free of termites the following month. Very little wood was consumed in these stations after treatment. A sixth monitoring station, located 32 feet from the barn by an infested tree, remained very active through the mid-November inspection, and will continue to be monitored when the weather warms this spring. Since no dyed individuals were recovered there, it may be part of a separate colony.

House #1. This small, one-story house in Lawrenceburg, Ky., was interesting for several reasons. The owner first discovered live termites under a wooden clothes chest, eating through the hardwood floor. The floor had moisture meter readings in excess of 20 percent, mainly from poor ventilation and only inches of soil clearance in the crawl space. The crawl space also had a severe infestation of brown recluse spiders — contributing to the difficulty (and risk) of treating the house conventionally. Eight exterior, below-ground monitoring stations were established; four within 2 to 4 feet of the foundation, and four others located around an abandoned well, 12 to 18 feet from the house. A heavily-infested wood pile, on top of the concrete pad covering the well, served as yet another (qualitative) monitor of termite activity. Perimeter dimensions of the house were 136 linear feet.

The house was trenched and rodded around the exterior with 80 gallons of Termidor on August 25, 2000. Three heavily infested monitoring stations adjacent to the foundation were inactive the following month, and only a few termites (mainly soldiers) remained in a fourth station until it, too, became inactive in November. Very little wood was consumed in these stations after treatment. Termite activity remained very high, however, in all four monitoring stations around the well and in the woodpile, through the end of November. Recovery of dyed termites in these areas, 12 to18 feet from the foundation, indicated they were of the same colony as those eliminated earlier near the structure. It will be interesting to see whether the connected feeding locations around the well remain active, or eventually disappear.

House 2. It is hard to quantify treatment effects inside a home. Finished walls and other obstructions permit few opportunities to monitor termite activity before and after treatment. Furthermore, few homeowners are inclined to let you "study" their termites for extended periods before attempting to kill them. House #2 afforded such an opportunity. The two-story brick home in Lexington was bought at a foreclosure sale, and the new owners anticipated a few months of remodeling before moving in.

The house was only 13 years old, but had a severe termite problem. A treatment had been performed in the past but no records were available. The entire front wall to the left of the porch had to be replaced, as did the threshold, door casing, and some ceiling plates along an adjoining wall. Termites were found in three locations inside the unfinished basement, feeding in cardboard boxes against the poured foundation wall. The boxes were removed, and long-term monitoring stations were established on top of the basement floor; a fourth station was established over an active mud tube inside the entrance to the garage.

Acoustic emission detector readings taken inside the basement also revealed high levels of termite activity along a 17-foot length of sill plate, behind the front porch. Fourteen active spots were monitored along the plate, between successive pairs of joists. A diagram of the house and monitored locations is shown in Figure 3.

The house was trenched and rodded with Termidor on October 13, 2000. Perimeter dimensions were roughly 200 linear feet, and 158 gallons were applied only around the exterior. No treatment was done inside the basement, garage or crawl space, and the only area of drilling was where the front porch slab abutted the foundation. On the north, south and east sides of the house, the footing was several feet below grade; rodding depth along the foundation, in these areas, was limited to about 2 to 2.5 feet.

Fig. 4. AED readings taken inside the basement of house #2, treated with Termidor on Oct. 13. Counts are average readings from 14 active spots along a section of sill plate.

Eleven days after treatment, emission detector readings inside the basement along the sill plate plummeted almost to zero. No termites were detected during two subsequent monthly readings (Figure 4). One month after treatment, dead soldier termites and no live workers were found in the three monitoring stations, located in the garage and along the bottom of the east wall in the basement below. At two months post-treatment, large numbers of dead and dying termites (soldiers and workers) were also found in the remaining basement station, opposite the wood deck. These results are remarkable, considering that neither the basement, crawl space or garage interiors were treated.

