The Microclimates and Characteristics of Roost Buildings of Bats (Chiroptera) In Urban Area of West Coast Of Sabah

Ehsan Hadi1, Nurul Najmaini1, Abdul Hamid Ahmad1 and Azniza Mahyudin1
1

Institute for Tropical Biology and Conservation, Universiti Malaysia Sabah.

Abstract This study was design to identify species of bat that roost in building and to collect information about the characteristics of the roosts in urban area of West Coast of Sabah. Bats were found opportunistically and examined in six buildings with a total of 11 roosting sites. Three species of bats were identified in the buildings, namely: 1) Taphozous melanopogon; 2) Taphozous saccolaimus and 3) Scotophilus kuhlii. Among these three, Taphozous melanopogon is the most abundant species. Bats were found roost in four types of buildings; commercial, mosque, tower and residential house. They roost on walls, attics, crevice and ceiling. Generally, rooting sites are situated high from ground and relatively covered from weather. Roosting sites also found located within radius 1 km distance from water resource and forest remnant. At the temperatures scale, roosting sites are characterized by warm temperature. However, the temperatures variables inside roosting sites were varied suggest that structure of buildings influenced the temperature. The roosting sites of Taphozous melanopogon and Taphozous saccolaimus were significantly warmer than Scotophilus kuhlii. Despite that, the temperatures occurred in Taphozous melanopogons and Taphozous saccolaimus roosting sites were relatively similar.

Keywords: Bats; Urban; Roost building; Roost characteristics; Roost temperature

1. Introduction Bats utilize a wide variety of structures as roost. Naturally they use trees, caves and overhangs. However, in urban environment where the natural roosts are scarce, bats might opt for many different types of roost including man-made structures such as bridge, tower, attic and even the hollow floor spaces of house as substitutes for tree cavities or caves (Altringham, 1996). Among those structures, the used of building as roost by bats is the most well-documented (Riskin and Pybus, 1998, Jenkins et. al, 1998, Kunz, 1982). During the past 40 years, evidences that certain species of bats is well adapted to building were recorded in increasing amount of literatures. Many species such as Taphozous melanopogon, Eptesicus fuscus and Pipistrellus pipstrellus whose originally roost in the caves has been link almost exclusively to building (Kunz, 1982).

The roost selection in bats involves many mutually important factors. Generally, bats choose area that has lower predation risk, reduced ectoparasite infection, higher from ground (Psyllakis and Brigham, 2006), and relatively close to water resource (Evelyn et al., 2004) and foraging area (Kunz, 1982). Thermal stability also one of the most important factor in bats roost selection. Lausen and Barclay (2006) explained the main reason for bats occupied human settlements is because of thermal advantages where warm temperatures can reduce energy expenditure and also offer benefit in term of productivity.

Despite the consistency of published studies, there is still paucity of roosting characteristics for many species especially around human settlement (Evelyn et al., 2004). Besides, far too little attention has been paid to the roosting information of bats in tropics region (Rodriguez-Duran and Soto-Centeno, 2003). Many studies have been conducted but mainly in temperate countries. Recent
2

research has identified roost as the most crucial element in bats conservation. Therefore, such information is urgently needed as the bats remain unprotected in many parts in the world.

This research was conducted to identify the species of bats that occupy the buildings in urban area of Kota Kinabalu, West Coast of Sabah. This study also examines the characteristics and microclimate of the roost. More understanding about the animals needs and their behaviour may be able to provide solution for human-bats conflicts since bats always regarded as nuisance and frequently persecuted.

2. Material and Methods Random surveys were conducted from April 2008 to June 2008 to locate possible bats roosting site in Kota Kinabalu. Residents, or workers of buildings were interviewed to get the information about bats occurrence. Positive feedback was then followed by examination of the roosting sites to confirm the existence of bats in the buildings.

2.1 Buildings Characteristics Seven variables were used to characterize roost site: temperature (C); building types, i.e house, factory, commercial/ species (if trees); Wall material (wooden, bricks etc), canopy cover (%); distance from brackish/ fresh water source (m); distance from forest remnant (m); roost height from the ground (m).

2.2 Microclimate Variables The microclimate of the sites was described with temperature used as a measurement. Microclimate was recorded using data loggers (Onset HOBO H8 Pro series, temperature range -30C to +50C). To reduce disturbance to roost, the data loggers were positioned 1 to 3 m away from roost.

Data were averaged per hour and categorised into hourly intervals over each period of 24-h period. To calculate variables that would show variation and pattern of 24-h temperatures, the hourly data were used to calculate: average temperature, temperature fluctuation (hourly); maximum; minimum; range of day; time of day maximum temperature occurred; and length of time maximum temperature of a day was maintained. The readings for each roosting sites were then compared.

2.3 Data Analysis Each variable was tested for normality using Kolmogorov-Smirnov test and most of variables were not normally distributed and could not be transformed successfully. Thus non-parametric tests were used in this study.

To test differentiation of temperatures between roosting sites, Kruskal-Wallis One Way Analysis of Variance test was performed on five temperatures variables: average temperature, fluctuation, range of temperature, time of maximum temperature occurs and length of time maximum temperature was maintained.

