Portal de Periódicos da CAPES

A comparison of visual and flagging methods for estimating adult Ixodes pacificus (Acari: Ixodidae) tick abundance

2010; Wiley; Volume: 35; Issue: 2 Linguagem: Inglês

10.1111/j.1948-7134.2010.00104.x

1948-7134

Martin B. Castro, J R Clover,

Mosquito-borne diseases and control

Determining the abundance of adult host-seeking ticks is a necessary component of assessing the risk of exposure to tick bites and, in turn, risk of tick-borne disease transmission. Determining abundance is also essential in the appraisal of the efficacy of any control methods applied. Host-seeking adult ticks are commonly collected from vegetation with the use of flags or drags. To ensure a reliable estimate, multiple passes are often made over the same location at different times, with ticks attached to the flags being enumerated after each increment of a fixed distance or time (Lane et al.1985, Falco and Fish 1989, Ginsberg and Ewing 1989, Lane 1990, Lane and Stubbs 1990, Maupin et al. 1991, Goddard 1992, Schmidtmann et al. 1994, Kollars et al. 1999, Li et al. 2000, Schulze et al. 2001). Other adult tick collecting techniques have been studied or employed, such as removal from hosts, walking surveys, inspection of leaf litter material and trapping including CO2 and pitfall types (Piesman and Spielman 1979, Ginsberg and Ewing 1989, Solberg et al. 1992, Lane 1996, Mejlon and Jaenson 1997, Schulze et al. 1997). For some study hypotheses it may be desirable to not remove the questing ticks from a given location yet still perform an accurate assessment of the tick population present. Kramer et al. (1993) found that biweekly removal sampling by flagging would have resulted in substantially underestimated abundance of questing Ixodes pacificus as well as underestimating the duration of adult seasonal activity. Li and Dunley 1998 found that tick density could be significantly underestimated under certain conditions due to drop-off of ticks with increasing subtransect lengths or flagging over rough vegetation. Schulze and Jordan (2001) found adults of Ixodes scapularis and Amblyomma americanum remained attached to drags three to four times greater in distance in sparse vegetation compared to dense vegetation. For our organization, traditional flagging has and will continue to be the primary technique for tick surveillance. In anticipation of conducting biweekly long-term surveillance of host-seeking adult ticks at permanent sites, a visual examination of vegetation has been proposed as an alternative to our traditional flagging. Our intent was to utilize a method that was less likely to disturb, displace, or damage ticks. We investigated the use of a visual surveillance method for counting adult ticks on vegetation within 1 m of a trail edge. The null hypothesis investigated in this study was that there would be no significant difference in tick abundance estimates calculated from flagging methods compared to visual methods. Tick surveys were conducted at two state parks in northern California over approximately eight weeks from January 22 to March 16, 1992. Annadel State Park is located in Sonoma County. This site (38° 27' 11.27" N × 122° 38' 51.51" W) is at 140 m elevation on a north-facing slope (Figure 1). Olompali State Park is located in Marin County. This site (38° 9' 06.76" N × 122° 34' 24.76" W) is at 58 m in elevation on a southeast-facing slope (Figure 2). The study sites are approximately 35 km apart. Both study sites are in coast live oak woodland habitat. The canopy cover is dominated by coast live oak (Quercus agrifolia), California laural (Umbellularia californica), and Pacific madrone (Arbutus menziesii). The understory is primarily introduced annual grasses of wild oats (Avena fatua), barley (Hordeum ssp.), rye grass (Elymus ssp.), and brome (Bromus ssp.) with occasional patches of poison oak (Toxicodendron diversiloba). A portion of a sampling site at Annadel State Park, Sonoma County. A portion of a sampling site at Olompali State Park, Marin County. At each site, two 45-m transects were established and marked at 3–m intervals. These transects were along the uphill edge of existing trails. On each survey date each transect was first visually inspected. A visual inspection of each 3–m section entailed scanning the vegetation along the transect length and up to one meter in from the trail edge. Observations focused on the tips of the vegetation. Subsequently the investigators traded transects and surveyed for ticks using a 1–m square flannel flag. Flagging was performed by sweeping the flag back and forth horizontally over the vegetation. The numbers of each