Butterfly Abundance and Diversity in Different Habitat Types in the Usangu Area, Ruaha National Park

Insects are key fauna species that respond quickly to disturbances and environmental changes. Tey act as good indicators of habitat, community, or ecosystem quality. Among the great diversity of insects, butterfies stand as ideal bio-indicators for ecosystem function and are sensitive to changes in habitat composition and structure. Tis study was carried out to examine the diversity and abundance of butterfies across the restored habitats in Usangu area part of Ruaha National Park (RUNAPA) from May 2022 to June 2022 using the walking transect method supplemented by sweep nets and butterfy baited traps. A total of six transects of 1km in length were laid in the four main habitat types selected in Usangu area including grassland, Miombo woodland, Vachellia/Commiphora woodland, and riverine forest. Searches were conducted in the morning and evening. Te Miombo woodland and riverine forest habitats exhibited relatively higher species diversity, richness, evenness, abundance, and a higher number of habitat-restricted species, while Vachellia/Commiphora woodland and grassland habitats recorded the lowest diversity and abundance as well as the lowest number of habitat-restricted species. Family Nymphalidae was the most dominant followed by Pieridae while Papilionidae and Hesperiidae were scarce in the study area. Tis study clearly shows the importance of Miombo woodland and riverine forest habitats in sustaining rich butterfy diversity and abundance in Usangu area. Te two habitats must be efectively managed and conserved for sustaining ecological health and integrity of Usangu area. Te Usangu area’s Miombo woodland and riverine forest habitats have immense potential for butterfy tourism and they can ofer an excellent opportunity to promote conservation eforts and raise public awareness. However, it is crucial to monitor these habitats closely as any environmental changes that may occur could harm the butterfy diversity and abundance in the area. Preserving this wilderness to maintain a thriving butterfy population is of utmost importance.


Introduction
Conservation and sustainable management of biodiversity in the face of ongoing global climate change require updated information to evaluate species' geographic distributions and changes over time [1,2].With an increase in human disturbances globally, numerous species are being adversely afected, hence accelerating biological diversity loss and extinction [3,4].Te loss and extinction of biodiversity is perhaps the most important concern in many African savannah regions, which are greatly rich in biodiversity but experiencing rapid habitat fragmentation and degradation [5,6].Terefore, there is an urgent need to understand how species respond to changing environmental conditions in African savannah ecosystems over time to improve the management of species or their habitats [7].
Invertebrates such as insects are key fauna species that respond quickly to disturbances and environmental changes and therefore act as good indicators of habitat, community, or ecosystem quality [8,9].Tis is because arthropods comprise the most tremendous taxonomic and functional diversity group of multicellular organisms on the planet, and they contribute signifcantly to vital roles in ecosystems such as pollinators, pest controllers, decomposers, scavengers, and prey species [10,11].Among the great diversity of arthropods, butterfies stand as ideal bio-indicators for ecosystem function and habitat conditions [12].Tey are the second largest, the most widespread, and most widely recognizable arthropods in the phylum Arthropoda, making up the insect order Lepidoptera.Tey have been widely promoted as potential land use change, climate, and habitat quality bio-indicators [13,14].Butterfies have often been used as indicator species for refecting environmental changes as well as habitat conditions or development after natural and anthropogenic disturbances, especially in highly diverse tropical regions of sub-Saharan Africa, where conditions are too difcult or expensive to measure directly [15,16].
Despite the ecological importance of butterfies as bioindicator species, no study has been conducted to document the status, distribution, and diversity of butterfies in Usangu area, in Ruaha National Park, Tanzania.Te Usangu area has been subjected to and thus experienced large-scale habitat fragmentation and degradation caused by both natural forces and anthropogenic activities [17][18][19].However, the Usangu area was annexed into Ruaha National Park in 2008 to protect the important wetlands and biodiversity.Various scholars have documented the importance of Usangu area and its restored habitats in protecting mammals of the area [20,21], but very little attention has been given to the understanding of insects such as butterfies, despite their signifcance as excellent indicator species for monitoring environmental conditions or assessing the efcacy of management.Terefore, the current study has been undertaken to address this knowledge gap by assessing the butterfy species' abundance, richness, and diversity in relation to restored habitats in the poorly studied Usangu area, part of Ruaha National Park.Because butterfies are sensitive to environmental perturbations, an efort has been made to understand the causes of changes in butterfy biodiversity with respect to their habitat type and assess the suitability of using these arthropods as indicators of the health of the environment in Usangu area.Te objectives of this study were to provide baseline information on the status and spatial distribution of butterfy species and to determine if there is a signifcant diference between butterfy species diversity, richness, and abundance in the four main habitat types.Tis study also intends to examine how the butterfy community structures vary in diferent surveyed habitat types in Usangu area.Findings from this study would be very important as it would serve as the frst approach of using the butterfy as part of bio-indicator tools for subsequent assessment and monitoring of restored habitat condition and quality within the Usangu area and its surroundings.

