Parasitol Res (2011) 109:1279–1292 DOI 10.1007/s00436-011-2372-x
ORIGINAL PAPER
Do climatic and physical factors affect populations of the blow fly Chrysomya megacephala and house fly Musca domestica? Ratchadawan Ngoen-klan & Kittikhun Moophayak & Tunwadee Klong-klaew & Kim N. Irvine & Kabkaew L. Sukontason & Chira Prangkio & Pradya Somboon & Kom Sukontason
Received: 7 March 2011 / Accepted: 25 March 2011 / Published online: 9 April 2011 # Springer-Verlag 2011
Abstract The blow fly, Chrysomya megacephala (Fabricius), and house fly, Musca domestica L., are medically and forensically important flies. The population dynamic of these flies is essential for both control and forensical aspects. The aim of this study was to investigate the climatic and physical factors affecting the population trend of both species in Chiang Mai province, northern Thailand, using the Geographic Information System (GIS). Based on systematic random sampling, 18 study sites were selected in three districts (Mueang Chiang Mai, Mae Rim, and Hang Dong). Six land use types were involved in the study sites, i.e., disturbed mixed deciduous, mixed deciduous forest, mixed orchard, lowland village, city, and paddy field. Adult flies were sampled every 2 weeks using an in-house prototype reconstructable funnel trap. Two types of bait were used—one with fresh beef viscera for luring M. domestica and the other with 1-day tainted beef viscera for luring C. megacephala. Collections were conducted from R. Ngoen-klan : K. Moophayak : T. Klong-klaew : K. L. Sukontason : P. Somboon : K. Sukontason (*) Department of Parasitology, Faculty of Medicine, Chiang Mai University, Chiang Mai 50200, Thailand e-mail:
[email protected] K. N. Irvine Geography and Planning Department and Center for Southeast Asia Environment and Sustainable Development, Buffalo State, State University of New York, 1300 Elmwood Avenue, Buffalo, NY 14222, USA C. Prangkio Department of Geography, Faculty of Social Sciences, Chiang Mai University, Chiang Mai 50200, Thailand
May 2009 to May 2010, and analysis of climatic factors (temperature, relative humidity, and light intensity) was carried out. Correlation bivariate analysis was performed initially to determine the relationship between climatic factors and the number of flies. Consequently, an ordinary co-kriging approach, in ArcGIS 9.2, was performed to predict the spatial distribution of flies with land use and climatic factors as co-variables. A total of 63,158 flies were captured, with C. megacephala being the most common species collected (68.37%), while only 1.3% were M. domestica, thus proving that C. megacephala was the most abundant species in several land use types. A significantly higher number of females than males was found in both species. Fly populations can be collected throughout most of the year with a peak in late summer, which shows a positive relation to temperature but negative correlation with relative humidity. C. megacephala was predicted to be abundant in every land use type, from lowland to forested areas, while the density of house fly was association with altitude and land use types.
Introduction The blow fly, Chrysomya megacephala (Fabricius) (Diptera: Calliphoridae), and house fly, Musca domestica L. (Diptera: Muscidae), are medically important flies distributed worldwide, including Thailand. According to previous surveys, these were the two most abundant fly species collected, comprising ≈90% of all specimens captured in Thailand (Sucharit et al. 1976). Since their breeding grounds and adult habitat are shared with the human environment, both fly species can cause an unpleasant impact on humans. Adult flies are not only annoying to humans and animals but also
1280
mechanical carriers of several pathogens (e.g., bacteria, viruses, parasites) to humans (Greenberg 1973; Sulaiman et al. 1988; Sukontason et al. 2007b). Bacterial organisms causing diarrhea have been isolated from house flies such as Shigella dysenteriae and Escherichia coli (Butler et al. 2010), while those isolated from C. megacephala are organisms such as Aeromonas hydrophila and Pseudomonas aeruginosa (Sukontason et al. 2007b). Their larvae are myiasis-producing agent (Zumpt 1965). Conversely, C. megacephala plays an important role in forensic entomology since the larvae found in human corpses can be used in forensic investigations (Goff et al. 1988; Sukontason et al. 2007a) not only to estimate the postmortem interval of a corpse (Goff and Flynn 1991) but also to detect organophosphate poisoning in a putrefying human body (Gunatilake and Goff 1989). C. megacephala and M. domestica have complete metamorphosis in their life cycle, with four distinct stages (egg, larvae or maggot, pupa, and adult). During the larval stage, the fly passes through three different instars by molting. Development of larvae is temperature dependent (Sukontason et al. 2008). The food and feeding behavior of both species are relatively similar—adults feed on human food, garbage, excreta, and animal dung. Adult C. megacephala shows a preference for the fresh remains of corpse and/or other carrion (Bohart and Gressitt 1951). During the day, female house flies are more predominant than male in or around