Impacts of Human Activities and Season on Species Diversity and Abundance of Butterflies in Mpanga Kipengere Game Reserve and Surrounding Farmlands, Tanzania ()
1. Introduction
Identifying high-value sites based on their biodiversity content is a crucial aspect of any conservation strategy [1] . Unfortunately, in recent years, human activity has put increasing pressure on biodiversity, posing challenges for biologists dealing with anthropogenic disturbances [2] [3] . Human activities are known to cause environmental changes that have adverse effects on plants and animal species in protected areas [4] [5] . Human activities that cause disturbances in natural areas can have a direct impact on important species’ needs, such as food, cover, and nesting sites, according to [6] . Among these species, butterflies are known to be particularly sensitive to environmental changes, as noted by [2] [7] . Natural habitats play a crucial role in the conservation of insects and other arthropods, providing them with essential elements like food, shelter, and nectar [8] [9] [10] . When natural ecosystems are disrupted, it can have negative consequences for plants and animal species [11] . Insects, as a major taxonomic group, have been particularly impacted, exemplifying these challenges [12] [13] .
The Mpanga Kipengere Game Reserve (MPKGR) is known for its rich biodiversity and high levels of endemism in both flora and fauna [14] . To effectively conserve this biodiversity, it is necessary to have a broad understanding of the classification, distribution, and biogeography of various indicator species [11] [15] . Among these, butterflies are considered a prime group for such assessments [16] as they can provide valuable information on environmental changes and help monitor and assess ecosystem health [13] [17] [18] However, the only available information on the butterflies in MPKGR is from a biodiversity survey conducted by Frontier Tanzania in collaboration with the University of Dar es Salaam and WWF back in 2003. Previous research has mostly focused on vertebrate species in the area, with little attention given to invertebrates beyond the game reserve where human activities take place.
To gain a better understanding of the diversity patterns of invertebrate species in different areas with varying conservation designations, it is essential to gather more data on a wider range of taxa [19] [20] . This will provide valuable insight into effective biodiversity management on a larger scale. By establishing a baseline for future monitoring, we can work towards preserving the biodiversity in this ecosystem, which has been impacted by human activities. Therefore, this study was conducted to provide valuable information to MPKGR Management regarding butterfly conservation, ecological monitoring, and ecosystem health assessment. Its objectives were to:
1) Evaluate the impact of human activities on butterfly species diversity and abundance in both MPKGR and adjacent farmlands.
2) Analyze the occurrence and seasonal variations of butterflies in MPKGR and compare them to those in adjacent Farmlands.
The study’s hypothesis is that butterfly species composition and diversity will differ between natural and disturbed habitats during various seasons.
2. Material and Method
2.1. Study Area
The Mpanga Kipengere Game Reserve (MPKGR) is situated in the Southern Highlands within Wanging’ombe and Makete districts of the Njombe Region, as well as the Mbarali District in the Mbeya Region. Its latitude ranges from 8˚50' to 9˚10' South, while its longitude ranges from 34˚00' to 34˚30' East. The reserve is surrounded by 24 villages, which are divided into five divisions: Wanging’ombe (7 villages), Imalinyi (3 villages), Ikuwo (9 villages), Lupalilo (1 village), and Rujewa (4 villages). To reach the reserve, one can use road, railway, or air transportation as it is located near Mbeya (135 km) to the southwest, Njombe (80 km) to the southeast, and Iringa (195 km) to the northeast. Other towns such as Makambako, Ilembula, Igawa, and Chimala are also nearby and are experiencing rapid growth. The rainfall distribution in MPKGR is greatest at higher altitudes and is peaking between March and May [21] . Rainfall is greatest in the southeast of the mountains, increasing from 1200 mm annually in the foothills to over 2300 mm at higher altitudes [21] . The dry season occurs from June to August and the wet season starts from November to May. The vegetation of these forests’ ranges includes lowland forests at 300 m on the Eastern side, sub-montane forests, and montane forests (Figure 1).
Figure 1. Map of Mpanga Kipengere Game Reserve showing location, districts, transects, and sampling points along MPKGR and adjacent Farmlands.
