METHODS

Three 10-acre test sites each with a 4.5-acre center evaluation plot were selected.  In Site A (IPM), treatment consisted of the release of two biological control agents, the decapitating phorid fly, Pseudacteon tricuspis (Fig. 1), and a pathogen of fire ants, Thelohania solenopsae (Fig. 2) plus the application of granular fipronil (0.1% at 12.5 lbs/acre) applied once during the study to the center 4.5-acre plot to lower the initial fire ant population.  The working hypothesis was that self-sustaining biological control agents would delay reinfestation by the fire ants into treated areas giving longer control, thus, reducing the need for chemical applications.

In addition, the reinfestation delay would allow the native ant populations to increase as the fire ant populations remained low.  Site B was similar to Site A except that only the granular fipronil treatment as above was applied.  Site C was an untreated control.  Evaluations were conducted 2X per year on 1/8-acre subplots inside and outside each 4.5-acre plots and all active mounds within each sampling area were assessed using the USDA population index rating system.  Active mounds were mapped using GPS to document fire ant populations, reinfestation rates, and spread of the biocontrol agents.  In addition, 8 pitfall traps per 1/8-acre circle were used to assess the native ant populations.

RESULTS AND DISCUSSION

Three years after initial treatment, the results in Figure 3 indicate that: a) in Site C (untreated control), fire ant populations increased by 7%, however, during the study, increases as high as 83% occurred; b) in Site B (chemical only treatment), the fire ant population decreased by 80%, and 72 weeks after the initial treatment, the fire ant control dropped below a preestablished 90% threshold.  Under normal circumstances, at this point, retreatment would have been necessary; c) in Site A (IPM), there was 95% reduction in fire ant colonies in the final evaluation.  In several evaluations during the study, the reduction was 97% or greater.  These results indicate that the IPM approach using a single chemical treatment in conjunction with the two biological control agents controlled fire ants for a longer period without the need for retreatments.

In addition, in the IPM treatment area (Site A), the native ant populations increased while the fire ant populations continued to decrease (Figure 4).  The native ant populations remained low in all other sites and did not replace fire ant populations.

Figure 5 shows the mean T. solenopsae infection rates for plots in the different areas.  A high infection rate occurred in the IPM untreated area of Site A, while the infection rate in the IPM fipronil-treated area was much lower because there were very few colonies remaining.   T. solenopsae was also detected in Site B (fipronil-treated area) thus the pathogen spread to this site even though it was never applied here.

Figure 6 shows the actual spread in the IPM area (Site A) of the pathogen, T. solenopsae, in fire ant colonies (yellow dots).  At this site, 138 of the 175 (79%) colonies were infected in September 2002, as opposed to only 30% observed in May 2002.  It also demonstrates detection of T. solenopsae in the chemical treatment site, with 35 of the 160 (22%) colonies infected in September 2002 compared to 4 of the 66 (6%) colonies in May 2002.  The pathogen may have been accidentally transferred to this site during our surveys, but the fact that this organism is so easily spread is encouraging.  The decapitating flies also spread and were found attacking fire ant workers within an area up to three miles around the initial release site.  In conclusion, these results are a good indication that it is possible to use an IPM approach for the control of fire ants when the right tools and conditions are available.