I. Introduction to the proposed research
Loss of biodiversity is besetting ecosystems worldwide (Tilman et al. 1994; Jantz et al. 2015). Reduction of a single species can adversely affect the entire ecological chain, consisting of several other species throughout the ecosystem, thus leading to an overall reduction in biodiversity (Terborgh & Estes 2010; Brodie et al. 2014). These cascades result in loss of functional redundancies of an ecosystem that removes its buffer against unexpected catastrophic events, like storms, flooding, disease, etc. (Clavel et al. 2011; Säterberg et al. 2013). This eventually makes them prone to turbulence in the animal community and possibly collapses in later stage (Flynn et al. 2009; Reich et al. 2012). Driving this trend are anthropogenic factors such as unsustainable use, agricultural expansion, conversion of native habitat to settlements, increasing resource extraction, and introduction of invasive alien species (Didham et al. 2005; Wittemyer et al. 2008; Barlow et al. 2016). Retaining ecosystem services and the component species is essential to maintain them is in the best interest of humankind (Grifo & Rosenthal 1997). For our own health and prosperity maintenance of biodiversity and ecosystem function is essential (Chivian & Bernstein 2008; Sandifer et al. 2015). All species are not equally affected by anthropogenic activities, some being unable to adapt are ultimately extirpated, while others thrive. Functional traits that promote adaptation to human activity include wide habitat tolerance, nocturnal activity, small body size and general diet (Larson et al. 2015; Šálek et al. 2015). Specialists are especially vulnerable to extirpation when their essential resource becomes unavailable to them, either through removal or monopolization by humans (Pardini et al. 2010; Clavel et al. 2011; Burin et al. 2016). In order to maintain functional ecosystems, it is important to identify these specializations and to ensure they are addressed in conservation action plans and in the management of Protected Areas, PAs (Schwartz et al. 2000).
Illegal resource extraction, particularly of wood and medicinal plants, alters habitats (Olupot et al. 2009; Mackenzie et al. 2012). As human populations expand, especially near the borders of PAs (Wittemyer et al. 2008), habitats become increasingly fragmented, and PAs come under increasing pressure. Community structures continue to change as species with broader or more flexible niches out compete specialists (Šálek et al. 2015). Even highly adaptable species may experience extirpation through exploitative competition with humans (Henschel et al. 2011). This creates more vacant niches to be filled and could hasten further trophic cascade. To ensure management decisions fully represent their needs, we need to understand the optimal and minimum habitat requirements, spatiotemporal activity, dietary preferences, and niche plasticity of individual species, and how these interact with other members of their community to maintain a stable community structure.
II. Literature survey
Human population growth, the major cause of biodiversity loss (Tilman et al. 1994; Wittemyer et al. 2008) is driven by deforestation, habitat alteration, unsustainable use of resources and removal with respect to perceived conflict with human interests (Tilman 2001; Dirzo et al. 2014). Such changes in ecosystem can cause reconfiguration in species assemblage and further resulting in cascading ecological degradation, extinctions and niche expansions throughout the trophic web (Estes et al. 2011; Valiente-Banuet et al. 2015). Anthropogenic activities like deforestation, conversion of land to farmland, mining and human settlement result in habitat alteration (Laurance et al. 2014; Ravikumar et al. 2017). Abrupt ecotones are created between native forest and deforested landscapes, and species assemblages changes over a very short distance (Lacasella et al. 2015; Tölle, Engler & Panitz 2017). Within a community, high proportion of specialists might result in dramatic assemblage shifts across these ecotones (Clavel et al. 2011; Alves et al. 2017). Along with habitat changes, when ecotone also represents a shift from low to very high human activity levels, species thriving across the ecotone must avoid direct conflict with humans (Ditchkoff et al. 2006). Species that are nocturnal, have a competitive advantage over diurnal species as they naturally avoid human activity patterns (Ditchkoff et al. 2006; Zapata-Ríos & Branch 2018). Species with greater niche plasticity persists longer, disperse farther through converted landscapes, and tolerate smaller patches of native habitat (Gehring & Swihart 2003; Moreira-Arce et al. 2015). Hyper-generalists are the most successful in human dominated landscapes (Larson et al. 2015; Šálek et al. 2015). Populations of these synanthropic species can increase dramatically, and can aggravate extinction risk for less competitive guild members (Newsome et al. 2015). Both due to habitat destruction and human activity, edge effects are created, where impacts of disturbance on community membership and demography decrease gradually with increasing distance from the habitat boundary (Laurance 1991). Carnivores are particularly sensitive to human activity and habitat alteration (Di Minin et al. 2016). Retaliatory killing, decrease in prey base through direct hunting by people, and poisoning, conflict with domestic dogs, accelerates mortality in carnivores (Gabriel et al. 2012; Farris et al. 2015; Zapata-Ríos & Branch 2018). To minimise risk of lethal encounters, small carnivores may adopt a nocturnal lifestyle, shift activity patterns only in areas with higher human activity or reactively avoid locations where humans are active until their activity has subsided to tolerable levels (Moreira-Arce et al. 2015; Wang et al. 2015). The degree of adaptation, amount of acceptable human activity, and mechanisms employed to minimise human encounter risk, are expected to vary widely among species (Schuette et al. 2013; Caruso et al. 2016).