2.2. Sampling Design and Data Collection
For the study, two transects measuring 1000 m each were created. One-half of each transect spanned 500 m inside the MPKGR while the other half extended 500 m into adjacent farmlands. Within each transect, 20 sampling points with 50-meter distance from one point to another were established, and the study period lasted for 12 consecutive months, from November 2021 to October 2022. During this time, butterflies were sampled for 10 days each month, during two time slots: 9:00-11:00 am and 3:00-5:00 pm. The sampling methodology outlined by [14] was followed, and two methods were employed to collect butterfly samples: baited traps and sweep nets.
2.3. Data Collection
Field data sampling
To catch butterflies attracted to fermenting fruit, we used traps baited with fermented bananas. We followed the process described by [22] [23] and constructed the traps from local materials based on the Van Someren-Raydon Trap design [24] . The bait was made by mashing ripe bananas and pineapples and then allowing them to ferment for three days. We placed traps at the center of each sampling point, 100 meters apart. Regular checks were performed, and the number of butterflies caught was recorded as individuals per trap per day.
To collect butterflies from areas where traps couldn’t be placed easily, we used sweep nets based on the methodology outlined by [25] . We spotted flying butterflies along the transect or around the traps and caught them using the nets. Once collected, we identified each butterfly species using the key described by [17] and counted them. For harder-to-identify individuals, we took photos since we couldn’t remove them from the Game reserve. These photos were then shared with butterfly taxonomists to confirm their identification.
Environmental data
To comprehend the impact of environmental factors on species richness, abundance, and community composition, we obtained data on annual temperature, mean annual precipitation, and solar radiation from (https://weatherandclimate.com/tanzania/njombe/kipengere#t3). Furthermore, we recorded the type of habitat at all sampling points. Topographic factors including elevation, slope, and aspect were extracted from the raster layer derived from the SRTM 30 m-based DEM-USGS Earth Explorer (https://earthexplorer.usgs.gov/).
2.4. Data Analysis
To compute variations between habitats (MPKGR and Farmlands) and seasons (Dry and Wet) we calculate their butterfly species abundance, diversity, richness, and evenness. In determining the species richness and abundance of butterflies, a species checklist was created. This checklist consisted of four variables: the name of the butterfly species, its family, the number of individuals counted, and its habitat and season of occurrence. Jaccard’s Similarity Index was used to measure similarities of butterfly species diversity between MPKGR and Farmlands. The formula for Jaccard’s Index was J(A, B) = |A ∩ B|/|A ∪ B| where the Jaccard’s Similarity Index value is 1 indicates that two datasets share the same members, and if there are no common members then the Jaccard Similarity Index will be 0.
Computer software Palaontological Analysis (PAST) was used to compute the Shannon Wiener Diversity Index (H) and to plot rarefaction interpolation and extrapolation curves to ensure that the sample size we use is enough and Pielou’s evenness index to compare butterfly species distribution between MPKGR and Farmlands. The formula for the Shannon index H = −Σ [(pi)*log(pi)] (Shannon-Wiener, 1949) was used where; Pi is a proportion (n/N) of individuals of one particular species found (n) divided by the total number of individuals found (N), ln is the natural log, Σ is the sum of calculations and s is the number of species while Pielou’s Index formula was J = H’/ln(S) where H’ is Shannon Weiner diversity and S is the total number of species in a sample, across all samples in a dataset. Indicator species analysis was performed to identify significant dominant species of the butterflies’ communities, using the computer software PAST.
3. Results
3.1. Butterflies’ Community Composition Diversity, Richness, and Abundance
In our study, we documented 124 butterfly species from five Lepidoptera families (Nymphalidae, Pieridae, Lycaenidae, Hesperiidae, and Papilionidae), with a total of 2799 individuals observed. Neptis morosa was observed to be the most dominant of all species with a total abundance of 0.03358 whereas Pseudacraea lucretia expansa were the most dominant species of all individuals recorded in farmlands (Appendix). Shannon wiener diversity index indicates that MPKGR is more diverse (H = 4.49) compared to Farmlands (Figure 2). Butterfly species in MPKGR and Farmlands have even distribution according to Pielou’s evenness index, with values of 0.947 and 0.91 respectively. Additionally, the results reveal that both habitats share most of the same species, as the Jaccard’s Index value calculated was 0.629.