Each of temperatures variables had expected to be different between roosting sites (at least one). The differences were suspected to be influenced by the buildings construction materials. Therefore, a comparison using Mann Whitney U-Test (p<0.05) was drawn to inspect the effect of construction material to roost temperatures. Mann-Whitney U Test was also used to find differentiation of temperatures variables according to species (Bonferroni Correction at 3 comparison, p<0.017).

3. Result During survey, eleven roosting sites were found in 6 roost buildings (Figure 1.). Four roosting sites were located in the City Mosque (CM), two in Wisma Kosan (WK), two in Telipok Health Clinic (THC) and one in each Wisma Merdeka (WM), State Mosque (SM) and Wisma LLT (WLLT). All buildings built with concrete and bricks except THC which is a wooden house. CM, WK and THC are located in suburban areas while the other roosting sites are in urban areas. There are 3 species occupied these roosts: Taphozous melanopogon (>800 individuals), Taphozous saccolaimus (18) and Scotophilus kuhlii (11). From these three, Taphozous melanopogon was the dominated species in this study and occupied all roost building except THC. Taphozous melanopogon was later regarded as T. melanopogon, Taphozous saccolaimus as T. saccolaimus and Scotophilus kuhlii as S. kuhlii.

Figure 1: Roost buildings in urban area of Kota Kinabalu, West Coast of Sabah

3.1 Roost Buildings All roost buildings are older than 10 years. WLLT, WM and WK are classified as commercial building, CM and SM as place for worship, while THC is a residential house. A higher proportion of the 11 specific roosts were found in wall (45.5%), others in attic (36.4%), crevices (9%) and ceiling (9%). All roost was communal and no solitary roost was found. Roosts are high from ground (med=16.90), highly covered (med=100.00) and located within 1 km from water (med=116.9) and forest remnants (med=268.89). Interestingly, all roost is surrounded by open space. Table 1 described the physical characteristics of the buildings.

Table 1. The characteristics of the roost buildings found in Kota Kinabalu. 6

Roost sites

Roost type

Species

Roost height

Entrance direction SE SE SE SE SE SE SE SE SE NW SW

Canopy cover 100 (%) 100 100 100 100 100 40 80 100 100 100

Distance from water 26.35 (m) 30.35 48.64 47.01 343.95 343.95 872.91 297.46 30.17 219 222

Distance from forest 556.58 (m) 556.58 575.38 592.18 76.48 72.24 268.89 574.74 260.75 33 33

CM1 CM2 CM3 CM4 WK1 WK2 WM SM WLLT THC1 THC2

Wall Wall Ceiling Wall Wall Wall Crevice Attic Attic Attic Attic

T. melanopogon T. melanopogon T. melanopogon T. melanopogon T. melanopogon T. melanopogon T. melanopogon T. melanopogon T. melanopogon T. saccolaimus S. kuhlii

5.6 (m) 9.3 45.4 20 21.6 16.9 4.5 67.6 18 6.3 7.1

SE=southeast, NW=Northwest and SW=southwest.

3.2 Temperatures Overall, most site recorded temperature above 27C. Highest temperature was 51.04C while the lowest was 23.24. Kruskal-Wallis Analysis showed that there were some variations of temperatures variables between roosting sites. Among all roost, CM1 was regarded as most stable site as it have lowest range (med=1.98) and maintained highest temperature longer (med=7:00) (Table 2).

3.3 Concrete Buildings versus Wood The temperature inside concrete roost buildings were more stable than wood building. The temperature in concrete were fluctuated slower than wood. The range of temperature occurs in concrete buildings were significantly lower and minimum temperature during night were
7

significantly warmer than wood building. Concrete roost buildings also reached maximum temperature earlier and it can hold up maximum temperature much longer compare to wood structures. Although, mean rank of temperature and maximum temperature during daylight (between these two sites) was not significantly differed (Table 3).

Table 2. Comparison of Temperatures Variable Between Roosting Sites.

CM1 CM2 CM3 CM4 WK1 WK2 THC1 THC2 WM SM WLLT H df P value

Temperature 27.91 (0.79) 27.91 (1.43) 27.5 (6.81) 27.91 (1.12) 28.31 (2.58) 26.67 (3.75) 27.85 (4.62) 27.39 (4.8) 27.33 (4.29) 27.56 (2.1) 29.5 (1.61) 728.41 10 p<0.05

Fluctuation 0.30 (0.41) 0.32 (0.44) 1.18 (1.36) 0.25 (0.32) 0.75 (0.88) 0.72 (1.00) 0.65 (0.94) 0.72 (0.98) 0.83 (1.42) 0.70 (0.72) 0.53 (0.70) 50.49 10 p<0.05

Range 1.98 (0.78) 2.36 (2.02) 23.17 (6.44) 2.36 (0.41) 4.80 (1.41) 6.73 (1.56) 8.24 (2.58) 8.47 (3.43) 7.91 (2.11) 3.57 (1.58) 2.40 (1.19) 268.40 10 p<0.05

LMT 7:00 (10:00) 5:00 (4:15) 1:00 (1:00) 5:00 (5:15) 3:30 (2:15) 2:00 (3:00) 2:00 (2:00) 2:00 (1:15) 3:00 (3:00) 4:00 (4:15) 6:00 (4:15) 110.43 10 p<0.05

TMO 11:00 (3:00) 11:00 (2:00) 10:00 (0:15) 12:00 (2:00) 12:00 (1:15) 12:00 (2:00) 11:00 (3:15) 13:00 (3:00) 10:00 (3:15) 13:00 (3:15) 13:00 (3:00) 83.29 10 p<0.05

Data presented as medians (interquartile range). H, Kruskal-wallis value; LMT, length of maximum temperature; TMO, time of maximum temperature occur.

