Temporal Foraging and Ranging Patterns Suggested the Niche Partitioning of Two Sympatric Herbivores, Axis axis and Bubalus bubalis , in the Nijhum Dweep National Park of Bangladesh

. For proper conservation measures and to elucidate coexistence mechanism of sympatric herbivore, we assessed the temporal foraging and ranging patterns of the Axis axis (spotted deer) and feral and/or semi domesticated Bubalus bubalis (bufalos) in the Nijhum Dweep National Park (NDNP) of Bangladesh. We have collected the data by day long scan sampling method for 12 months. We found that spotted deer and bufalos, respectively, spent 50.34% and 36.41% of their total day time in foraging. To avoid clash with the larger sized bufalos in the grazing ground, the spotted deer choose a slightly diferent time for grazing. At least three foraging peaks were found for spotted deer, whereas, bufalos showed two foraging peaks in a day. More importantly, spotted deer relied more on browsing for their food collection although they are natural gazer, whereas, bufalos relied more on their natural grazing habit for food collection. Spotted deer spent most of their time inside the forest and forest edges, whereas, bufalos mostly spent their time in the open grazing grounds. Te range of total distance moved (TDM) per day for spotted deer and bufalos was 1.56 to 2.67km and 1.02 to 3.30km, respectively. Te total area ranged (TAR) per day were 0.23km 2 to 0.8km 2 for spotted deer and 0.03km 2 to 0.35km 2 for bufalos. Although, these two parameters varied seasonally for both species, only in case of TAR of spotted deer the variation was statistically signifcant ( P < 0 . 05). We conclude that because of the presence of a larger sympatric herbivore, the spotted deer did some alterations in their temporal foraging and ranging pattern (TFRP) to survive in the small island which has very limited resources for their existence and survival.


Introduction
Since its introduction in the late 1980s, the spotted deer Axis axis (Erxleben, 1777) population in the Nijhum Dweep Island of Bangladesh, which is later ofcially declared as the Nijhum Dweep National Park (NDNP) in 2001 as per International Union for Conservation of Nature (IUCN) protected area category II [1]. Te number of the species has increased rapidly in the absence of any natural predators to encounter a population burst in the late 1990s [2]. Feeroz and Uddin [3] reported the estimated number of spotted deer to be 10,000 and 14,000, respectively, for 2001 and 2006 census in the island. Tereafter, the population size has been gradually decreasing in the island, which estimated about 2000 or less spotted deer in the island [3]. One of the reasons for this decline may be due to the increasing human population. Te consequence of human activities resulted in habitat fragmentation and subsequence destruction, and also other anthropogenic disturbances. It is now almost universally accepted that habitat loss and over exploitation are putting many of the world's mammal species at risk of extinction [4]. Hunting, habitat degradation, and invasive species are some of the main leading factors which are responsible for the poor status of many mammal species [4].
Local people in the NDNP nurture a lot of bufalos Bubalus bubalis (Linnaeus, 1758) in a semidomesticated manner, which means all bufalos have an owner but always remain in the forest under the supervision of cowboys, and some of these sometimes become feral in the forest. Since both spotted deer and bufalos are herbivores having a more or less similar food preference [5], they compete for forage resources. Te high diet similarity between the sympatric species indicates competitive interaction at high density and with limited food resources [5]. Although smaller in size, the spotted deer still remain and coexist with the bufalos in the island. However, sympatric species difer in their feeding styles [6] due to diferences in morphological [7] and physiological [8] characteristics among them. Hence, the spotted deer might have done some modifcation in their activity patterns to cope with the competitors.
Information on the temporal activity and ranging patterns of animals is crucial for the implementation of suitable conservation measures, as it refects the responses of subject animal to anthropogenic disturbance [9]. Moreover, the coexistence of animals ecologically rely on the temporal activity pattern information [10]. Among the closely related sympatric species, the temporal activity patterns are critically important to understand their coexistence mechanism in relation to interspecifc competition and/or niche separation [11]. Home range extents of an animal depend largely on resource requirements [12]. Animals not only need sufcient space to procure food and other resources, but also have to stay away from predators [13] and other larger competitors. Te present study is aimed to perform the comparative analysis of foraging activity (FA) and ranging patterns (RP) of spotted deer and bufalos in the NDNP of Bangladesh. Te objective of this study was to understand how the two sympatric species of herbivore, having very close and/or similar feeding habits, partition their ecological niche to coexist in a small island with limited resources.