3.2. Species Found in Specific Habitat
A total of 124 butterfly species (MPKGR-110 and Farmland-92) were observed, whereby 78 butterfly species, which account for 62.9% of all recorded species, were observed in both MPKGR and Farmlands. Out of these, 32 species (25.8%) were only seen in MPKGR, while 14 species (11.29%) were exclusive to Farmlands. The family Nymphalidae accounted for the majority of species observed in both locations. Most of the butterfly species observed to be specific in either MPKGR or Farmlands belong to the families Lycaenidae, Pieridae and Nympharidae (Appendix). We believe that our butterfly sampling was thorough enough for the time and season of our survey, as the rarefaction extrapolation curves almost reached asymptotes in both MPKGR and farmlands (Figure 3).
Additionally, the rarefaction and extrapolation curves based on coverage percentage suggested that the diversity in our study region was well-represented with a sample coverage percentage above 95% as shown in (Figure 4).
Figure 2. Shannon wiener diversity index (H) computed from Computer software Palaontological Analysis (PAST). The result indicates the diversity is higher in MPKGR compared with Farmlands.
Figure 3. The size-based rarefaction and extrapolation curves show the species richness at MPKGR and adjacent Farmlands, with the inner line representing interpolation and the outer line representing the 95% confidence intervals.
Figure 4. The rarefaction and extrapolation curves based on size display butterfly diversity and sample representation at MPKGR and adjacent Farmlands with 95% confidence intervals.
3.3. Seasonal Variations and Monthly Occurrences of Butterflies
According to the results, the month of April during the rainy season had the highest number of species (96) with a total of 524 individuals making up 77% and 18%, respectively of all species and individuals collected. The average temperature and rainfall were recorded at 26.44˚C and 6.7 mm. On the other hand, during the dry season in September, the number of species dropped to 39 (31.45%) with only 146 individuals (5.22%). The average temperature and rainfall were reported at 24.94˚C and 0.02 mm, respectively (Table 1).
3.4. Indicator Species and Endemic Species
We have identified a total of 9 butterfly species that serve as indicators for both MPKGR and Farmland. Of these 9, 6 species (Neptis morosa, Neptis serena serena, Colotis antevipe zera, Colotis auxo incretus, Azanus ubaldus, Graphium antheus) are considered generalists as they can be found throughout the year in both habitats. The remaining 3 species are specific to their respective habitats and can only be found during the wet season. These species are Acraea servona, Pseudacraea lucretia expansa, and Acraea pudorina. This study also observes the presence of two endemic species, Charaxes congdoni and Harpendyreus juno which are endemic to the southern highlands, including the MPKGR area.
4. Discussion
Species diversity and abundance
Our findings indicate that MPKGR recorded a significantly larger diversity and abundance of butterflies than the adjacent Farmlands. The reason behind the high number of individual and species of butterflies observed in Mpanga
Table 1. Monthly occurrence of butterfly species in relation to the influence of temperature and rainfall.
Kipengere Game Reserve could be due to the miombo woodland present there, as opposed to the cultivated areas in the Farmlands. This suggests that woodland and wooded grassland habitats may provide a better quality of life for butterflies as explained by [26] [27] . These qualities include the availability of larval host plants and food resources in these habitats [28] [29] . Additionally, the low abundance and species richness recorded in the Farmlands could be attributed to the habitat disturbance caused by human activities like tree cutting [30] [31] [32] . Such habitat disturbances directly remove the required conditions for butterfly breeding, thus affecting their overall abundance and richness [33] [34] .
Species found in specific habitat
A total of 78 butterfly species, which account for 62.9% of all recorded species, were observed in both MPKGR and Farmlands. Out of these, 32 species (25.8%) were only seen in MPKGR, while 14 species (11.29%) were exclusive to Farmlands. The family Nymphalidae accounted for the majority of species observed in both locations. Most of the butterfly species observed to be specific in either MPKGR or Farmlands belong to the families Lycaenidae, Pieridae and Nympharidae (Appendix) This suggests that the environmental and climatic conditions in both habitats are favorable for some species. Vegetation is also a factor that affects butterfly composition across habitats. Conversely, butterfly species that are only observed at specific locations indicate that they have specific requirements for survival at that particular location [17] [29] .