Table 3. Comparison of temperatures variables of construction materials and species

Materials Day (Max) Night (min) Temperature Fluctuation Range LMT TMO Species Temperature Fluctuation Range LMT TTO Species Temperature Fluctuation Range LMT TTO Species Temperature Fluctuation Range LMT TTO

Concrete 29.10 (2.40) 27.12 (1.57) 27.91 (2.26) 0.23 (0.33) 3.57 (4.53) 4:00 (4:00) 12:00 (3:00) Tm 27.91 (2.26) 0.54 (0.78) 3.57 (4.53) 4:00 (4:00) 12:00 (3:00) Tm 27.91 (2.26) 0.54 (0.78) 3.57 (4.53) 4:00 (4:00) 12:00 (3:00) Ts 27.85 (4.62) 0.65 (0.94) 8.24 (8.28) 2:00 (2:00) 11:00 (3:15)

Wood 31.50 (4.47) 26.34 (1.56) 27.69 (4.71) 0.45 (0.77) 8.38 (2.85) 2:00 (2:00) 12:00 (3:00) Ts 27.85 (4.62) 0.65 (0.94) 8.24 (8.28) 2:00 (2:00) 11:00 (3:15) Sk 27.39 (4.80) 0.72 (0.98) 8.47 (9.59) 2:00 (1:15) 13:00 (3:00) Sk 27.39 (4.80) 0.72 (0.98) 8.47 (9.59) 2:00 (1:15) 13:00 (3:00)

U 4494196 4400301 4545969 2797536 3090 4826 6729.50 U 2201043 1443815 1591.5 2549 3826 U 2158848 1353721 1498.5 2277 2903.5 U 230613.5 242028 414 405 338

p value ns p<0.05 ns p<0.05 p<0.05 p<0.05 p<0.05 p value ns p<0.017* p<0.017* p<0.017* ns p value p<0.017* p<0.017* p<0.017* p<0.017* p<0.017* p value p<0.017* ns ns ns ns

Effect size (R ) -0.014 -0.027 -0.006 -0.261 -0.089 -0.056 -0.024 Effect size (R ) -0.02 -0.18 -0.33 -0.19 -0.03 Effect size (R ) -0.03 -0.20 -0.34 -0.23 -0.15 Effect size (R ) -0.04 -0.02 -0.05 -0.04 -0.10

Data presented as medians (interquartiles range) where Mann-Whitney U tests were used. *Significant level were adjusted at p<0.017 (Bonferroni correction). Tm = T. melanopogon, Ts = T. saccolaimus, Sk = S. kuhlii.

3.5 Species Roost Temperature Overall, roost temperature of S. kuhlii is significantly lower than the other two species. Despite that, the roost temperature of T. melanopogon and T. saccolaimus was not significantly differed (Table 3).

4. Discussion
9

In this study, roosts were found exclusively in buildings that shared common characteristics. All roost building aged more than 10 years and historically, showed no sign of major modification. The walls were not smooth with some cracks; allowing bats to cling. Bats accessed the interior buildings through windows, doors or attic which were opened all the time. Moreover, the existence of ruined rough bricks or concrete which comparable to crevices; create a good roosting habitat for bats.

4.1 Species distribution Only three species of bats were found in this study. The low number of species is expected as species richness, abundance and activity of bats were reduced in disturbed habitats (Hourigan et al., 2006; Lumsden and Bennett, 2002; Kirsten and Klomp, 1998). The species in urban ecosystem are also dominated by one or a few abundant species (Furmankiewicz and Grniak, 2002). This is because not many species of bats have the ability to survive in open and edge habitat. Therefore, the community structure of bats may have changed; particularly for microchiropteran which have narrower and specialize habitat requirements. This may also explain why the number of species found in buildings was low (See Hourigan et al., 2006; Kurta and Teramino, 1992). T. melanopogon was the most abundant species found in this research (>800). Discovered roost buildings also dominated by this species revealed that T. melanopogon has successfully adapted to the current environment in this city. Roost buildings which were occupied by T. melanopogon were found distributed along coastal area. This record was different from study of Bates and Harrison (1997) in India, which found T. melanopogon occur near fresh water areas such lakes, rivers and ponds.

10

Conversely, T. saccolaimus and S. kuhlii showed a restrictive distribution in this study. This finding is in contrast to what was found by previous studies. Both species are intimately associated with modified habitats especially S. kuhlii which were known to occur in all type of mosaic particularly urban or suburban parklands or rural areas (Pottie et al., 2005). A possible explanation for this might be that both species prefer to roost in rural area rather than sub urban or urban mosaic. This result also fit with population status of this species which is known to be in low density in Sabah (Yasuma et al., 2003).