PREMISE: Barn #1. This infested tobacco barn in Nicholasville, Ky., afforded many opportunities to monitor exterior treatment effects of Premise. The barn had a concrete block stem wall and typical wood siding. Ten below-ground monitoring stations were established in areas where termites were feeding in wood debris. Five were located inside the barn, 2 to 5 feet from the exterior treated trench; five others were located outside, between a line of wooden support poles supporting an attached breezeway. Active mud tubes were located throughout the barn, including one, 3-inches wide, on an interior door casing to which another monitoring station was affixed.

Sixty-five gallons of Premise were applied to the outside trench (168 linear feet) on June 30, 2000. The side of the barn with the attached breezeway could not be fully treated because there was heavy equipment stored against the wall. However, the wooden support poles that were within 2 to 4 feet of the outside monitors were trenched and rodded. Five consecutive months of post-treatment monitoring showed no appreciable decline in termite activity in any of the monitoring stations or within mud tubes, a number of which were within 2 to 3 feet of the treated trench. Termites also could be found in ground debris.

Barns 2 and 3. Both of these barns, located in Paris, Ky., and Woodford Co., Ky., had plenty of termites and apparently single colonies. The Paris barn had seven active below-ground monitoring stations on the interior, while the Woodford Co. site had six; 4 interior, 2 exterior. All stations were within 2 to 7 feet of the outside trench. Neither barn had a concrete stem wall, but the wood siding contacted the ground.

The respective barns were trenched and treated around the exterior with 80 and 105 gallons of Premise on July 25 and August 31, 2000. Although some stations eventually became inactive, the effects were not as rapid or pronounced as with Termidor. Three months after treatment, 3 of the original 7 monitoring stations in the Paris barn no longer had termites. Two months after treatment at the Woodford Co. site, 3 of 6 stations became inactive. While only two stations in each barn had termites in mid to late-November, colder temperatures may have caused abandonment of some stations. Because of their transient nature, wood consumption tends to be a more accurate measure of ongoing termite activity. Feeding rates in both barns, during October and November, did not decline appreciably from pre-treatment levels. Monitoring will resume when the weather warms this spring.

House #1. This house, located in Mt. Sterling, Ky., had a severe termite problem with swarms the past several years. There had been prior treatments but no records were available. Damage was extensive in the basement and in three rooms on the main floor. Six active, below-ground monitoring stations were present on the front and left sides of the house. Five of the outdoor stations were within 2 feet of the exterior trench, and one was 9 feet away, off the front corner of the porch. Termites were also monitored inside the basement, along a 15-foot span of sill assembly, adjacent to where the active monitors were located outdoors. Acoustic emission detector readings were taken monthly at several spots along the infested wood members.

The house was treated on the exterior with 125 gallons of Premise on August 24, 2000.

Perimeter dimensions were 150 linear feet. Although in some areas the footing was several feet below grade, rodding depth was limited to about 2 feet. Attached slabs, including the front porch, were drilled conventionally. Despite most monitors being relatively close to the treated zone, all but one remained active through the end of November. Wood consumption was comparable to pretreatment levels. Acoustic emission counts in the basement remained high in September and October, but were lower in late-November. Monitoring will resume this spring.

STUDY IMPLICATIONS. Liquid termiticides have been applied in much the same way for 50 years. In a painstaking effort to treat every conceivable termite entry point, we drill floors, pull back carpet, cut tile, and pump hundreds of gallons of pesticide under places where people work and live. With at least one product we tested, much of this effort may be unnecessary.

Consider the six structures treated with Termidor: all 34 infested monitoring stations within 2 to10 feet of the exterior treated trench were inactive after one to three months. Cessation of termite activity was supported by "flat-lined" acoustic emission readings, absence of termites in previously active mud tubes and debris, and termite "body counts." These results are remarkable, considering the high pre-treatment infestation levels, and that more than half of the active monitors were on the interior of structures.

Imagine that the active monitoring stations were clusters of termites, feeding in and around a home. It seems they would have been eliminated with an exterior treatment of Termidor.