Study Area.
Te Nijhum Dweep National Park (NDNP) comprises of a cluster of several small islands, located about 31 km southwest of Hatiya Upazila under Noakhali District, Bangladesh. Te geographical extent of the NDNP is in between latitude 22°01′25″ to 22°06′11″N and longitude 90°56′44″ to 91°06′07″E. Te NDNP is located at the confuence of the Meghna River mouth on the Bay of Bengal ( Figure 1).
Te main island is about 10 km long in north-south and 8 km wide in east-west directions [3]. Te NDNP is separated from the main land Hatiya by Mokteria channel and bounded by the Meghna River in the eastern part, Mokteria channel in the northern part, the Shahabaj River in the eastern part, and the Bay of Bengal in the southern part. Tere are three major types of land use available in the NDNP, such as, forested area, grazing land, and human habitation.  [14] was used to record the activity budget, foraging behaviour, and habitat use of the two species. Although many deer species are crepuscular in nature [15,16], spotted deer are inclined to be diurnal [17,18] and bufalos are completely diurnal animal, hence the data collection was carried out during the day.

Methods of Data
Before starting the data collection, one herd of spotted deer was selected which forage mostly on the remote side of the island where least anthropogenic activities were evident. After that, the herd was habituated to continuous presence of the observer(s) in the vicinity wearing local dress as they appeared to recognise external clothing and persons. A scan interval of 2 minutes was chosen. Te feld observations were carried out with naked eyes and with the aid of one pair of binoculars (10 × 4 binoculars (Model: Zen-Ray 2015 ZRS HD (Summit) 8 × 42) depending upon the prevailing feld conditions, from a distance of 100 to 150 m, so the focal herd did not get disturbed. Te behavioural states were as follows: Moving (when an animal is in motion in the form of walking or running without doing any other activities); Resting (when an animal take rest either standing or laying without ruminating); Ruminating (when an animal ruminate either standing or laying); Foraging (when an animal move either for grazing or browsing); Social Behaviour (it included fghting and snifng for spotted deer, and fghting, snifng, and bathing for bufalos); Alarm (when an animal give alarm to its herd members. Here, standing alert and call alert for spotted deer, and standing alert alone for bufalos. Te sampling was carried out once in a month for each species continuously from dawn (06.00 h) to dusk (18.00 h). Five individuals from each herd of the two species were selected each time-of diferent age and sex-for recording diferent activities. As in every two minutes a single activity record was taken, so for fve individuals, 150 activities were recorded in every hour, and in this way a total of 1,800 activities were recorded, in 12 hours (06.00 h to 18.00 h), in a single day of the representative month for each of the two species.
Data recordings were started from March 2018 as this month represents the frst month of summer in Bangladesh. In the present study, three broad seasons were considered-summer (March to June), monsoon (July to October), and winter (November to February). Te focal spotted deer and bufalo herds were followed on foot from dawn to dusk in the feld and recorded their activities. Te samples for each activity were averaged on a daily basis and their standard deviations were calculated. Te detailed activities were merged into broader activities for monthly, seasonal, and yearly analysis. For analysis of data, IBM SPSS statistics 20 programme was used.

Ranging Pattern.
Two parameters were considered for the analysis of ranging pattern of the two species, viz., total distance moved (TDM) per day and total area ranged (TAR) per day. For determining these parameters for both the species, 7 to 9 GPS (using a hand held GPS, model: GARMIN GPSmap 60 CSx) readings were taken, which represented the major activities (waking, resting, grazing and browsing, and bathing in case of bufalos) performed in the whole day (06.00 h to 18.00 h). By connecting the GPS points in ArcGIS 10.8 programme, the TDM for each species in each month was calculated. For determination of TAR, a polygon was constructed by connecting the GPS points in the same software and resulting areas of the polygon was considered as the ranging area of the respective species in the respective months. Te seasonal data of both species have been presented in real earth image of "Google Earth," and also the TAR of both species were overlaid on seasonal basis (summer, monsoon, and winter) in the same image to compare the ranging areas and activity partitioning.

Comparative Activity Budget of Spotted Deer and Bufalos.
During the study periods, both the species were found to spent more time, 50.34% and 36.41%, respectively, for spotted deer (Axis axis) and bufalos (Bubalus bubalis) for foraging than those of the other activities (Table 1). Within the foraging activity, spotted deer relied more on browsing activity (32.91%), whereas bufalos relied more on grazing (28.19%) ( Table 1).