Study Area.
Te study was carried out in Usangu area part of Ruaha National Park (RUNAPA) in Southern Tanzania (Figure 1).Usangu is located in latitude 08 °30′ south and longitude 34 °15′ east with an elevation of 1,026 meters above mean sea level (amsl) [17].Te Usangu area is made up of wetlands including the Usangu wetland and upper drier alluvial fan ecosystems [19].Te mean annual temperature of Usangu is 24 °C, and the average annual rainfall is 650 mm.Usangu is characterized by a range of habitat types including the montane forest, Miombo woodlands, riverine forest, and lowland savannah with Vachellia and mixed woodland [22][23][24].Te Usangu area was previously a game reserve that was fragmented and degraded by uncontrolled human activities [25].However, in 2008, the Usangu area together with other nearby important and remarkable wetlands was annexed into Ruaha National Park (RUNAPA), making it the second largest National Park in Tanzania and East Africa with an area of about 20,226 km 2 [26].All sampling sites for this study were located in the Usangu area within Ruaha National Park.

Study Design and Data
Collection.Ground surveys were carried out along the main administrative and patrol roads between May and June 2022.Walking transects of 1 km long were established perpendicular to roads to complement the efects of roads on butterfy species distribution.Tese transects were established in the four main habitat types selected in Usangu area including grassland, Miombo woodland, Vachellia/Commiphora woodland, and riverine forest (Figure 1).At least, 6 transects of 1 km each were established in every habitat making a total of 19 transects.Te research team walked a straight line of 1 kilometer perpendicular to the park administrative roads (left or right side) using a handheld GPS to maintain the straight line depending on the terrain and vegetation cover of the area.Each transect was sampled using three complementary methods.Firstly, the transect method was used by visual watching of fying butterfies along transects by the team of specialized identifcation skills [27][28][29].Te visual observation method was conducted in the same transect before a handheld butterfy net and the baited trap were used.Te method was used to record butterfy species that are common and easy to identify to avoid overcollection.Te method involved counting the number of fying butterfies that crossed a strip of known length (between 30 m and 60 m) and 20 m wide for 10 minutes, and time lapsed was recorded by a handheld stopwatch [28].
Secondly, the collection of butterfies using a handheld sweep net (35 cm in diameter) following [29] was commenced after visual watching of fying butterfies.Searches in each transect were conducted by two to three collectors for 5 hours for each trap day from 09:00 to 12:00 in the morning and from 15:00 to 17:00 in the evening.Timed sweep netting was conducted within the transect established.Data on species type and associated habitat types in its specifc transects were recorded on standardized data sheets.Unidentifed individuals were kept in special envelopes for later identifcation at the species level with the aid of feld guidebooks [30][31][32].
Tirdly, butterfies were sampled using cylindrical butterfy traps baited with fermented banana along transects (measuring 1000 m), and baited traps were used to capture fruit-feeding butterfies that may not be captured in 2 Psyche: A Journal of Entomology handheld sweep net [33].Traps were placed at 30 m intervals from each other making a total of 16 traps per habitat selected; three habitat types were selected except grassland because of the absence of trees used to hang traps.A total of sixteen traps were placed in each habitat, summing up to a total of 48 traps in the study area.Te traps were baited with yellow ripe bananas that were smashed, uniformly mixed, and allowed to ferment in a plastic bucket for six days.Tree teaspoonfuls of bait were placed on a small plastic plate ftted on a cotton base.Traps were baited on the frst day of setting and hanged at 50-100 cm height above the ground [34,35].Te baited traps were checked once a day for three consequent days between 09:00 and 17:00.Caught butterfy individuals were identifed using standard identifcation feld guides [30][31][32], and numbers of each species were recorded.

