Accepted Manuscript Family composition and temperature in fly assemblages: Community temperature index using family temperature index Tae-Sung Kwon PII: S2287-884X(17)30078-X DOI: 10.1016/j.japb.2017.07.003 Reference: JAPB 242 To appear in: Journal of Asia-Pacific Biodiversity Received Date: 13 June 2017 Revised Date: 30 June 2017 Accepted Date: 6 July 2017 Please cite this article as: Kwon T-S, Family composition and temperature in fly assemblages: Community temperature index using family temperature index, Journal of Asia-Pacific Biodiversity (2017), doi: 10.1016/j.japb.2017.07.003. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT 1 Title: Family composition and temperature in fly assemblages: Community temperature index 2 using family temperature index 3 4 Short running title: Family temperature index of flies 5 6 Paper Type: Research Paper 7 8 Address: Forest Insect Pests and Diseases Division, National Institute of Forest Science, 57 9 Hoegi-ro, Dongdaemun-gu, Seoul 02455, Republic of Korea 10 11 Author: Tae-Sung Kwon 12 Tel: 82-2-961-2655 13 Fax: 82-2-961-2679 14 E-mail: [email protected] 15 16 17 18 19 20 MANUSCRIPT 21 22 23 24 25 26 27 28 29 30 31 32 ACCEPTED 33 34 35 36 37 38 ACCEPTED MANUSCRIPT 39 Abstract 40 This study aimed to examine relationship between temperature and family composition of 41 fly assemblages. Flies were surveyed at six locations in the Baekdudaegan mountain range 42 (South Korea) of which mean annual temperature (MAT) ranged from 7.3 to 9.3 . Ten 43 sampling sites were selected at each location, and flies were sampled at each site by ten 44 pitfall traps with an interval of 5 m in August 2009. Flies were identified in family level. 45 Family temperature index was used to calculate community temperature indexes (CTI) of six 46 locations. Through the sampling, 8526 flies in 41 families were collected. Phoridae was most 47 abundant in six locations. Sciaridae, Muscidae, Heleomyzidae, and Mycetophilidae were also 48 abundant. Values of CTI were positively related with values of MAT. Difference of 49 temperature between locations influenced more on the family composition of fly assemblages 50 compared to distance (km) between locations. These findings indicate a significant influence 51 of temperature on fly assemblages. 52 53 Key words : fly, family temperature index, community temperature index, climate warming, 54 pitfall trap 55 MANUSCRIPT 56 Introduction 57 58 Global warming has significantly influenced biota in the world such as range shifts, 59 phenological shifts, shits in species composition, and shifts of interactions (Parmesan and 60 Yohe 2003). Range shifts have been mostly studied among the climate change researches, but 61 its results were mainly obtained from the well-studied groups with long-term robust data such 62 as birds, plants, and butterflies. The range shifts are likely to lead to change of biotic 63 assemblages, but the studies on community levels were rare compared to range shifts (Kwon 64 et al 2010). In addition, most studies on the climate change influences are conducted in 65 Europe and NorthACCEPTED America. Therefore, influences of climate warming on the ecosphere 66 remains uncertain due to the limitation of studied groups and regions. 67 Insects are most diverse organisms in the world, comprising 56% of total species 68 (McGavin 2001). As insects are cold-blooded animal, growth and performance of insects are 69 heavily dependent on change of temperature. In addition, insects perform main ecological 70 functions such as pests, natural enemies, nutrient cycling, pollination, food production for ACCEPTED MANUSCRIPT 71 animals in the terrestrial and aquatic ecosystem (McGavin 2001). Therefore, studies on the 72 influence of climate warming on insects are very important to understand and predict change 73 of ecosphere in the warming future. However, majority of studies are on the handful flagship 74 insects such as butterflies. Butterflies are hervivorous insects, and heavily dependent on 75 plants. Therefore, it is questionable that butterflies can represent the overall response of 76 insects to climate warming. Flies (Diptera) are one of most diverse insects groups, and they 77 conduct various ecological functions such as detritivores, herbivores, pollinators, parasites, 78 and predators. Therefore, flies are likely to be one of candidate insects to represent the overall 79 response of insects (Hövemeyer 2000). 