4.2 Roost Characteristics The differences in roost distribution revealed that there are might be some flexibility and preferences behaviour among these species. However, the roost characteristics might be underestimated for T. saccolaimus and S. kuhlii because of the small sample size. Nevertheless, in general rooting sites are located high from ground and relatively covered from weather. The temperatures in all roosting sites were also warm. Roosting sites also found located within radius 1 km distance from water resource and forest remnant.

4.3 Temperature Overall, all species found in this study roost in warm roosting site. Preference of bats toward warmer temperature has been observed in several studies in other tropical area (Table 4). However, the factors guiding bats in this study to roost in warm site are poorly known. To date, for other species high roost temperatures offer several advantages to bats particularly reproductive females. Roosting in warmer roost enables reproductive females to save energy maintaining their body temperature and use them for reproduction purposes. It has been hypothesized that warm
11

roosts accelerate embryonic development (which resulted in early birth) and increase milk production.

Table 4. Research investigating roost temperature in tropical bats.

Bat species T. melanopogon T. saccolaimus S. kuhlii P. quadridens E. sezerkoni L. silvicolum C. brevicauda

Type Building Attic Attic Cave Cave Termite nest Cave

Temperature (C) 28.3 2.7 28.7 2.9 28.3 3.1 30.2 4.6 28.2 2.3 27.9 1.0 19.2 2.7
12

Country Malaysia Malaysia Malaysia West Indies West Indies Panama Mexico

Source This study This study This study Rodriguez-Duran & Soto Centeno (2003) Rodriguez-Duran & Soto Centeno (2003) Dechmann et al., (2004) Avila Flores & Medellin

B. plicata A. jamaicensis E. alba

Cave Cave Tents

26.7 3.1 21.9 2.6 23.3 0.7

Mexico Mexico Costa Rica

(2004) Avila Flores & Medellin (2004) Avila Flores & Medellin (2004) Rodriguez-Herera et al., (2008)

Temperatures data presented in mean (SD).

The high temperature also benefits young bats which had low thermoregulating ability. During early birth, when their mother was away warm roosts help them maintain their body temperature. When they are able to regulate their body temperature, warm temperature lower the thermoregulatory cost and will increase their growth rates (Lausen and Barclay, 2006; Kerth et al., 2001). In temperate, this strategy was used to increase winter survival (Kunz, 1982).

4.3.1 Thermal Variation among Roost The variation in microclimate found among roost buildings (Table 2) indicated that temperature was influenced by the diverse microstructures or substrates of buildings; such as crevices, walls, attics, wood, bricks or other features of the buildings architecture. Some structures may expose more to solar radiation or can retain heat longer than others. Besides, the ventilation system also contributed to temperature fluctuation in the buildings. The existence of vent-holes and peripherals such as ventilator could avoid heat entrapment.

The bats also were most common roost in concrete buildings. One reason bats dwelled in this type of buildings might be because of the compacted stable temperature and ability to retain heat longer.
13

Similarly, the bats preferences for bricks or concrete structures have been seen in several past studies (Lausen and Barclay, 2006; Lance et al., 2001). In Sweden during winters, bats occupied concrete bat boxes more common than wooden. The research of exposed roost in Alberta also found that most bats roost on bricks which suggest bricks or concrete may have higher capacity of trapping heat (Riskin and Pybus, 1998).

4.3.2 Temperatures Difference among Species Microclimate variation also can encourage colonization by several species (Hill and Smith, 1985) and may influence pattern association of bats in certain roosts (Rodriguez and Soto-Centeno, 2003). Despite wide range of temperatures recorded in many roost buildings, only T. saccolaimus and Scotophlus kuhlii were found sharing roost. The reason anyhow might be related to the condition site of study which concentrated around urban environment that known to have limited number of species and the limited number of available roosts.

Roost selection is a complex processes. It influenced by many factors that mutually dependent between each other. Rodriguez-Duran (1995) suggested the occurrence of certain species in a given cave was determined by physiology of the bats which influenced by body mass, diet and microclimate condition. For example, Baudinette (2000) found microchiropterans species in a cave in Australia preferred different temperatures and humidities. Similarly, the study of bats in West Indies also revealed that species of Pteronotus quadridens which is smaller species, roost in warmer site compare to Erophylla sezekorni although they roost in the same cave (RodriguezDuran and Soto-Centeno, 2003).

14

Suprisingly, the result of this study was inconsistent with the previous studies. S. kuhlii which is smaller, roost in cooler site compare to T. saccolaimus and T. melanopogon. The inconsistency maybe due to morphological features of S. kuhlii. This species longer tail is probably significant in conserving heat as seen in Lasiurus borealis (Davis and Reite, 1967). The comparison of other temperatures variables also showed that T. saccolaimus is closely related to S. kuhlii. This similarity maybe explained by the fact that they shared common preference for roost type; that is in attic and hollow trees (Yasuma et al., 2003).

4.4 Canopy Cover and Surrounding Areas The finding of this study indicated most roosting sites are situated in highly covered area (average 92.7 %). Most bats roost under roof except WM which is located in the narrow slit between two buildings. In natural roosting site, several other bat species also have been shown to roost preferentially in highly covered roost (Table 5).