sex of adult ticks were recorded for each 3–m section. After identification any flagged ticks were released back to the same 3–m section from which they were collected. Although observations focused on the tips of vegetation, grass stems were also scanned. Individual tick measurements were not taken at this time but very few ticks were not near the tips of the vegetation. This is in agreement with other northern California study sites where greater than 97% of ticks were noted within 7.5 cm of the tips of vegetation (M.B.C., unpublished data. n>500). Simultaneous visual and flagging surveys eliminated bias for temperature, humidity, wind, and time of day between sampling methods. Biases attributed to investigator sampling techniques were avoided by using the same personnel for flagging and visual surveillance of the same transect each sampling period. Host-seeking immatures were not intended to be part of this study and none were recorded visually or by flagging. Adult ticks sequestered in the soil and leaf litter area were also not taken into consideration as these ticks would not be enumerated by either of our study methods. The numbers of ticks observed by visual inspections were compared to those collected by flagging by paired t-test. Values of p<0.05 were considered statistically significant. Mean (M) and standard deviation (SD) were calculated for both the flagging and visual methods. The respective variabilities of each technique were evaluated by comparing the coefficients of variation (C.V.).A total of 16 surveys was conducted at the two study sites over eight weeks. All ticks observed were adult Ixodes pacificus Cooley and Kohls. Estimates of ticks within a 45 m2 area by the visual method (M=38.5, SD=26.44) were significantly greater than the number of ticks captured by flagging (M=29.56, SD= 20.29; p=0.0026) (Table 1). The flagging and visual methods were also evaluated by t-test to ascertain if collection differences with respect to tick sexual status occurred. There was no difference between methods with regard to detection of males (p=0.3948) (Table 2), but more females were visually counted than were flagged (p= 0.0036) (Table 3). It is unclear why this occurred. As noted earlier, very few ticks were not at the tips of vegetation when our survey was performed and those few ticks would not account for this difference. It may be that females were lost as part of the flagging process. Further study is warranted, such as performing an additional visual check subsequent to the flagging survey to see if ticks remain on the vegetation. There was no difference in the variability of tick number estimates between the visual (C.V.=68.68) and flagging (C.V.=68.64) methods. Numerous factors are recognized to complicate and impede mechanical collection of ticks with a flag. If these factors are themselves subject to change (e.g., wind), estimates based on flagging may be unstable upon repeated sampling. In contrast, visually counting ticks on vegetation is less susceptible to these changes in conditions and estimates based on the visual method are more invariant. However, in this study, there was no apparent difference in variability between flagging and the visual method, suggesting that these compromising factors have equivalent influence on both methods. The time period to flag or visually check each transect was recorded. The total time for visual inspections for the 16 surveys was 496 min (M=31) while total flagging time was 317 min (M=19.81). The paired times for each observation/flagging event were also evaluated with a paired t-test (t=4.5290, two-tail p=0.0004). This test revealed that the time to visually inspect for the presence of ticks took significantly longer than to flag for their presence (Table 4). Despite the fact that the visual method took longer to execute than the flagging method and also identified more females than the flagging method, we found that surveying for adult ticks visually provided a consistent alternative method to estimate relative tick numbers compared to our current standard flagging techniques and should be considered when designing long-term surveillance studies to assess adult host-seeking tick populations at permanent sites. This study was performed before the typical emergence period of adult Dermacentor occidentalis and Dermacentor variabilis for this region, but we feel this visual surveillance method would be applicable to these species as well. We thank personnel from Annadel and Olompoli State Parks for permission to utilize park property for this work. We are grateful to Curtis Fritz for constructive review of the manuscript.