Butterfy Species Composition.
In this study, butterfy data from walking transect, sweep-netting, and baited traps were pooled to obtain total butterfy diversity and abundance per habitat type.Several community parameters were used to compare butterfy species diversity.Te Shannon-Wiener index (H′), species richness (S), relative abundance (A), and evenness (E) parameters were computed for each surveyed habitat type using the PAST 4.04 statistical software [36].Te butterfies were classifed into fve categories based on the number of sightings in the study area as follows: VC, very common (>100 sightings), C, common (51-100 sightings), NR, not rare (16-50 sightings), R, rare (3-15 sightings), and VR, very rare (1-2 sightings) [37].
Butterfy species restricted to only one habitat type were categorized as habitat-restricted species.Butterfy richness in the study area was estimated using the Chao 1 estimator.Chao 1 estimator is suitable for nonparametric data containing single and doubletons and uses abundance data [38].Species richness was computed for each habitat type in the program EstimateS v.9.0 [39], using 100 randomizations.Te species accumulation curves were generated for each habitat type by using EstimateS v.9.0 [39], to evaluate the completeness of butterfy sampling.Species accumulation curves represent the observed butterfy species number (Sobs) and the estimated butterfy species number which was collected as the number of samples approaches the population size.Te diferences in butterfy species richness and abundance types were tested with Kruskal-Wallis tests after confrming nonnormal distribution [40].Species evenness and diversity across habitats were compared using a one-way analysis of variance (ANOVA) [41].

Comparison of Butterfy Assemblages among Habitat
Types.Te similarity matrix using the Bray-Curtis index was generated to assess how butterfy communities organized across the surveyed habitat types in Usangu area, using the square root of log transformed data, pooled within sampled days.Furthermore, the metaMDS function of the vegan R package was used to perform nonmetric multidimensional scaling (NMDS) to understand the ordination of butterfy communities between habitat types [42].When clustering was evident, a one-way analysis of similarities (ANOSIM) was used to analyze the signifcant diferences in butterfy assemblage across the habitat types (α � 0.05) [43].Psyche: A Journal of Entomology All statistical analyses were undertaken using R software (R Development Core Team, 2022), with the Community Ecology Package "vegan" [42].2).Belenois aurota aurota was the most abundant species in all four habitat types with 742 individuals (37.19% of total individuals).Pinacopteryx eriphia melanarge was the second most abundant species, with 125 individuals (6.27%).On the other hand, a total of 40 habitat-restricted species were recorded in the study area.Among these, 55% (22 species) were restricted to Miombo woodland habitats, 22.5% (9 species) to riverine forest habitats, 17.5% (7 species) to grassland habitats, and 5% (2 species) to Vachellia/Commiphora woodland habitat (Tables 2 and 3).Te variation in the number of species belonging to each family was also observed in the study area.Nymphalidae had the highest representative species, comprising 46 species (47.92%), followed by Pieridae 30 species (31.25%) and Lycaenidae 16 species (16.67%) while Hesperiidae and Papilionidae families were represented by two species each (2.08%) (Figure 2(a)).Family-wise composition of butterfies in terms of species richness recorded in each habitat type in Usangu area is presented in Table 3.Only two out of fve butterfy species families were recorded in all the habitat types in the study area.Both Nymphalidae and Pieridae families had the highest species richness recorded in Miombo woodland (Table 3).Lycaenidae was the codominant family with 7, 9, and 7 individual species recorded in grassland, Miombo woodland, and riverine forest habitats, respectively (Table 3).Hesperiidae had the highest number of species recorded in Miombo woodland while Papilionidae family was represented by only two species that were recorded in grassland and riverine forest (Table 3).