Table 5. Research investigating roost canopy cover in tropical bats and other part of the world.

Bat species T. melanopogon T. saccolaimus S. kuhlii E. alba M. molossus D. rotundus N. albiventris T. georgianus H. diadema H. ater

Canopy cover (%) 91.1 (n=9) 100 (n=1) 100 (n=1) 83.1 (n=50) 81 (n=5) 62.7 (n=7) 95 (n=1) 53.5 (n=16) 48.5 (n=1) 48.5 (n=6)

Country Malaysia Malaysia Malaysia Costa Rica Australia Bolivia Bolivia Australia Australia Australia

Source This study This study This study Rodriguez-Herera et al., (2008) Aguirre et al., (2003) Aguirre et al., (2003) Aguirre et al., (2003) Milne et al., (2005) Milne et al., (2005) Milne et al., (2005)

Data presented as mean, (n=number of roost). 15

A highly covered roost can protect bats from extreme weathers and avian predators. It also provides bats with more stable temperature (Kunz, 1982). The finding of this study seems to be consistent with the theory. The temperatures fluctuations of most roosting site were very low with an average 0.51 C. Only CM3 were considered as less stable with high range of temperature (Table 2).

On the other hand, the surrounding roosts are open areas. The reasons of this preference are not fully understood. Studies for other species such as Chalinolobus tuberculatus, Eptesicus fuscus and Myotis bechsteinii suggest the main reasons why bats prefer open space were due to their ecomorphology, predator avoidance, ease at roost relocation and thermal benefits (Vonhof and Gwilliam, 2007; Psyllakis and Brigham, 2006; Sedgeley and ODonnell, 1999).

Species that was found in this study is consistent with ecomorphological theory. All three species in this study have been described as having high wing loadings (Yasuma et al., 2003). The morphology allows these species to fly fast in open areas, above treetops and along forest gaps. However, this feature reduced the flight maneuverability and may limit bats from accessing cluttered habitats (Wei et al., 2008). Therefore, roosts which are far from ground in open spaces would facilitate bats that lack maneuverability (Sedgeley and ODonnell, 1999).

Open spaces also allow bats to detect predators and this area most likely receive more direct sunlight compared to dense area. Exposure to solar radiation produced warmer temperatures which

16

could benefit bats in many ways (Perry et al., 2007). The surrounding habitats might be the cause dictating the temperatures in the roosting sites.

4.5 Higher from ground Another important finding is those roosts located high from the ground (at least 4.5 m above the ground). This characteristic was reviewed as one of main important variables in roost selection (Agosta, 2002). High roost is commonly selected by bats because it can avoid terrestrial predators such as snake, rats and etc. However, in urban area, cats and dogs are more abundant.

Domestic animals were known to prey bats in many studies (Sedgeley and ODonnell, 2004; Riskin and Pybus, 1998). In this study, these animals were found abundant in the study sites. Cats were also seen tried to catch bats when they forage close to ground and during bats-released from mist nets in THC. Taller roost also accommodates bats to gain velocity during early flight. This roosts characteristic is particularly important for bats that are in less maneuverable conditions such as pregnant adults, adults moving non-volant young and young bats during their first flight (Norberg and Rayner, 1987). Roosts that have greater height from ground are also more exposed to solar radiations which offer warmer condition to bats (Russo et al., 2004). The preference of bats for high roost to gain microclimate advantage by choosing higher roost was confirmed in this study which was observed in T. melanopogon in CM3. Many other research finding also has showed preference of bats to roost in high roosting site. However, it is interesting to note that roost height in this study was unusual. Mean roost height found is 20.2 m and the maximum roost height recorded is 67.6 m; both exceeds those all recorded in other published studies (Table 6).
17

Table 6. Research investigating roost height in tropical bats and other part of the world.

Bat species T. melanopogon T. saccolaimus T. saccolaimus S. kuhlii S. kuhlii E. alba S. lilium E. furinalis N. albiventris N. noctula A. lituratus

Type Building Attic Tree Attic Building Tree Tree Tree Tree Building Tree

Roost height (m) 23.2 (n=9) 6.3 (n=1) 7 (n=1) 7.1 (n=1) 12 (n=1) 1.3 - 2.6 (n=26) 2.1 to 7.9 (n=8) 9.5 (n=2) 8.4 (n=1) 7.5 (n=131) 8.1 (n=15)

Country Malaysia Malaysia Australia Malaysia Philippines Costa Rica Belize Bolivia Bolivia Hungary Venezuela

Source This study This study Schulz & Thomson (2007) This study Rickert et al., (1989) Timm & Mortimer (1976) Fenton et al., (2000) Aguirre et al., (2003) Aguirre et al., (2003) Bihari (2004) Munoz-romo et al., (2007)

Data presented as mean, (n=number of roost).

4.6 Distance from Water Source Being close to permanent water resource is always associated with roost selection by many species of bats (Rabe et al., 1998; Entwistle et al., 1997). The finding of this study seems to be consistent with other research. Compare to other studies, roosting sites in this study are located relatively close to water (<1 km). However, the close species of T. melanopogon and T. saccolaimus (T. kapalgensis and T. georgianus in Australia) were described to roost considerably far from their roosting site (table 7).