Due to increasing demands from the growing human population, associated livestock grazing pressure and land transformation for agriculture and development, forests get highly degraded and fragmented (Shamoon et al. 2018; Gupta 2011; Kalle et al. 2013; Kumara et al. 2014; Mukherjee et al. 2010). With time, fragmentation along with habitat loss have resulted in the increase in patchiness, increase in isolation distances and decrease in patch size (Kumar 1985; Puri et al. 1983). These factors exert different kinds of pressures on biodiversity at community levels (Fahrig 2003; Pardini et al. 2005; Magrach et al. 2014), whilst specifically impacting species with specialised habitat requirements or large home-ranges, resulting in modified and compromised ecosystem (Ries et al. 2004; Hector et al. 2001) with reduced ecosystem services (Allan et al. 2015). Fragmented forest patches are described as habitat islands in a sea of modified landscapes (Broadbent et al. 2008; Laurance et al. 2009; Gibson et al. 2013). MacArthur and Wilson’s (1967) Island Biogeography Theory is based on the principle that large, connected habitat patches support higher species diversity in comparison to small, isolated patches. Studies on fragmentation of tropical rainforest have shown that area of available habitat influences changes in animal occurrences and densities, with larger areas usually having a greater number of species (Laurance et al. 1997; Umapathy & Kumar 2000). However, small, isolated forest fragments are also repositories of biodiversity (Turner & Corlett 1996). Where habitat and niche availability are reduced, species richness may remain constant while functional diversity decreases, as specialist species get replaced by generalist species (Safi et al. 2011) thus habitat fragmentation drives functional trait loss, causing relaxation or extinction of species which possess pressured traits. So, it is critical to understand the influences of landscape-scale habitat fragmentation on both taxonomic and functional diversity to address ecological questions and conservation challenges (Mason and de Bello 2013). There are few studies on the impact of habitat fragmentation on population of Lion-tailed Macaque (Umapathy & Kumar, 2003), elephant (Kumar et al. 2004), leopard (Athreya 2014) and arboreal mammals (Umapathy & Kumar 2011) but none of them have looked at its impact on diverse mammalian community assemblages in detail.
Large predators perform crucial roles in ecosystems (Steneck 2005;Ripple et al. 2014). Through direct predation and lethal interference competition they influence the trophic web (Polis et al.1989; Miller et al. 2001; Zhang et al. 2015), create a landscape of fear among prey and subordinate predators (Lima 1998; Laundre et al. 2014; Ramesh et al. 2016). In order to avoid encounters with large predators, species at risk generally modify their activity patterns and habitat use (Berger & Gese 2007). This restriction of access to feeding area and prey items, may result in alteration of diet (Laundre et al. 2009; Laundre et al. 2014). Therefore, large predators keep population density of smaller predators and prey under control (Creel & Christianson 2008). The interference competition and predation promote biodiversity by limiting movements, abundance, and population densities of subordinate species and prey by mitigating overutilization of their food sources (Ritchie & Johnson 2009; Suraci et al. 2016). In many places, anthropogenic changes leads to the extirpation of large predator (Estes et al. 2011; Ripple et al. 2014). According to the Mesopredator Release Hypothesis, the extirpation of apex predators creates a trophic cascade that can have dramatic impacts on biodiversity throughout trophic food webs (Roemer et al. 2009; Terborgh & Estes 2010; Estes et al. 2011). As the landscape of fear is altered or removed, population densities and niche breadths of species in the trophic level below the extirpated large predator increases (Ritchie & Johnson 2009; Swanson et al. 2016; Droge et al. 2017). The released mesopredator may take over the former apex predator’s role, and may become the new apex predator (Berger et al., 2008; Ritchie & Johnson 2009; Wallach et al. 2015). Thus, alters the landscape of fear for smaller carnivore species as they face greater interference competition from the released species, this possibly decreases their densities and releases the next trophic level below them, thereby continuing the cascade (Levi & Wilmers 2012). If a new apex predator emerges that can sufficiently control other small predators, the system may stabilise and biodiversity may be retained at some levels (Gehrt et al. 2013).