Patterns of Butterfy Species Richness, Diversity, and
Abundance in Four Habitat Types in Usangu Area.Te highest species richness was recorded in Miombo woodland followed by riverine forest and grassland habitats while Vachellia/Commiphora woodland habitat had the lowest species richness (Table 4 and Figures 3(a) and 3(b)).Species richness was not signifcantly diferent among the habitat types (χ 2 = 5.15, df = 3, P � 0.16).Te comparison of the estimated species richness (Chao 1) is shown in Table 4 and Figure 3(b).Te estimation of species richness in the four habitat types in Usangu area by Chao 1 shows that species richness was not close to the observed values in all habitat types except for Vachellia/Commiphora woodland (Figure 3(b) and Table 4).Te highest variation between the observed and expected richness was recorded in the three habitat types in Miombo woodland, grassland, and riverine forest, and this was evident from the low values of their sampling completeness as indicated in Table 4. Te rarefaction curves for the three habitat types also failed to reach an asymptote when plotted against the cumulative number of individuals (Figure 4).Te rarefaction curves for Miombo woodland, grassland, and riverine forest continued to rise even when all individual butterfies were accumulated (Figure 4).However, the rarefaction curve for Vachellia/ Commiphora woodland seems to approach asymptote gently, showing that species saturation had been reached and sampling efort was adequate (Figure 4).

Butterfy Community Structure in the Four Habitat Types in Usangu
Area.Te results of butterfy community similarity based on the Bray-Curtis index show that the butterfy species composition in Usangu area was structured into two main clusters (Figure 5).Te similarity of butterfy communities showed that butterfy species between Miombo woodland and riverine as well as Vachellia/Commiphora woodland and grassland habitats resembled each other, while those in grassland and Miombo woodland were greatly distinct (Table 5 and Figure 5).Te scatter plots based on NMDS ordination plot (Figure 6) also showed a separate clustering of points for Miombo woodland and riverine suggesting similar distribution patterns.Vachellia/Commiphora woodland and grassland habitats clustered together, while the butterfy species in Miombo woodland were clearly separated from the grassland habitat.Te analysis of similarity (ANOSIM) test showed a signifcant diference in the butterfy communities across the four