Table 7. Research investigating roost distance from water source in tropical bats and other part of the world.

Bat species

Distance from water (km)

Country

Source

T. melanopogon T. saccolaimus S. kuhlii Myotis spp B. barbastellus M. yumanensis T. georgianus T. kapalgensis H. diadema H. ater

0.23 (n=9) 0.22 (n=1) 0.22 (n=1) 0.48 (n=22) 1.79 (n=33) 0.13 (n=20) 2.7 (n=16) 6.6 (n=4) 1.6 (n=1) 1.6 (n=1)

Malaysia Malaysia Malaysia Canada Italy USA Australia Australia Australia Australia

This study This study This study Jung et al., (2004) Russo et al., (2004) Evelyn et al., (2004) Milne et al., (2005) Milne et al., (2005) Milne et al., (2005) Milne et al., (2005)

Data presented as mean, (n=number of roost).

Roost near water body offers bats benefit in several ways. First, the insect availability is usually abundant of near water. Hence, by roosting near water, bats are close to prospective foraging location (see Evelyn et al., 2004). Second, waterway provides bats open a flight passage which is benefit most fast flier bat. However, water bodies that lack of vegetation are often avoided by bats. It is believed that loss of buffer zone resulted in loss of invertebrate diversity and abundance (see Russ and Montgomery, 2002).

In this study, all three species are described as fast fliers with high wing loadings and high aspect ratios. It also means these species lack maneuverability. Therefore, the water bodies might be useful as foraging site for these species as they prefer to forage in open space. However, exploitation of water body as foraging site only has been seen in T. melanopogon near CM and WLLT.

4.7 Distance from forest To date, distance from forest plays a major role in roost selection. In general closer to forest offer several benefits to roost dwellers. With more complex structure of vegetation, forest remnants offer
19

bats variation and abundant of insects. Even small or poor quality remnants harbour resource for bats particularly where large forests are rare (see Law et al., 1999). Therefore, by being close to forest bats could reduce energetic consumption during flight to the foraging sites (Jenkins et al., 1998).

Roosting near forest also may protect bats from avian predators. The presences of trees in forest allow bats to emerge with low detection of predators (Russo et al., 2007). There is several other bat species have been shown to roost preferentially near forest remnant (Table 8).

Table 8. Research investigating distance of roost from forest resource in other studies.

Bat species Distance from forest (m) T. melanopogon 392.6 (n=9) T. saccolaimus 33 (n=1) S. kuhlii 33 (n=1) C. tuberculatus 15 (n=1) Mystacinidae sp 220 to 530 (n=8) M. sodalis 14.2 (n=47) Data presented as mean, (n=number of roost).

Country Malaysia Malaysia Malaysia New Zealand New Zealand USA

Source This study This study This study Sedgeley & O' Donnell (1999) O' Donnell et al., (1999) Carter & Feldhamer (2005)

During investigating period, T. melanopogon were seen forage in open space near forest edge in two places: WK and WLLT while S. kuhlii also use the same space in THC. Unfortunately, this

20

study was unable to trace foraging site of T. saccolaimus. T. melanopogon and S. kuhlii which known to forage in open habitats also seems cant utilized benefits of forest remnants to its fullest.

These two species might be gain advantage of food supply at the corridor but, they are most likely not covered from predators because they are open space flier. The protection might just apply to low wing loadings bats that have slow flight but better maneuverability (e.g, Russo et al., 2007).

5. Conclusion Taken together, these results suggest that bats choose roost based on warmer temperature, high level from ground and high percentage of canopy cover. An implication of this is the possibility that any renovations done on the buildings should not change the microclimate. Human activities near roost area also at least must not disturb the animals. The current findings add substantially to our understanding that bats need certain type of roost which influences most aspect of their life. This research will serve as a base for future studies and more investigations are needed. Knowledge of bats metabolic rates relative to warmer roost will help to answer the questions that unsolved in this study. Further research using radiotelemetry also could help tracking the bats and consolidate the results of this study.

Acknowledgement We gratefully acknowledge and thank Mr. Hairul Hafiz Mahsol for his statistics advice. We thank Kota Kinabalu Health Office and manager of Wisma Merdeka for giving us permission to conduct
21

this research in their buildings. We also thank staff of City Mosque especially Hj. Gustin, Hj. Asli, Abdullah, Mr. Mudim and Mr. Rosli for their incredible assistant all through the studies that we did there. Further thank people who helped us during the fieldwork in Telipok Health Centre and Wisma Kosan.

References

Altringham J D. 1996. Bats biology and behaviour. Oxford University Press, United State.

Agosta S J. 2002. Habitat use, diet and roost preferences of the big brown bats (Eptesicus fuscus) in North America: a case of conserving an abundant species. Mammal Review 32: 179-198.

Aguirre L F, Lens L, Matthysen E. 2003. Patterns of roost use by bats in a neotropical savanna: implications for conservation. Bulletin of the American Museum of Natural History 111: 435-443.