feeding and breeding places, where mating and resting also take place (West 1951). In a previous preliminary survey in Chiang Mai, the largest city in the north of Thailand, C. megacephala was found predominantly in March and September, while a very low number of M. domestica was reported (Lertthamnongtham et al. 2003). However, this survey was carried out in only three woodland areas in Chiang Mai province, which did not reflect the real situation of both flies in this area. Furthermore, there also has been no previous environmental factor study for fly density or population dynamic study of these two fly species in this area, which is essential for both control measures and forensic entomological application. Various environmental factors such as temperature, rainfall, relative humidity, and land use types have been considered essential since they can influence directly the population dynamic of insects (Feliciangeli 2004; Hackenberger et al. 2009; Kassem et al. 2009). Many vector-borne diseases such as dengue fever, leishmaniasis, lymphatic filariasis, malaria, and trypanosomiasis have been investigated for their distribution and relationship with environmental variables using Geographic Information System (GIS) methodology (Eisen and Lozano-Fuentes 2009; Rutto and Karuga 2009; Tonnang et al. 2010), which can be used for clarifying the complex nature of spatial relationships among many parameters (William 2004).
Parasitol Res (2011) 109:1279–1292
To understand the population dynamics of these two fly species, the systematic randomization method was used for selecting appropriate locations for 12-month fly surveys. The spatial and temporal distribution of both fly species across a variety of land use types, and the relationship of seasonal fly abundance with climatic factors (temperature, rainfall, and relative humidity) were studied using the GIS method in urban and suburban areas of Chiang Mai, Thailand.
Materials and methods Study areas This study was performed from May 2009 to May 2010 in the central zone of Chiang Mai province, northern Thailand, which has a total area of 873 km2 including three districts—Mueang Chiang Mai (MU), Mae Rim (MR), and Hang Dong (HD) (Fig. 1). MU was the urban area in this study, while MR and HD districts were suburban areas to the north and south of MU, respectively. The central zone of Chiang Mai province is mostly plain that spreads around the Ping River at an average altitude of 310 m above sea level, while in the west it is mountainous. The climate is defined as tropical savannah
Fig. 1 Map of the study area, Chiang Mai province in northern Thailand, showing three districts—Mueang Chiang Mai, Hang Dong, and Mae Rim
Parasitol Res (2011) 109:1279–1292
1281
and characterized by monsoon, which creates three distinct seasons: rainy or wet (June to October), cool (November to February), and dry (March to May). The average temperature, relative humidity, and annual rainfall in Chiang Mai province during the years 1981 to 2010 were 25.8°C, 71%, and 95 mm, respectively (data from Northern Meteorological Center). Survey sampling location design Based on a systematic random sampling method, the entire area of the three districts was stratified initially into a sampling frame of 3×3 km for MU (urban) and 5×5 km for MR and HD (suburban) using the topographical maps of Chiang Mai (MapMagic™ scale 1:150,000 on the UTM projection type, Everest Spheroid and the Indian 1975 Datum). Based on this procedure, the sampling frame (Sn) comprised all the points (x and y coordinates) located in the intended geographical region, which was divided into a set of equal plots. This was obtained as Sn ¼ ðxmin ; xmax Þðymin ; ymax Þ, where n was a fixed plot number. Consequently, the collection sites were derived using systematic sampling. This approach was achieved by sampling every kth item from the sampling frame (Sn) after the first k item had been selected at random from a random number table (Rogers and Williams 1993), with each item in the population having
Table 1 Land use classification and altitude of the three study sites in Chiang Mai based on systematic sampling
Study sites
k¼
TSn n
where TSn was the total number of sampling frames and n the number of study areas of interest. From these procedures, 18 study sites (six from each district) were selected for fly collection (Table 1). Fly collection Adult fly collection was performed every 2 weeks using an inhouse prototype reconstructable funnel trap kit invented by K. Sukontason. The trap is lightweight and portable since it was designed for quick assembly and use in remote areas. The trap consists of three components: (1) a polyvinyl chloride (PVC) frame box (30×30×50 cm), (2) a fly entrance module, and (3) a black fly net (30×30×80 cm) (Fig. 2). Two replicate sets of traps were placed initially in a single row, ≈100 m from the nearest trees in each study site. Each trap set consisted of two prototype traps (10 m apart) (Taylor and Berkebile 2006). Two types of bait were used in each trap set, one with 250 g of fresh beef viscera and the other with 250 g of 1-day tainted beef viscera, for luring M. domestica (Upakut et al. 2007) and C. megacephala (Bunchu et al. 2008), respectively.