Now consider where most structural infestations originate: although termites occasionally appear in the centers of buildings, most infestations are associated with perimeter walls and adjoining areas. This has much to do with location of termite entry points, but probably also has a lot to do with moisture — arguably the most important factor in where termites appear around buildings. Inside our barns, we seldom find termites in the center, where the soil is drier. They are much more abundant toward the edges, where the ground remains moist. The same should hold for most houses, where the ground around the perimeter is moistened by rainfall and irrigation — not to mention the added attraction of mulch beds, plant roots, etc.

It is likely that many of today’s construction challenges (e.g., sub-slab heat ducts, foam insulation/stucco below grade) will be less of a problem with Termidor and, perhaps, other liquid non-repellents, as we’ve seen with some baits. (If termites are dying in monitoring stations, two or more feet from the exterior-treated zone, why should they survive, inches away, behind foam board?). If exterior liquid treatments eventually are adopted, it still would be prudent to ‘spot- treat’ infested or high-risk interior areas, as is commonly done with exterior bait applications. However, the 50-year habit of drilling every conceivable termite entry point "by the tape measure"— with the associated labor, liability, intrusiveness, and cost to the consumer — seems unwarranted.

With some products, deep-rodding of foundations and drilling of basement floors (other than perhaps walk-outs), may also prove unnecessary. Most termite foraging occurs in the upper few feet of soil, since this is where most of the oxygen and food resources, such as plant roots, are located. Any termites tunneling deeply around foundation walls should eventually encounter non- repellent termiticide residues nearer the surface. At a time when the public, the EPA, and virtually everyone else is demanding less pesticide, perhaps it’s time to rethink the four-foot minimum depth requirement when rodding deep foundations.

The limited effect on termites away from Premise-treated zones was somewhat unexpected, considering how well the product has performed commercially. Our findings are consistent, however, with results at USDA Forest Service research sites, where termite activity is subsiding in untreated control plots in the vicinity of Termidor plots, but not near plots treated with Premise (Kard 2001). Other researchers have reported population decreases following application of Premise (Oi, 2000).

The fact that we could not consistently eliminate termites in areas away from exterior-treated zones does not mean Premise is ineffective. On the contrary, long-time Premise users frequently report single-digit retreatment rates, which have been documented in surveys by us and others (see "The New Termite Killers," http://www.pctonline.com/articles/article.asp?MagID=1&ID=1220&IssueID=126), resulting in a high satisfaction rating for the product. While it did appear, in our study, that Termidor was more effective in killing termites away from treated areas, it is possible that similar effects will be seen with Premise when monitoring resumes this spring.

At the beginning of this article, we mentioned that many termiticide manufacturers are now backing their products with five-year performance commitments, provided they are applied according to label directions. It must be stressed, however, that neither Premise nor Termidor are currently labeled for exterior-only application, and the legality of such treatments will vary state-by-state. Companies should consult the manufacturers and their state regulatory agencies for official opinions on the experimental treatments discussed in this article.

References
Atkinson, T.H. Use of dyed matrix in bait stations for determining foraging of subterranean termites (Isoptera: Rhinotermitidae: Reticulitermes spp. and Termitidae: Amitermes wheeleri). Sociobiology 36: 149-167.

Grace, J.K., C.H.M. Tome, T.G. Shelton, and R.J. Oshiro. 1996. Baiting studies and considerations with Coptotermes formosanus (Isoptera: Rhinotermitidae) in Hawaii. Sociobiology 28: 511-520.

Kard, B. 2001. Gulfport studies stay the course. Pest Control 69(1): 30-33, 73.

Oi, F. 2000. Impact of Premise 75 on field populations of Reticulitermes spp. 2000 Joint Annual Meeting of Entomological Societies of America & Canada. Montreal, CAN.

Su, N.-Y., and R.H. Scheffrahn. 1996. A review of the evaluation criteria for bait-toxicant efficacy against field colonies of subterranean termites (Isoptera). Sociobiology 28: 521-530.