Avila Flores R, Medellin R A. 2004. Ecological, taxonomic and physiological correlates of cave use by the mexican bats. Journal of Mammalogy 85(4): 675-687.

22

Barclay R M R. 1991. Population structure of temperature zone insectivorous bats in relation to foraging behaviour and energy demand. Journal of Animal Ecology 60: 165-78.

Bates P J J, Harrison D L. 1997. Bats of the Indian Subcontinent. Harrison Zoological Museum Publications, 258.

Baudinette RV, Churchill S K, Christian K A, Nelson J E, Hudson P J. 2000. Energy, water balance and the roost microenvironment in three australian cave-dwelling bats (microchiroptera). Journal of Comparative Physiolology B 170: 439446.

Bihari Z. 2004. The roost preference of Nyctalus noctula (Chiroptera, Vespertilionidae) in summer and the ecological background of their urbanization. Mammalia 68(4): 329-336.

Carter T C, Feldhamer G A. 2005. Roost tree use by maternity colonies of indiana bats and northern long-eared bats in southern Illinois. Forest Ecology and Management 219: 259-268.

Chruszcz, B. J., and Barclay, R. M. R., 2003. Prolonged foraging bouts of a solitary gleaning / hawking bat, Myotis evotis. Convergence, 826, 823-826.

Davis W H, Reite O L. 1967. Responses of bats from temperate regions to changes in ambient temperature. Biological Bulletin 132: 320328.

23

Dechmann D K N, Kalko E K V, Kerth G. 2004. Ecology of an exceptional roost: energetic benefits could explain why the bat Lophostoma silvicolum roosts in active termite nests. Evolutionary Ecology: 1037-1050.

Entwistle A C, Racey P A, Speakman J R. 1997. Roost selection by the brown long-eared bat Plecotus auritus. Journal of Applied Ecology Vol. 34 (2): 399-408

Evelyn M. 2004. Conservation of bats in suburban landscapes: roost selection by Myotis yumanensis in a residential area in California. Biological Conservation 115(3): 463-473

Fenton M B, Vonhof M J, Bouchard S, Gill S A, York N, Johnston D S, Reid F A, Riskin D K, Standing K L, Taylor J R, Wagner R. 2000. Roosts used by (Chiroptera:Phyllostomidae). Biotropica 32: 729-733. Sturnira lilium

Furmankiewicz J, Grniak J. 2002. Seasonal changes in number and diversity of bat species (Chiroptera) in the Stolec mine (SW Poland). Przyroda Sudetw Zachodnich Suplement 2: 49-70.

Hill J E, Smith J D. 1985. Bats, A Natural History. British Museum (Natural History), England, 53.

Hourigan C L, Johnson C, Robson S K A. 2006. The structure of a micro-bat community in relation to gradients of environmental variation in a tropical urban area. Urban Ecosystems 9(2): 67-82.

24

Hutson A M, Mickleburgh S P, Racey P A. 2001. Global Status Survey and Conservation Action Plan Microchiropteran Bats. IUCN, Gland, Switzerland and Cambridge, United Kingdom, 35-51.

Jenkins E V, Laine T, Morgan S E, Cole K R, Speakman J R. 1998. Roost selection in the pipistrelle bat, Pipistrellus pipistrellus (Chiroptera:Vespertilionidae) in northeast Scotland. Animal Behaviour 56: 909917.

Jung T, Thompson I, Titman R. 2004. Roost site selection by forest-dwelling male in central Ontario, Canada. Forest Ecology and Management 202(1-3): 325-335.

Kerth G, Weissmann K, Konig B. 2001. Day roost selection in female Bechsteins bats (Myotis:bechsteinii): a field experiment to determine the influence of roost temperature. Oecologia, 126(1): 1-9.

Kirsten I, Klomp N. 1998. Microchiroptera in urban, rural and forest areas of southern NSW. Australasian Bat

Kunz T H. 1982. Roosting Ecology of Bats. in Kunz, T. H. (ed.) Ecology of bats. 2-3.

Kurta A, Teramino J A. 1992. Bat community structure in an urban park. Ecography 15: 257-261.

25

Lance R F, Hardcastle B T, Talley A, Leberg P L. 2001. Day-roost selection by Rafinesques bigeared bats (Corynorhinus Rafinesquii) in Louisiana Forests. Journal of Mammalogy, 82(1): 166 172.

Lausen C L, Barclay R M R. 2006. Benefits of living in a building: the big brown bats (Eptesicus fuscus) in rocks versus buildings. Journal of Mammalogy 87(2): 362-370.

Law B S, Anderson J, Chidel M. 1999. Bat communities in a fragmented forest landscape on the south-west slopes of New South Wales, Australia. Biological Conservation: 88.

Lumsden L, Bennett A. 2005. Scattered trees in rural landscapes: foraging habitat for insectivorous bats in south-eastern Australia. Biological Conservation 122(2): 205-222.

Milne D J, Armstrong M, Fisher A, Flores T, Pavey C R. 2005. Structure and environmental relationships of insectivorous bat assemblages in Tropical Australian Savannas. Austral Ecology 30: 906-919.

Munoz-Romo M, Herrera E A, Kunz T H. 2007. Roosting behavior and group stability of the big fruit-eating bat Artibeus lituratus (Chiroptera:Phyllostomidae). Mammalian Biology 73(3): 214221.