III. Research gaps identified
Most of the studies focused on single species and a few studies provided information on general distribution of particular group of mammals (Kumara and Singh, 2007; Mudappa and Kumar 2007) but none of them have addressed the patterns shaping the diverse mammalian community assemblages in detail across land use gradients (PAs, Reserved Forests (RFs) and mosaic farmlands) in India. Since it’s one of the poverty prone districts, people are highly dependent on NTFP collection, and own large number of livestock for the sustenance/economic prosperity and even encroach the forested land for farming thereby exerting more pressure on the remnant forest patches. Thus, it is imperative to study how the leftover forest patches can still play a major role in conservation of certain mammal species that are vulnerable to habitat fragmentation. Therefore, it is important to investigate land use changes and effect of habitat fragmentation (patch size and isolation) driving mammalian assemblages. Fragmentation effects have been tested largely on birds (Ehlers Smith et al. 2018), this will be the first detailed study on mammals in India. With this habitat condition, our study also addresses how large predators can shape the behavioural patterns of prey and subordinate predators if diverse mammalian species need to be conserved. Such studies are urgently required in India as landscape transformation is taking a toll at an enormously alarming rate in the name of development.
IV. Objectives
1. To determine land use gradients shaping mammalian communities.
a) How does species abundance and richness vary across different land use types?
I expect that mosaic areas of farmlands would have lower mammal richness and occupancy compared with RFs/PAs where habitat is prone to higher human disturbance. I expect that sites with higher habitat heterogeneity would support higher mammal richness and high occupancy of forest associated species as natural vegetation diversity would decrease in farmland mosaics than in PAs/RFs and hence a concomitant decrease in mammal richness in farmland mosaics.
b) How does anthropogenic pressure drive temporal behaviour of mammals in a fragmented landscape?
Understanding how anthropogenic activity affects mammal behaviour is important as behavioral changes can affect species’ temporal patterns and influence habitat preferences. I expect that mammalian species sensitive to human activity would show largely nocturnal activity pattern in farmlands than PAs/RFs. I expect the increase in concentration of activity i.e. a shorter activity of window with increasing habitat disturbance.
2. To assess the impact of patch characteristics (size and isolation) on mammalian assemblage.
a) Is diversity of forest-associated mammal communities influenced by habitat fragmentation?
Here we expect that the structural complexity (heterogeneity) of habitat patches would decrease as patch size decreases, as niches will reduce along the reducing patch-size gradient. Decreasing patch size and structural complexity would have a significantly negative effect on richness and diversity. I expect a higher diversity of forest dependent species which have a specialized niche in larger patches, while there would be gradual loss of this specialized niche in smaller patches with increasing diversity of generalized species in smaller patches.
b) Are rare species more vulnerable to patch size and isolation?
A species will be categorized as rare based on its distribution pattern relative to other species across all patches. I hypothesize that the larger fragments will support higher detection probabilities and occupancies of rare species compared to the smaller patch sizes. Furthermore, the proximity of smaller patches to larger patches would contribute to increasing detections and occupancies of rare species.
3. To assess species interactions driving behavioural responses of mammals.
a) Can the absence of apex predator lead to meso-predator release?