Discussion
Tis study presents the diversity, richness, and composition of butterfy species community into four main habitat types in the Usangu area, mainly the grassland, Miombo woodland, Vachellia/Commiphora woodland, and riverine forest.
Since there had been no study of butterfy composition in Usangu area, the present study forms the frst checklist records on butterfy species of Usangu area, in Ruaha National Park.Te generated checklist serves as baseline information for future studies, butterfy conservation action plans, and attraction for tourists as well as a monitoring tool for assessing the impacts of environmental changes such as global climatic change in Usangu area.Te occurrence of fve families in Usangu accounts for 71.43% of the total butterfy families that are known to occur in Tanzania [29,30].Te occurrence of 10 species (10.42%) in all four habitat types of Usangu area indicated that the species recorded are habitat generalists and therefore they are described as ubiquitous [44,45].Te general occurrence of generalist butterfy species would beneft them by providing a greater distribution as well as maintaining a bigger population size [45].
On the other hand, 40 species (41.67%) were restricted to single habitats only (Tables 2 and 3).Te presence of these species in a single habitat indicates that these species might be habitat specialists and are sensitive to changes in plant community diversity, composition, and structure.Many butterfy species prefer only a particular set of microenvironments that is found in each habitat [46,47].In the present study, we reported 96 butterfy species occurring in the Usangu area, and these results can be compared with the study by Nkwabi et al. [29] who reported a total of 96 species of butterfy species from Ntakata Forest in Western Tanzania.However, the number of butterfy species and genera found in this study was lower compared to the number of butterfy species found elsewhere in Southern Tanzania.For example, Jew et al. [15] found 104 species of butterfy  [48] have reported a smaller number of butterfies (90 species) compared to the current study.Te diferences in habitats, seasons, study period, size of the study area, and sampling design could be the plausible reason for the variation in the number of butterfy species and genera recorded in these areas.For instance, this study was a one-time butterfy survey that was conducted in a dry season for eight days as compared to that of the Kipembawe Division which sampled butterfies in four months [15].Te rapid survey is commonly acknowledged to be missing the essential rare species and is usually biased toward the common popular species [49,50].In addition, diferent methods of data collection used by the researchers may have contributed to such variation in the number of butterfy species.Furthermore, the higher number of butterfy species obtained from the current study compared to those in the wildlife management areas, in Southern Tanzania [48], could be due to geographic and climatic variations as well as diference in the level of anthropogenic disturbances.Wildlife management areas of the Ruvuma landscape are signifcantly being infuenced by anthropogenic disturbances which might be negatively afecting its butterfy species compared to Usangu area which has a restriction of human-induced activities and strong protection after being annexed into Ruaha National Park.In this study, Nymphalidae had the highest species richness (46 species) followed by Pieridae (30 species) which were sampled in all surveyed habitats.Te fndings echo those of other researchers in Southern Tanzania and other parts of the country who reported Nymphalidae had the highest species richness in their studies [29,51].Te highest number of Nymphalidae species could be attributed to several factors.First, Nymphalidae species owing to their polyphagous nature and greater ability to fy are capable of inhabiting diferent habitats in search of resources [52][53][54].Second, the Nymphalidae family is one of the most diverse groups of butterfies [55,56].It is worth noting that butterfy species within Nymphalidae and Pieridae families were also sampled within all surveyed habitat types in Usangu area.Te broad range of adaptation and the broad range of habitat preferences within Usangu area might be contributing to their occurrence in all surveyed habitats.Te Nymphalidae are mostly fruit-feeding butterfies; therefore, they occurred in large numbers in Miombo woodland followed by riverine forest [57][58][59].Pieridae also showed a greater dominance in all four surveyed habitat types in Usangu area.Teir higher dominance might be attributed to the availability of host plant species for Pieridae butterfies in all surveyed habitat types in Usangu area [60].Furthermore, it is worth noting that the greater dominance and highest number of Pieridae emphasize the fact that the majority of species in this family are indications of disturbed environments [61,62].Although Usangu area has been annexed into the national park with strong protection measures, the area still contains habitat patches that were exposed to intense anthropogenic pressure and ornamental plants which contribute to the greater dominance of these species.Results also show for the frst    12 Psyche: A Journal of Entomology time that even after long-term habitat restoration (over 15 years), it was not enough to eliminate the impacts of past anthropogenic disturbance on butterfy fauna.Conversely, Lycaenidae was a co-dominant family with 16 species that were recorded.Lycaenidae was found in the grassland, Miombo woodland, and riverine forest habitats and no other species of this family were recorded in the Vachellia/ Commiphora woodland habitat.Lycaenidae are commonly known to associate negatively with disturbance and they usually decrease in richness in disturbed environments [61,63].Based on the impacts of anthropogenic activities previously reported in the Usangu area [20], it is fair to suggest that the average number of species belonging to Lycaenidae has been infuenced by past disturbances and   changes in their habitat conditions in the area.On the other hand, Hesperiidae and Papilionidae families were poorly represented in the surveyed habitat types.Hesperiidae family showed low abundance and richness in grassland, Miombo woodland, and riverine forest while members of this family were completely absent in Vachellia/Commiphora woodland habitat (Tables 2 and 3).In addition, only two individuals of Papilionidae family were sampled only in grassland and riverine forest (Tables 2 and 3).Tis fnding is consistent with the results of the authors of [51,64] who reported that the group members of the Hesperiidae and Papilionidae families were least represented in their research studies in the country.Te low family abundance Hesperiidae and Papilionidae that was observed in grassland, Miombo woodland, and riverine forest may be attributed to the type of habitat and host plant availability as butterfy species richness and abundance are linked to the host specifcity and niche preferences [65].In addition, the diffculties in detection and identifcation by observers which are attributed to their small size, inconspicuousness, dull color, rapid and erratic fight patterns after any disturbance, and their low dispersal ability may also be among the probable reasons for low richness in families in Usangu area [65,66].