Temporal Variations in Foraging.
According to GPS tract records, both species overlapped with each other frequently during foraging, other than this activity they remained separated; that is why, day long comparison of foraging of both species was made.
In all seasons, spotted deer started their foraging long before dawn (06.00 h) and continued up to 10.30 h (Figures 2-4). Mostly before and after sunrise they respectively grazed and browsed. From 11.00 h to 14.00 h, they did not forage at all, rather they rested at that time; after 14.00 h, they again started foraging and became more and more active towards the dusk (18.00 h) and to continue it after sunset (Figures 2-4). On the other hand, bufalos started the day with resting in all seasons; they did not start foraging as late as 07.00 h-seasonally this time varied-became fully active in foraging at about 08.00 h (Figures 2-4). Teir foraging was predominated by grazing, especially in the morning, but some browsing activity was also noticed before noon. From 13.00 h to 15.00 h, they stopped foraging and went for resting/ruminating (overall resting) (Figures 2-4). After 15.00 h, they again started foraging and continued up to 17.00 h, after that they reduced foraging and started returning to the resting place (Figures 2-4).
Spotted deer showed three foraging peaks, whereas bufalos showed two in all seasons (Figures 2-4). Te frst and last peaks of spotted deer were difused as they started foraging before sunrise and in the evening they continued foraging even after dusk. During winter, the peak of spotted deer foraging was found to be sharper than that of the other two seasons (Figure 4), whereas in case of bufalos, foraging peak of winter was found to be broader than that of the other two seasons (Figure 4). Te percentage of time spent in foraging was also more in winter than that of the other two seasons for both spotted deer and bufalos.
During summer, spotted deer and bufalos overlapped in foraging between 08.30 h and 10.00 h (i.e., before noon) and 15.30 h and 16.30 h (i.e., late afternoon) (Figure 2). Overlapping of resting in summer was not that evident as spotted deer started resting at 11.00 h and stopped at 13.00 h, whereas bufalos started resting at 13.00 h and ended at 14.00 h (Figure 2).
During monsoon, spotted deer and bufalos overlapped in foraging between 08.30 and 11.00 h and 15.30 h and 16.30 h (Figure 3). Overlapping of resting in monsoon was not that evident as spotted deer started resting at 12.00 h and fnished at 14.00 h, whereas bufalos started resting at 14.00 h and fnished at 15.00 h (Figure 3).
During winter, spotted deer and bufalos overlapped in foraging much less times-10.00 h in the morning and 15.30 h to 16.00 h in the late afternoon ( Figure 4). Overlapping of resting in winter was not that evident as spotted deer started resting at 12.00 h and stopped at 14.00 h, whereas bufalo started resting at 14.00 h and ended at 14.30 h (Figure 4).

Habitat Preference for Diferent Activities by Spotted Deer and Bufalos.
Te GPS data of ranging pattern revealed that spotted deer grazed mostly in the early morning and late evening in the grass land of the forest edges and small grassy patches within the forest; whereas bufalos grazed mostly in the open grassland in the late morning and during evening (Figures 5-10). Te foraging pattern of the spotted deer was dominated by browsing within the forest, whereas in case of bufalos that was dominated by grazing in the open grassland ( Figures 5-10). Te resting pattern of both species also showed dissimilarities, such as the spotted deer rested entirely within the forest, whereas bufalos rested in the open feld or at the edge of the forest (Figures 5-10). When the seasonal TAR of both species overlaid on each other, it revealed that both species share some common areas but there were diferences in time of the activity within the shared areas during all seasons (Figures 5-10). Te TAR of both of the species moved from north to south with the advent of monsoon, whereas again they shifted from south to north during winter and summer (Figures 6, 8, and 10).