80 Despite the ecological importance of Diptera, studies on fly assemblages are rare due to the 81 difficulty of species identification of flies (Frouz 1999). Therefore, studies of dipteran 82 assemblages are mainly conducted in Europe where species identification of Diptera is well- 83 established. Although 66 families are listed in the checklist of Korea insects (Paek et al 2010), 84 the non-taxonomists are nearly impossible to identify species in most dipteran families due to 85 the scarcity of species identification resources. In reality, only family specialists (taxonomists) 86 can identify species of each family. Even the dipteran taxonomists are only capable to 87 identify species in his (or her) and related families.MANUSCRIPT Therefore, studies in family level can be 88 alternative to the difficult species level studies of flies. 89 This study aimed to examine relationship between family composition of fly assemblages 90 in forests and temperature. I sampled flies with standardized method (pitfall traps) in six 91 forested areas with different thermal conditions ranging from 7.3 to 9.3 . Pitfall traps 92 are an easy and efficient method to collect flies, but this method is not widely used for 93 ecological studies of fly assemblages. However, two studies (Kwon et al 2013, Lee et al 2015) 94 suggest that pitfall traps are very efficient to collect diverse flies. In recent, community 95 temperature index (CTI) using species temperature index (STI, species optimum temperature) 96 is increasingly used to study range shifts of butterflies, birds, and plants in Europe. Each 97 family would haveACCEPTED family-specific temperature optima according to their evolution origin and 98 family specific range. Therefore, the family temperature index can be also used to calculate 99 CTI using family composition of fly assemblages. I used data of 66 family temperature index 100 provided by Lee et al (2015), and examined the relationship between CTI using FTI and 101 temperature. In addition, I examined whether distance between locations or temperature 102 difference between locations determined more the family composition of fly assemblages. ACCEPTED MANUSCRIPT 103 Material and Methods 104 105 Sampling of flies was once conducted in August 2009 in six uphill passes (locations) of the 106 Baekdudaegan mountain range (Table 1). Ten sampling sites were selected at each location: 5 107 sites in northern forests from motor road, and 5 sites in southern forest. The sites are located 108 in healthy deciduous forests (S, J, E, U, Y) or larch plantation (H). Tree ages were 30-40 109 years old with 8-15 m heights. Understory vegetation (herbs and shrubs) was well developed 110 and litters covered the surface. Five sites (each five in northern side from road and each five 111 in southern side) were selected at 10 m, 50 m, 100 m, 200 m, and 300 m away from motor 112 roads. The distance was not linearly measured, but crudely measured by number of walks 113 (about 50 cm per a walk) along the trail. Ten pitfall traps were linearly set up (flush with 114 surface) with 5 m interval at each site, and lasted for ten days. The survey design was 115 graphically provided by Kwon (2015). Polyethylen glycol was used for preservative, and 116 filled one third of each trap. Fly bodies were sorted and stored in 100% ethyl alcohol. Fly 117 adults were identified to family by the adult family keys (Papp and Shumann 2000, 118 Triplehorn and Johnson 2005). The fly samples were deposited at the insect diversity 119 laboratory of the National Institute of Forest Science.MANUSCRIPT 120 Mean annual temperature of six locations was measured by the multiple regression model 121 of Kwon et al. (2012) from data of latitude, longitude, and altitude. Determination index (R 2) 122 of the regression temperature model is 91%. Non-metric multidimensional scaling (NMDS) 123 was used for ordination of fly assemblages. Data of 60 sampling sites (10 sites per location) 124 were used for NMDS. Dissimilarity of fly assemblages (pooled 10 sites) between locations 125 was calculated by the Bray-Curtis index (Oksanen et al 2015), and distances (km) between 126 locations was measured by Google Earth (http://www.google.com/intl/ko/earth). Family 127 temperature index ( ) was provided by Lee et al. (2015), in which the fly sampling was 128 conducted at 299 forest sites in the whole region of South Korea using pitfall traps. Family 129 temperature indexACCEPTED was average of MATs of recorded sites in the national fly survey. 130 Community temperature index ( ) of fly assemblages was calculated by the following Μ 131 equation, CTI = (∑ΗͰͥ Ai ∗ Fi )/N, where Ai = abundance of i family, Fi = FTI value of i family, 132 and N=number of total individuals. Adonis was used to examine influence of environmental 133 factors on family composition of fly assemblages (Oksanen et al 2015), and ANCOVA was 134 used for abundance (number individuals) and richness (number of families).