Land use type
Altitude (meters above sea level)
x
y
495312 490465 500430 494617 491862 500893
2081814 2079323 2080649 2076438 2068567 2074324
Disturbed deciduous forest Mixed deciduous forest Mixed orchard Lowland village Mixed deciduous forest City
385 950 309 349 356 311
HD1 483280 HD2 486521 HD3 491893 HD4 489894 HD5 498572 HD6 492895 MR1 490497 MR2 482569 MR3 494184 MR4 489752 MR5 482295 MR6 494507 Median (range)
2077488 2074391 2071744 2068367 2066598 2065575 2102275 2096220 2096377 2092744 2088362 2085306
Paddy field Mixed deciduous forest Disturbed deciduous forest Mixed orchard Paddy field Mixed orchard Mixed orchard Mixed deciduous forest Mixed orchard Mixed deciduous forest Mixed orchard Lowland village
498 363 332 326 303 306 353 407 310 348 746 335 348.5 (303–950)
MU1 MU2 MU3 MU4 MU5 MU6
MU Mueang Chiang Mai, HD Hang Dong, MR Mae Rim district
UTM coordinate
an equal probability of selection. The value of k was calculated as:
1282
Parasitol Res (2011) 109:1279–1292
and Shinonaga (1977) for C. megacephala and M. domestica, respectively. The land use types The data of land use types in the central area of Chiang Mai province were received from the Geo-Informatics and Space Technology Development Agency (Public Organization), Northern Region, Thailand. Statistical analysis
Fig. 2 The prototype trap used for adult fly collection invented by Kom Sukontason. The trap consists of three components: (1) polyvinyl chloride (PVC) frame box, (2) fly entrance module, and (3) black fly net
The bait was kept in a translucent plastic container (12 cm diameter, 12 cm height) and placed underneath the fly entrance module after it had been tied to a fly net with elastic rubber bands and installed with the PVC frame box. Flies from all directions could be attracted by the odor from the bait. After landing on and/or feeding from the bait, the flies would fly up toward the light through the fly entrance module to the fly net. Fly collection was carried out for a 1-h period between 9:30 A.M. and 12:00 noon. The physical data of each study site were recorded including the co-ordinates [Garmin™ eTrex Handheld GPS (China)], temperature (°C), relative humidity (%) [Digital Hygro-Thermo Meter (DHT-1); Daeyoon Scale Industrial Co., Ltd (China)], and light intensity (lx) [LUX/FC light meter TM-204 Tenmars (Taiwan)]. Fly identification After trapping, the inlet of the fly net was sealed with elastic rubber bands to prevent the flies from escaping. Then, each fly net was labeled, detached from the PVC frame, and transported back to the laboratory within 1 h. Immediately after reaching the laboratory, all adult flies were sacrificed by placing them in a freezer set at 0°C for 2 h. All deceased flies were counted individually, sexed, and identified under a dissecting microscope using the taxonomic keys of Tumrasvin et al. (1979) and Tumrasvin
To examine the relationships between fly population and climatic data, a base-10 logarithm (log10) transformation of the raw data (n + 1) for improving the normality of distribution was applied to subsequent statistical and geospatial analyses. The influence of climatic factors on fly density was determined using bivariate correlation analysis for each of the three regions, while the data for all land uses of these three regions were combined before analysis. Oneway analysis of variance (ANOVA) and the Games–Howell post hoc test were employed to investigate the relationship between land use and mean total number of C. megacephala and M. domestica using SPSS 12.0 for Windows (α=0.05 for all tests). If the relationship was significant, a co-kriging approach (in the Geostatistical Analyst tool of ArcGIS 9.2) was utilized to predict the spatial distribution of fly population in the different study areas. Co-kriging was an extension of kriging used to combine spatial data for several co-variables to make a single map of the target variable, using information on the spatial correlation of the target variable and cross-correlations between it and other co-variables. Basically, this technique allowed samples of the co-variables to be used when predicting the target value in unsampled locations. In this study, default parameters were used on the data including temperature, relative humidity, and the total number of C. megacephala and M. domestica. Additionally, the data of land use types defined as dummy variables were also incorporated into the ordinary co-kriging analysis. The choice of the mathematical models for approximating the semivariogram/covariance function was accepted when the model resulted in the lowest root-mean-square prediction error.