Overall, various habitat types of Usangu area in Ruaha National Park are suitable for the survival of butterfies.In this study, we aimed to assess the biodiversity of arthropods using butterfies as an indicator species and setting up the foundation for diferent habitat type preservation and butterfy conservation in Usangu area.Our study design included various habitats with diferent species composition and coverage, permitting us to conclude which habitat type management was sufcient to improve biodiversity conservation in Usangu area.Te fndings showed that the four habitat types in Usangu area did not have diferential efects on butterfy species diversity, richness, evenness, and abundance.However, there was evidence that Miombo woodland and riverine forest habitats exhibited relatively higher species diversity, richness, evenness, abundance, and higher number of habitat-restricted species, which supports the arguments that habitats with high abundance of trees and canopy cover contribute more to biodiversity conservation than in open areas such as grassland habitats [45,67,68].Tis study has shown that Miombo woodland had the highest species diversity, richness, evenness, and abundance compared to other habitat types; this concurred with previous fndings [29,48,69,70].Furthermore, the availability of open areas in the Miombo woodland such as glades, bogs, and clearcuts that could serve as alternative habitats for grassland butterfy species is another possible reason for the highest species richness, diversity, evenness, and abundance in the Miombo woodland [71][72][73].Te riverine forest was the second most diverse and rich habitat in our study area.Higher species richness and diversity in riverine forest habitats might be attributed to the availability of water which is well known for providing an ideal environment for butterfies, allowing for increased abundance, diversity, and richness due to the availability of resources such as basking sites with high sunlight, supporting butterfy behavior like puddling and through its benefcial efects to the host plant species for most butterfy species [74][75][76].Te fndings of the present study are also in line with the fndings 14 Psyche: A Journal of Entomology in Miombo woodland ecosystems, where the increase in woodland and canopy cover and high tree density resulted in higher butterfy species abundance, evenness, richness, and diversity in Miombo woodland habitats compared to deforested or grassland habitats [51,77].Furthermore, additional indirect evidence for the importance of high tree density and high canopy cover in the tropics is provided by this research indicating that the negative efects of habitat fragmentation on butterfies in Miombo woodland ecosystems decrease as the environment becomes more dominated by high tree density and greater forest cover.On the other hand, Vachellia/Commiphora woodland and grassland habitats recorded the lowest butterfy diversity and abundance among the habitats in the Usangu area perhaps due to their small areas or shortage of varied microclimatic conditions.Tis observation supports other studies comparing habitats with high tree density and forest cover and low tree density or grassland habitats, which found low species abundance and richness in habitats with low tree density or grasslands compared to those with high plant species richness and tree density [15,29].Habitat and larval host plants are among the fundamental resources that are required by butterfies for their survival and reproduction [40,78].During the study, it was observed that the Vachellia/Commiphora woodland and grassland habitats have a low species diversity, richness, and abundance.Te main reason behind this could be attributed to the lack of vegetation heterogeneity.Tis lack of heterogeneity can cause the lack of adequate larval food plants as well as reduce suitable microhabitat conditions that are essential for the survival of butterfy species [43,79,80], low plant species richness and density [80][81][82], and limited nectar sources for adult butterfies and developing larvae due to insufcient host trees, shrubs, and herbs [83,84] can negatively impact butterfy species richness, abundance, and community structure.
Although Miombo woodland, riverine forest, and grassland habitats had the highest species richness compared to Vachellia/Commiphora woodland habitat, the species accumulation curve and the estimated species richness (Figure 4 and Table 4) indicated that the sampling efort in the three habitats in Usangu was incomplete and there was a possibility to discover more butterfy species at each of these habitats with a substantial investment of more sampling efort [85].Te low species richness in Miombo woodland, riverine forest, and grassland habitats shows that short-time insect surveys tend to be biased toward common, well-known species and therefore leave behind the rare species which are usually added later with an increased sampling efort [86].Te rare and less common species are usually very important from a conservation perspective, and rapid diversity assessments may miss important elements of conservation interest [86,87].Te rapid and short inventory assessment might have missed many species in the surveyed habitats in the Usangu area.Terefore, our results suggest that diferent survey eforts may be required to equally assess butterfy biodiversity in diferent habitat types in the Usangu area.On the other hand, the small sizes, and habitat separation as a result of habitat fragmentation and sampling season, might have also contributed to the low butterfy species richness obtained from these habitats [88,89].Aaden et al. [74] denoted that most insects in tropical regions usually have high fecundity in the wet season and drops in the dry season.Tis study was conducted during the dry season between May and June, and hence most butterfy species could have remained undetected or unrecorded in the surveyed habitats during this nonfowering season in the Usangu area.Te estimates of species richness are essentially a scale-dependent process and are very strongly infuenced by the sampling period [87,90].
Te Bray-Curtis similarity index of butterfy species between the four habitat types in Usangu area showed that the level of species similarity between habitats of Usangu area was generally low, except similarity between Miombo woodland and riverine forest habitats.Generally, the composition of butterfy species had many similarities in the Miombo woodland and riverine forest followed by Vachellia/Commiphora woodland and grassland habitat, while the lowest similarity was observed between Miombo woodland and grassland habitats (Table 5 and Figure 5).Te lowest species similarity recorded between habitats in Usangu area could be contributed to habitat specifcity of butterfy for host plant species that provide food for butterfy.Jj and Emana [45] stressed that the distribution of any species is regulated by the distribution of its habitats as well as the availability of food, shelter, and other essential resources found within that particular habitat.Furthermore, habitat loss, fragmentation, and isolation which resulted from past unregulated anthropogenic activities in the Usangu area could also be plausible reasons for the low species similarity between these habitats.Habitat loss and fragmentation have also been acknowledged as among the leading causes of the current biodiversity reduction globally [91].Fragmented habitats can constrain species movement, disrupt species dispersal behavior and population, and reduce connectivity by increasing spatial isolation between suitable patches and hence negatively impact biodiversity [92][93][94].Te study fndings align with other previous works in the country [29,51] that reported the same conclusions.It is worth noting that Miombo woodland and riverine forest were the favorable habitats for the butterfy in the Usangu area.Tese two habitats provide them with suitable environmental conditions for their survival and growth all year round.Tey support high tree abundance, host plants to carry out their life cycle, and high richness and diversity of vegetation, thus meeting butterfy requirements such as food, shelter, and physical environmental factors, such as the presence of water fow and favorable weather conditions [80,82,95].