As the presence of apex predator induces fear/ interference competition over mesopredators, this apparent response has important ecological implications in shaping carnivore communities. I will investigate pair wise interactions within the carnivore community. I would expect that the presence of apex predator like leopard would decrease detection and occupancy probabilities of mesopredators.
b) Does body size matter in behavioural patterns of prey and subordinate carnivores in response to top predator?
Body size as a life-history trait aids in species coexistence patterns through time partition like the adjustments of prey and subordinate predator behaviour in response to top predator. I would expect that the photographic time lag for e.g. between detections of leopard – jackal/hyena and leopard – wild pig/hare) would be longer at the same location to minimize direct interaction while the photographic time lag between leopard – jungle cat, leopard – hare, among subordinate prey species would be shorter at the same location. This interactive effect gradually reduces from large/similar body sized carnivores/prey to smaller body sized predator/prey.
V. Detailed methodology
Study Area
Bellary forest division is located in the central region of the eastern sector of the State of Karnataka and forms a part of the southern portion of the Deccan peninsula. At its extremes it is situated between 14° 30′ and 15° 50′ north latitude and 75° 40′ and 77° 11′ east longitude. The study will be conducted within 85 forest patches covering PAs (Daroji and Gudakote Wildlife Sanctuaries), RFs and adjoining fringe mosaic farmland areas up to 1 km from PAs and RFs. In total 1200 km2 area will be surveyed. Bellary has two different seasons: Long dry season (October – May) and short wet season (June – September). In December, the mean daily maximum temperature ranges from 16.7° C to 29.7°C. April and May happen to be the hottest months and the heat is oppressive. The mean daily maximum temperature is 39.2°C and the mean daily minimum temperature is 25.2°C. The general elevation of the district is between 500m to 700m above mean sea level. Except the hilly region in the west, the whole of the area in the east is open and plain with dreary and sand less expanses of black-cotton soil particularly in Siruguppa and Bellary taluks. The average annual rainfall is 574.9mm. The geographical area of the district and consequently of the division is 8447 sq. km which accounts for 6% of the State’s geographical area. The lands classified as ‘forest’ constitute 16.28% of the gross area of the district. Forest area cover eight ranges: Bellary, Daroji, Gudekote, Hadagali, Hospet, Kudligi, Sandur south and Sandur north. According to Champion and Seth (1968), the forests of the division has been broadly classified as tropical dry deciduous forests (predominantly composed of Anogeissus latifolia, Albizzia amara, Albizzia lebbek, Boswellia serrata, Chloroxylon swietenia, Dalbergia paniculata, Feronia elephantum, Hardwickia binata, Erythroxylon monogynum, Grewia tiliaefolia and Azadirachta indica) and tropical dry thorn forests (largely dominated by Albizzia amara, Acacia sundra, Cassia fistula, Carissa carandus, Cassia auriculata, Dodonea viscosa, Euphorbia spp, Randia dumetorum and Zizyphus spp.). Bellary district is rich in mineral wealth, especially in iron ore and manganese deposits. Most of the mines are situated in forest land. Over 5000 hectares of forest land is under mining activity at present. In mined areas the ecosystems are drastically altered and the resultant environment is not suitable for any productive use. Since it’s one of the poverty prone districts, people are highly dependent on NTFP collection, and hold large number of livestock for the sustenance/economic prosperity and encroaching the forested land for farming which is putting more pressure on the remaining forest patches. Therefore, the natural forest is being fragmented into patches. With huge human and livestock population pressure, changing land use pattern and encroachment of forest land to extensive agriculture on the forests lead to increasing human-animal conflict like livestock predation, human injury or death, wild animal mortality and crop damage particularly from leopard and sloth bear. The major mammalian species found within this habitat are leopard, striped hyena, Indian wolf, fox, golden jackal, jungle cat, rusty-spotted cat, Indian pangolin, small Indian civet, ruddy mongoose, Common mongoose, common palm civet, Porcupine, langur, bonnet macaque, wild pig, black-naped hare, blackbuck and four-horned antelope (Ranavath 2016).
Figure 1. Map of study area overlaid with 1km2 sampling grid covering Protected Areas, Reserved Forest and fringe Mosaic Farmland of Bellary District, Karnataka.