Discussion
In the present study, the foraging behaviour accounted for 50.34% and 36.41%, respectively, for Axis axis (spotted deer) and Bubalus bubalis (bufalo) in the Nijhum Dweep National Park (NDNP) ( Table 1). Foraging included both grazing and browsing. However, grazing included all feeding on grass and other low height vegetation, whereas browsing included feeding on leaves, soft shoots, or fruits/pods of highgrowing, generally woody plants [19]. Relatively high amount of time spent in foraging by both species indicated that there is shortage of food or competition for food for both species in the island. Dave [5] also found that spotted deer and bufalos spent more than 30% time in foraging in the Gir National Park, which has similar results of this study. According to previous theoretical [20,21] and empirical [22] studies, greater time spent in foraging always indicated poor habitat resources, especially food resource. Te present study indicated poor food resources for both the sympatric species, especially very poor food resources for spotted deer. Although spotted deer is smaller in body size than that of the bufalos, it had to forage more to fll up its relatively smaller stomach further indicated very poor food resources available for spotted deer in this national park.
Spotted deer grazed very limited amount of time than that of the bufalos (Table 1). Generally, spotted deer were found to be grazing in the forest edges and small grassy patches of the forest, whereas bufalos grazed mostly in the open feld. Tis might be due to anthropogenic disturbances (agricultural works, sound of engine boats, and so on) that prohibited spotted deer to graze in the open feld during the day; as earlier research [5] mentioned that spotted deer spent more time in grazing than browsing in the Gir National Park of India. On the other hand, spotted deer were found to be browsing high amount of time than that of the bufalos, as they were found preferring feeding on keora (Sonneratia apetata) leaves and fruits from the forest foor, which were easy options for them.
Te foraging patterns of spotted deer and bufalos showed clear diferences as the former showed three distinct peaks of foraging, whereas the later showed two (Figures 2-4). Tis result disagreed with the observations of Dave [5], who found two peaks for both spotted deer and bufalos in the Gir National Park of India. Te second foraging peak of the spotted deer was formed during 08.00 h to 10.00 h, and it is exceptional for most of grazing ungulates might be formed due to walking of the species in search of available food in the forest as they could not avail the food from the grassland at that time because of anthropogenic  International Journal of Zoology 5 activities and the presence of bufalos. Broad patterns of alternating foraging and resting for both species were little bit diferent. Little diference in the seasonal patterns of foraging of individual species was also evident. Moreover, the morning and evening peaks of spotted deer were difused and refers that the morning peak started much before the dawn and the evening peak ended after dusk (Figures 2-4). Distinct foraging peaks in the daily cycle were also found in most of the studies [23][24][25][26] with an intensive peak in the morning hours.
Te spotted deer found to be grazing less during the day than that of the bufalos, when the later occupied the grazing   land. Fisler [27] mentioned the infuence of larger body size in determining the dominance among the species of the same trophic level. In this case, interspecifc behavioural hierarchy of the two species probably determined the partitioning of the ecological niche [28], as spotted deer, although predominantly a grazer [5], had become a predominant day time browser in the NDNP. In areas where larger species of the same trophic levels got established, the larger one either monopolizes the resources directly from the smaller one or caused smaller species to shift niches [5]. Te diference in the foraging patterns could also be attributed to the size, quality, and quantity of the food items; gut capacity and process; and digestion of the forage [29][30][31].
Both abiotic factors (such as slope and distance to water) and biotic factors (such as forage quality and quantity) might govern the foraging or ranging pattern of large herbivores   Figure 8: TAR (shaded spotted deer and solid bufalo) of a single group of spotted deer (orange) and bufalo (pink) in the Nijhum Dweep National Park in monsoon-2018. [32]. In this study, the major abiotic factors that infuenced the ranging pattern of both the species might be the tidal water and anthropogenic disturbances; major biotic factors might be forage quality and quantity, and the species themselves as resource competitors. Te average TDM and TAR of spotted deer were higher than that of the bufalos, indicating that the resources available for spotted deer were limited in the island compared to the bufalos. Seasonally, spotted deer moved more during monsoon, whereas bufalos moved more in summer. Both spotted dear and livestock (cattle and bufalo) move more in summer than winter and monsoon in the Gir National Park, India. Tis is because during summer, increased walking appeared to be the expense of foraging time, although other activities did not change much [5]. But in the NDNP, due to regular tidal periods, the grazing land never get completely dried; hence, no notable diferences in grass availability were evident. On the other hand, during monsoon, the grazing and forest lands got submerged more intensely, so food became very limited for spotted deer as they could not tolerate high water levels, hence they moved towards the high dry land. In contrast, bufalos are more comfortable in the water by nature, so they do not face any problem in fnding foods during monsoon, and as a result they moved less in this season; Moreover, their TAR is also found less during this period of the year. During summer period both TDM and TAR of bufalos were found to be more than that of the other two seasons. Tis might be due to searching of suitable bathing place, as during this period the island become very hot and dry, so the available water pools in the open grassland for bathing become limited. Te abovementioned fndings supported the general concept of the negative relationship between the foraging area and resource availability for herbivores as mentioned by Cohen et al. [33,34] in case of white-tailed deer and Dave [5] in case of spotted deer and bufalos. But seasonal data of TDM and TAR deferred with that of Dave [5], who found greater foraging area during summer for the spotted deer in the Gir National Park of India; this might be due to the diference of the geographical position of the two study areas. Animals ranging in an area with better food availability have a lower net displacement and decrease the chance of leaving the high resource density area, thereby increasing the utilization of resources [35][36][37][38].
As information on resource distribution is considered to be limited in animals [39], so a number of past studies used random movement strategy while foraging within the theory of optimal foraging [38,40]. On the other hand, Bailey et al. [32] observed that prior information on the resource distribution increases their chances of encountering the target, hence form a defnite shape of animal foraging path. In this study, the ranging pattern and foraging path of both the species indicate that they probably have a clear idea about the resource distribution of the island as they are more or less specifc in terms of timing and use of resources.

Conclusion
An animal's foraging and ranging habits may help to solve the issue of adequate management and conservation measures, which are important considerations for park managers. Tis study can be utilized as a starting point, and longterm research may reveal more information about these two sympatric ungulates.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.