26

Norberg U, Rayner J M V. 1987. Ecological morphology and flight in bats (Mammalia: Chiroptera): wing adaptations, flight performance, foraging strategy and echolocation. Philosophical Transactions of the Royal Society of London, B 316: 335427.

O Donnell F J, Christie J, Corben C, Sedgeley J A, Simpson W. 1999. Rediscovery of short-tailed bats (mystacina sp.) In Fiordland, New Zealand : preliminary observations of taxonomy, echolocation calls, population size, home range and habitat use. Society 23: 21-30.

Perry R W, Thill R E. 2007. Roost selection by male and female northern long-eared bats in a pine dominated landscape. Forest Ecology and Management 247: 220-226.

Pottie S A, Lane D J W, Kingston T, Lee Y H. 2005. The microchiropteran bat fauna of Singapore. Acta Chiropterologica 7(2): 237-247.

Psyllakis J, Brigham R. 2006. Characteristics of diurnal roosts used by female Myotis bats in subboreal forests. Forest Ecology and Management 223(1-3): 93-102. Rabe M J, Morrell T E, Green H, deVos J C, Miller C R. 1998. Characteristics of ponderosa pine snag roosts used by reproductive bats in Northern Arizona. Journal Wildlife Management 62: 612 621.

Rancourt S, Rule M, O Connell M. 2007. Maternity roost site selection of big brown bats in ponderosa pine forests of the Channeled Scablands of Northeastern Washington State, USA. Forest Ecology and Management 248(3): 183-192.
27

Rickert E A, Heideman P D, Utzurrum R C B. 1989. Tents roosting by Scotophilus kuhlii in the Philippines. Journal of Tropical Ecology 5: 433-436.

Riskin D K, Pybus M J. 1998. The Use of Exposed Diurnal Roosts in Alberta by The Little Brown Bat, Myotis lucifugus. Canadian Journal of Zoology 76, 767-772.

Rodrguez-Durn A. 1995. Metabolic rates and thermal conductance in four species of neotropical bats roosting in hot caves. Comparative biochemistry and physiology. Part A, Physiology 110(4): 347-55.

Rodriguez-Duran A, Soto-Centeno J A. 2003. Temperature selection by tropical bats roosting in caves. Journal of Thermal Biology 28(6-7): 465-468.

Rodrguez-Herrera B, Medelln R A, Gamba-Rios M. 2008. Roosting requirements of white tentmaking bat Ectophylla alba (Chiroptera:Phyllostomidae). Acta Chiropterologica 10(1): 89-95.

Russ J M, Montgomery W I. 2002. Habitat associations of bats in Northern Ireland : implications for conservation. Journal of Wildlife Management 108: 49-58.

Russo D, Jones G, Migliozzi A. 2002. Habitat selection by the Mediterranean horseshoe bat, Rhinolophus euryale (Chiroptera: Rhinolophidae) in rural area of Southern Italy and implication for conservation. Biological Conservation 107: 71-81.
28

Russo D, Cistrone L, Jones G, Mazzoleni S. 2004. Roost selection by Barbastelle Bats (Barbastella barbastellus, Chiroptera:Vespertilionidae) in Beech Woodlands Of Central Italy: Consequences For Conservation. Biological Conservation 117(1): 73-81.

Russo D, Cistrone L, Jones G. 2007. Emergence time in forest bats: the influence of canopy closure. Acta Oecologica 31(1): 119-126.

Schulz M, Thomson B. 2007. National recovery plan for the bare-rumped sheathtail bat Saccolaimus saccolaimus nudicluniatus. Report of Department of the Environment and Water Resources, Canberra. Queensland Parks and Wildlife Service, Brisbane.

Sedgeley J A, O Donnell C F J. 1999. Roost selection by the long-tailed bat, Chalinolobus tuberculatus, in temperate New Zealand rainforest and its implications for the conservation of bats in managed forests. Biological Conservation 88 (2): 261-276.

Sedgeley J A, O Donnell C F J. 2004. Roost use by long-tailed bats in South Canterbury: examining predictions of roost-site selection in a highly fragmented landscape. Society Newsletter 11: 28-30.

Timm R M, Mortimer J. 1976. Selection Of roost sites by honduran white bats, Ectophylla alba (Chiroptera: Phyllostomatidae). Ecology 57: 385389.

29

Vonhof M J, Gwilliam J C. 2007. Intra- and interspecific patterns of day roost selection by three species of forest-dwelling bats in Southern British Columbia. Forest Ecology and Management, 252(1-3): 165-175.

Wei L, Han N, Zhang L, Helgen K M, Parsons S, Zhou S, Zhang S. 2008. Wing morphology, echolocation calls, diet and emergence time of black-bearded tomb bats (Taphozous melanopogon, Emballonuridae) from Southwest China. Acta Chiropterologica Vol.10 (1): 51-59. Yasuma S, Azniza M, Bernard H, 2005. The Bornean Mammals Vol. 2 & 3. Research and Education Component, BBEC Programme c/o Institute for Tropical Biology and Conservation, Kota Kinabalu.

30