Conclusions
Te current investigation is the frst to generate a butterfy checklist of Usangu area and document the spatial distribution of butterfy species in the four main habitat types that are found in Usangu area, Ruaha National Park (RUNAPA).Regardless of its size, Usangu area is rich in butterfy species that are known to occur in Tanzania.Te situation refects the availability of diverse and rich habitats that provide suitable habitats for butterfy in Usangu area, Ruaha National Psyche: A Journal of Entomology Park.Te study fndings also have provided baseline information to conservationists and researchers for more studies in similar restored areas annexed to National Park.Te study recommends that further studies on abiotic characteristics of habitat types, degree of connectivity or isolation between habitats, and types of host plants that provide food for butterfies in various habitats in Usangu area should be done.Te investigation should consider diferent seasons in Usangu area as this study was only carried out during the dry season only with repeated sampling eforts.From a conservation perspective, the results presented here suggest that the habitat types in Usangu area are still recovering from past anthropogenic impacts, and for butterfy conservation, RUNAPA should focus on preserving the Miombo woodland and riverine forest by maintaining the habitat stability of the two habitats as they harbor greatest species richness and abundance of butterfies in Usangu area compared to other habitat types.Te Miombo woodland and riverine forest habitats in Usangu have the potential for butterfy tourism, which can aid in conservation eforts and raise public awareness.However, it is crucial to carefully monitor the impact of tourism development on these areas, as any alteration to the Miombo woodland, plant biomass, and water resources in the riverine forests could harm the butterfy diversity in this region.To ensure the preservation of this unique ecosystem, conservationists and park managers should establish a plan for ongoing monitoring and research to observe any changes in butterfy species composition, diversity, and richness in the Usangu area.

Figure 1 :
Figure 1: Map showing the location of four sampled habitat types in Usangu area, Ruaha National Park.

Figure 2 :
Figure 2: (a) Family-wise composition of butterfy species in Usangu area.(b) Status of butterfy species in Usangu area.C � common, NR � not rare, R � rare, VC � very common, and VR � very rare.

Table 2 :
Te distribution and abundance of butterfy species in the four habitat types in Usangu area, Ruaha National Park.

Figure 4 :Figure 5 :
Figure 4: Sample-based rarefaction curves of estimated species richness in four habitat types in Usangu area, Ruaha National Park.

Figure 6 :
Figure 6: Nonmetric multidimensional scaling (NMDS) plot based on a Bray-Curtis abundance matrix, showing patterns of butterfy community structure among the four habitat types in Usangu area.
Psyche: A Journal of Entomology surveyed habitat types in Usangu area (Global R � 0.2902, P � 0.0022) using the Bray-Curtis similarity index.

Table 3 :
Family-wise composition of butterfies in terms of species richness and number of species restricted in four habitat types in Usangu area.

Table 4 :
Species richness and sampling completeness in the four habitat types in Usangu area.

Table 5 :
Bray-Curtis index for observed butterfy assemblages across diferent habitat types in Usangu area.