Results A total of 63,158 flies were captured, of which C. megacephala was the most common (45.6%; 28,822/ 63,158). In contrast, the number of M. domestica collected was minimal at only 1.3% (821/63,158) (Table 2). A significantly higher number of both species was collected from tainted beef viscera rather than fresh viscera at 13.64
Parasitol Res (2011) 109:1279–1292 Table 2 Distribution of adult flies collected from traps at different study sites in Chiang Mai from May 2009 through May 2010
1283 Family
Species
Calliphoridae
Number of flies (% of total)
Chrysomya megacephala Other Calliphoridae
Muscidae
Musca domestica Other Muscidae
Sarcophagidae Others Total
and 6.40 times for C. megacephala and M. domestica, respectively (Mann–Whitney U test; P=0.000) (Table 3). Seasonal analysis showed that high numbers of C. megacephala and M. domestica were collected in the dry season (March to May), with the peak being May (Fig. 3). After the dry season, fly populations decreased rapidly, beginning from the start of the rainy season (June to July) and retaining low numbers in the of rainy and cool season. The lowest numbers of C. megacephala were found in November, with no M. domestica collected from any sites in January. After the cool season, the number of both fly species collected increased; however, the population of C. megacephala fluctuated with a tendency to increase, while the increase in M. domestica was relatively fast (Fig. 3). Figure 4 represents the number of females per male of C. megacephala and M. domestica obtained from the total number of specimens collected from 18 sites, showing significantly higher numbers of females than males (Mann–
Table 3 Number of C. megacephala and M. domestica using different bait in a fly trap during the fly survey Bait
C. megacephala total [median (range)]
M. domestica total [median (range)]
Fresh beef viscera One-day tainted beef viscera Mann–Whitney U test
1,969 [0.0 (0–198)] a
111 [0.0 (0–22)] a
26,853 [16.0 (0–831)] b
710 [0.0 (0–58)] b
P=0.000
P=0.000
Different letter (a, b) shown as a significant difference within groups when P 40
12 75 136 60 15 4
17.5 37.0 51.0 52.0 54.0 44.5
−0.294 0.092 0.268 0.126 0.058 0.838 0.215 0.442 −0.089 0.752 −0.302 0.274 −0.069 0.699 −0.154 0.384
most suitable bait to use for collecting blow fly and house fly in the field. Higher numbers of females than males were found in both C. megacephala and M. domestica. For C. megacephala, the year-round occurrence without any peak of density corresponded with other reports. Lertthamnongtham et al. (2003) proposed that predominate females were at 1.5:1 to 30:1 female/male ratios. Likewise, more female M. domestica were collected than males during the dry season and early rainy season (March to May), which corresponded with the study of Bong and Zairi (2009) in Malaysia. In their 1-year survey, the peak density of house fly was found during March to July and low density during August to February. In addition, Avancini and Silveira (2000) revealed that the female to male ratio for M. domestica showed a strong prevalence of females at ratios from 1.5:1 (in the morning) to 26.8:1 (at noon). More females than males being collected from the field survey suggested that the potential for finding protein was from natural sources used for oviposition, food source, and/or breeding places (Spradbery 1979). In this study, C. megacephala revealed a relatively high dispersal potential, with most collections made at temperatures ranging from 15°C to 40°C, thus maintaining longterm population stability throughout the entire sampling period, with a high peak in the late summer, and low peak in the wet and cool dry weather. Sukontason et al. (2008) revealed that the developmental rate of C. megacephala larvae rapidly appeared in April at an average temperature of 31.4°C, but more slowly in the rainy season and winter. Reigada and Godoy (2005) suggested that adults of this species cannot overwinter because low temperatures restrict their growth. As for M. domestica, the number of flies captured in winter declined to nearly zero by the end of October to January in some study areas, especially in HD. No house fly was collected when the temperature dropped
−0.053 0.600 0.061 0.546
−0.157 0.117 0.041 0.686
Data are presented as median number (range) of the total fly samples according to the temperature ranges; different letter (a, b, c) shown as a significant difference between groups when P 45°C and