Seasonal variation and Temporal occurrence of butterfly
During the wet season (January to June), there is a significant increase in butterfly species diversity compared to the dry season. In MPKGR, there is a small number of species recorded during the dry season (July to December) which may be due to environmental conditions. During this season, most of the plants are dry and affected by the dry conditions. Although trees and shrubs are present, many of them cannot be used by butterflies due to their deciduous phenology.
The study noticed a seasonal population fluctuation as there were more butterflies during February, March, April, and May, but fewer during June to December. Research has indicated that seasons play a role in influencing the quantity and variety of insects present. It is also revealed that during the transition from short to long rains (January to March), we observed a higher number of butterfly species and a greater abundance of them than during the long to short rains transition that is April to June (Figure 5) These findings align with previous research conducted on Kihansi gorge, which found that butterfly species richness and abundance were higher during the dry season compared to other seasons [23] .
The months of June and July are the coldest in Mpanga Kipengere and adjacent farmlands, with maximum temperatures around 16˚C (Table 1 and Figure 6). Additionally, from June to mid or late October, a dry season occurs in the reserve which can result in scarce butterfly food. These cold conditions and low food availability may have contributed to poor detection or resulted in lower numbers of Hesperiids, Lycaenids, and Pierids in the reserve and surrounding farmlands. The Nymphalidae is currently the largest butterfly family in Tanzania, with 657 species [24] [25] and therefore has the highest occurrence throughout the year compared to other butterfly families (Appendix).
Figure 5. Butterfly species observed each month in MPKGR and adjacent Farmlands during the wet season (January-June) and dry season (July-December).
Figure 6. Show the butterfly species and individual counts in relation to climatic variables (temperature and rainfall).
Indicator Species and endemic species
It is crucial to identify a set of indicator species for long-term environmental monitoring in conservation and biodiversity management [35] [36] [37] . We have discovered 9 indicator butterfly species that are associated with different habitat quality in the ecologically sensitive areas. These species can be helpful for future monitoring and assessment of biodiversity in the area. The majority of these indicator species are habitat generalists and polyphagous, while a few are habitat specialists, monophagous, and have a small wingspan, such as Euryphula concordia, Colotis danae, and Neptis jordani. This suggests that they have limited dispersal ability and are highly dependent on specific habitats that may only occur in certain environmental conditions, as observed in previous studies [38] [39] . Furthermore, studies have shown that the plant-abundance relationship is strongest for butterfly species that are habitat specialists, monophagous, and less mobile, as reported by [40] [41] . This study highlights the presence of two endemic species, Charaxes congdoni and Harpendyreus juno, in the Livingstone mountains and southern highlands, including the MPKGR area as previously reported by [21] [42] . Their existence in this region underscores the significance of MPKGR as a crucial area for biodiversity preservation.
5. Conclusion
Understanding the impacts of human activities and seasonal variations on species diversity and the abundance of butterflies in an ecosystem is important to inform the conservation of existing Game Reserves. Our findings suggest that there is a significant difference in butterfly diversity and abundance between MPKGR and adjacent farmlands with higher diversity in MPKGR where the land is free from anthropogenic disturbance. The large and significant variation in butterfly diversity and species community explained by anthropogenic and environmental factors suggests a need for conservation plans for the natural habitats of MPKGR which is under threat from anthropogenic disturbance from adjacent farmlands. The butterfly species that were specific to certain locations may serve as ecological indicators because they appear to be favored by the environmental conditions of those locations. Future studies looking into how various individual butterfly species are influenced by the available qualities of the habitats will be necessary in generating information that will be useful in identifying species-specific needs for improving the conservation of the butterfly community in MPKGR. This study reveals that Southern Highlands, including the MPKGR area, is home to two unique species, Charaxes congdoni and Harpendyreus juno. Their presence in this region emphasizes the importance of MPKGR as a critical area for conserving biodiversity. We need further research to determine how the anthropogenic activities on the farmlands could be affecting the diversity and abundance of these indicator species in MPKGR which has already recorded the extinction of large mammals in the recent decade due to human development activities.
Appendix
Table A1. Abundance, species richness, and species occurrence in specific habitat i.e., in MPKGR, Farmlands, or both. The abbreviation GR-Represent species recorded from MPKGR only, FL-Represent species recorded from farmlands only, and GR + FL-Represent species found in both MPKGR and Farmlands.