Data collection and analysis
1. To determine land use gradients shaping mammalian communities
The study will be conducted within 85 forest patches covering PAs (Daroji and Gudakote Wildlife Sanctuaries), non-PAs (RFs) and adjoining fringe mosaic farmland areas up to 1 km from the PAs and RFs as large number of mammals disperse and use adjoining mosaic farmland habitats because of water and availability food resources in the irrigated areas. As my study covers wide range of mammals, I would consider smaller grid size of 1 km2 as sampling unit which is assumed to capture distribution of all mammalian study species. Using ArcGIS 10.1 (Environmental Systems Research Institute Inc., Redlands, CA, USA), I will overlay 1 x 1 km grids over entire study areas. The land cover layer obtained from Bhuvan GIS (http://bhuvan.nrsc.gov.in/bhuvan_links.php) will be laid at the background to identify the gradient of land cover types in the study area. In total nearly 1200 grids of 1km2 will be surveyed using camera traps to assess the distribution and abundance of mammals across land use gradients that represent various levels of anthropogenic pressure. In each grid, trained forest staff and myself will walk at least 1 km to identify potential camera sites based on sign evidences (sightings, scats/pellet and tracks of mammals) and existing database/information from field forest staff for placement of single passive-infrared digital camera traps: Moultrie® M-50i, EBSCO Industries, USA (Jhala et al., 2008; Kalle 2013; Karanth and Nichols 2002; Ramesh et al. 2012). Each grid will have a single camera trap deployed. Two hundred camera traps will be operated at an inter-trap distance of ca. 500 m, simultaneously for 20 days in one section of the study area covering ca. 200 km2 and then moved to other parts of the study sites covering different land uses. This way we will cover many spatially representative sites across a gradient of land cover and habitat types for Bellary district. The GPS coordinates and the habitat variables including terrain, altitude, vegetation type, tree height, canopy cover, distance to water sources, distance to road and distance to settlements will also be estimated at each camera location. Whilst setting up cameras, we will record the microhabitat structure at each camera trap location within a 20 m radius around each camera trap site. A foliage profile for each site will then be compiled. To do so, a visual estimation of percentage coverage of each vegetation class relative to other classes will be made within the individual quarters of the circular plots (totalling 100% coverage in each). The vegetation classes considered in this study include bare ground, grass cover, leaf litter, herbaceous plant cover (including seedlings), canopy cover and woody vegetation cover (including saplings). Furthermore, the stem density of trees will be recorded at different height bands (2–5 m, 6–10 m, and 11–15 m), vegetation type, dominant tree and shrub species within 20 m radius (Jhala et al., 2017; Smith et al. 2017). Indications of human disturbances such as fire, number of cut/lopped trees, and cattle grazing/presence of cattle will be recorded within 20 m radius. In each 20 m radius plot, I will use a 5 m radius plot for quantifying dominant shrubs and weeds, % shrubs and weeds, and 1m radius for dominant herbs & grass, % of dry leaf litter, % of grass cover, % of herb cover and % of bare ground (Jhala et al., 2017). Plant species will be identified with the help of field guides (Keystone Foundation 2006; Balasubramanian 2015). I will collect the information regarding presence of mining sites, density of people, cattle, sheep, domestic dogs, and goats at block levels from local village Panchayat records of Bellari District office. Occupancy model (MacKenzie et al., 2006) will be applied to estimate the site occupancy of study species using the package unmarked (Fiske & Chandler 2011) in program R. I will then assess the community differences based on Bray–Curtis similarity index of species present between land use categories. I will estimate mammalian richness on the basis of presence–absence data using a first-order of nonparametric Jacknife Estimator in EstimateS 9.0 (Colwell, 2013). Based on the time stamp of each encounter, I will infer a 24-hour activity distribution function for each species at each land use category. Activity pattern for mammals in form of circular histograms will be analysed from independent time records using the program Oriana 4.0 (Kovach 2011). This will allow us to assess the 24-h activity of all mammal species across land use gradients.
2. To assess the impact of patch characteristics (size and isolation) on mammalian assemblage.
Field sampling will be carried out in 85 forest patches of varying sizes; the smallest is 2 km2 and the largest patch is 300 km2. Testing along a gradient of patch sizes aids the inclusion of mammals with varying home range and body sizes and their vulnerability to habitat fragmentation. For more detailed methods on field sampling please see the details in the 1st objective. The number of cameras placed in each patch will be in proportion to the patch size. I will firstly classify patches as mainland (intact, mostly protected large patches >100-300 sqkm) and fragment (patches <100 sqkm), and then calculate the area of each patch and it’s mainland isolation distance (i.e. the distance of the habitat patch from the nearest mainland patch). Patch characteristics include the following: patch area, distance to the nearest patch, distance to the mainland/largest patch, and the number of adjacent patches along distance categories. These patch characteristics will be measured using Patch Analyst in ArcGIS 10.1 (Environmental Systems Research Institute Inc., Redlands, CA, USA). The Island Biogeography Theory as described by MacArthur and Wilson’s (1967), I will include forest patch size, mainland isolation effects and patch interconnectivity, and their influence on mammalian richness and diversity. Thus, I will collate information on life-history traits such as all functional and biological traits present in the mammalian community at a given patch: I will develop a species-trait matrix incorporating the following traits: main diet (carnivory, frugivory, insectivory, omnivory), social structure and body mass in grams. All traits will be derived from the known literature (Menon 2016). I will determine the mammalian assemblage exhibiting a nested pattern i.e. whether nestedness patterns are related to size and degree of isolation to determine whether life-history traits (body mass, niche breadth, sociality, trophic level, arboreality & dispersal ability) influence the occurrence of nested patterns using BINMATNEST software which is a binary matrix ‘temperature’ calculator to determine the degree of nestedness of the assemblage matrix (Rodriguez-Girones & Santamaria, 2006, Atmar & Patterson, 1993; 1995; Rodriguez-Girones & Santamaria, 2006). To assess the intensity of habitat degradation due to anthropogenic disturbances, I will assess the 85 study patches using habitat structure, disturbance parameters such as presence of people/mines, livestock, trails, and tree-cutting signs, as described in the 1st objective along with patch characteristics. To assess the vulnerability of rare species to habitat fragmentation, I will relate the distribution and occupancy patterns of the rare species with patch sizes. Occupancy analysis (MacKenzie et al., 2006) will be applied to estimate the probability of site occupancy of mammals using the package unmarked (Fiske & Chandler 2011) in program R. To test relationships between patch characteristics, life history traits, disturbances and habitat structure on species richness, and diversity indices, I will identify the patch-size threshold that constraints the species richness and diversity patterns along a decreasing patch size gradient. This analysis will be conducted in program R v3.3.1 (R Core Team, 2015) using the following packages; “ade4” (Dray & Dufour, 2007), “arm” (Gelman et al., 2015), “ecodist” (Goslee & Urban, 2007) and Generalized Linear Models within the package “lme4” (Bates et al., 2015).
3. To assess species interactions driving behavioural responses of mammals.
As predation, interference competition, induction of apparent fear are important factors governing patterns in natural systems, and adjustments of prey and subordinate predators behaviour in response to top predator stimulus can have important ecological implications on their communities. I will investigate the effects of top predator (leopard) on the behaviour of different sized prey (Black-naped hare to wild pig) and subordinate predators (mongoose to hyena). The time lag distance between prey and predator photos in the same camera trap site will be used to address the heterogenous behavioural response of heterospecific prey species (different species; black-naped hare, four-horned antelope, black-buck, wild pig), and subordinate predators (hyena, jackal, fox, jungle cat, rusty spotted cat, small Indian civet, common palm civet, mongoose) to the presence/absence of leopard. I will get this information from time printed camera trap photos. So, the prey species largely eaten by top predator (leopard) substantially exhibit more time lag in response to top predator presence photos in the same site (Camera site) and it varies across different body sized prey species which possibly occurred due to increased vigilance or efforts to reduce the possibility of detection by potential top predators. This study will suggest an anti-predator strategy in different body sized prey species of top predators by maintaining a safe time interval in the same location. With subordinate predators, top predators often induce apparent fear/avoidance behaviour which can also vary across different body sized carnivores. In a community with species of similar foraging strategies and body size, interactions between them may lower their foraging success, and thus increasing the chance of interference encounters. Therefore, the photographic time interval (between photos of leopard and hyena) will be longer in the same location to avoid the direct encounter (intraguild interference interaction). This interactive effect gradually reduces from similar body sized carnivores to smaller sized carnivores where habitat types might play a major role. Single season two-species occupancy models will be applied to assess whether the presence of a species affect the occupancy and detection probability of a second species at camera sites (MacKenzie et al. 2004, 2006) in program PRESENCE 8.4 (Hines 2006). An independent photograph of a species regardless of multiple photographs recorded within 5 min at the same camera location to investigate the temporal overlap between pairs of species. The kernel density estimates of activity patterns of temporal overlap will be measured between pairs of species using the coefficient of overlapping (Ridout and Linkie 2009; Meredith and Ridout 2016). The temporal overlap analysis will be conducted in Program R Package version 3.0 (R Development Core Team 2014) using package overlap (Meredith and Ridout 2016).
VI. Expected outcomes
My study outcome will show how land use gradients drive the mammalian communities in Bellary district, Karnataka. It will highlight the importance of fine-scale habitat structure determining the habitat requirement of the mammalian species. Behavioural avoidance of mammals reflects their responses to human induced changes in the ecosystem. Decrease in patch size and habitat isolation will show how degradation of habitat reduces the habitat heterogeneity for diverse mammals. This study highlights the importance of integrating forest fragment attributes such as patch size and isolation affecting species with varied ecological traits that determine species in dire need of conservation action to prevent further loss of forest associated species. Two species interaction measured using occupancy model will enhance our understanding about the role of large predators on subordinate predators and prey if diverse mammalian communities are to be conserved and thus maintaining the ecosystem function. At least four research papers will be published in peer reviewed journals (national and international) and national magazines. This will encourage many future researchers to study elusive mammals. Presentations of all outcomes of this study will be made to different stake holders including government sources, and local organizations involved in conservation efforts and in various scientific platforms like seminars, conferences and workshops.
VII. Importance of the proposed research
The categorization of habitat quality from the spatial models can be used to prioritize areas requiring immediate protection for the species conservation. Maps with high, medium and low probability of occurrence of mammals generated from the study can be used to facilitate protection and restoration of critical habitat, and hence have broad applicability for conservation and wildlife management in the region. Predictive maps will be shared with forest managers, and stakeholders to pin point areas that require instantaneous species-specific site management and to generate awareness of the conservation threats to mammals in the region. As the study area has large numbers of mining sites, through this study we can identify their impact on the ecosystem that can later help the forest managers to regulate the activities of these mining sites in order to restore the forest. Habitat fragmentation attributes suggest the vulnerability of species to habitat degradation that will help in increasing the integrity of fragments and improving connectivity in the landscape for maintaining intact mammalian assemblage. The findings of this systematic study will give baseline information regarding the abundance and distribution of the mammals across land use gradients, which can be used by the forest managers for the management of biodiversity and expand the protection of mammals even in the adjoining mosaic landscapes of the whole district.
VIII. Research plan schedule
Research Activity Time in Months
6 12 18 24 30 36 42 48
1. Course Work
2. Literature Survey
3. Introduction
4. Objective 1
5. Objective 2
6. Objective 3
7. Half Yearly Reports
8. Communication of Papers
9. Preparation of Final Thesis
10. Submission of Thesis
IX. Pilot study/Preliminary work done
As part of the proposed study, I visited the study sites in Bellary district and interacted with all the concerned forest officers and gathered preliminary information required for this study. Information like maps and the spatial layers of GIS, and necessary logistics required for the study have been arranged. During this visit, I gained knowledge on various threats to the species and their habitat, and about how human induced land use changes around the fringe areas of PA and RF are affecting the mammalian community structures and their behavioural responses. This has helped me in developing many scientific questions for the proposed study.
X. Expenses and funding
My personal expenses will be covered under the Inspire Fellowship Scheme funded by Department of Science & Technology (DST), Govt. of India. While all the field expenses, field accommodation, equipment like Camera traps, GPS, compass, field travel and other logistics will be provided by Karnataka Forest Department, Bellary District. Office space, access to computer and library facilities, printing facility and other administrative requirements will be provided by my host institute: Sálim Ali Centre for Ornithology and Natural History, Coimbatore.
XI. References
...(download the rest of the essay above)