Lepidoptera as a tool for human disturbance assessment on biodiversity in Choke Mountains, Ethiopia

doi:10.21203/rs.3.rs-2064956/v1

Download PDF

Research Article

Lepidoptera as a tool for human disturbance assessment on biodiversity in Choke Mountains, Ethiopia

https://doi.org/10.21203/rs.3.rs-2064956/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background:

Recent studies indicated that frequent anthropogenic disturbances in tropical countries are primary drivers of the reduction in community diversity and local extinction of many insect taxa, including Lepidoptera. Biodiversity assessment provides information on the responses of different taxa to these anthropogenic changes, useful to develop evidence-based conservation strategies to reduce biodiversity loss. We assessed the impact of anthropogenic disturbances on lepidopteran assemblages across five different land uses (forest fragments, crop fields, Pasture land, rural settlement and undisturbed natural forest) in Choke Mountains, Ethiopia. Automatic light traps have been used in 16 sites, in 3 sites manual samplings have been performed. A total of 4,559 moths representing 14 families and 339 species were sampled. The highest diversity was obtained from the Natural Forest (269 species), followed by Forest Fragment (174), Pasture Land (148), Crop Fields (138), and Rural Settlement (136). The highest species-rich families were Geometridae (131 taxa), Erebidae (93 taxa), Tortricidae (40 taxa) and Sphingidae (17 taxa). Values calculated for the Shannon index were high, ranging between 4.78 for Natural Forests and 4.15 for Crop Fields. Simpson index indicated very high values of evenness, with all land use types above 0.97. The highest values of dominance were observed in Crop Field and Pasture Land, with the percentage of the dominant species around 10%. Estimated species richness by Chao-1 was highest within Natural Forest (366), followed by Pasture Land (256), Forest Fragment (241), Rural Settlement (233), and lowest in Crop Field (186), with the highest increase, if compared to the observed number of species, in Pasture Land (+ 73%) and Rural Settlement (+ 71%).

Conclusions:

The results of this study will be useful for guiding conservation management priorities for preventing irreversible biodiversity losses and maintaining ecosystem services provisioning that are essential conditions for the sustainable development of rural communities.

moths

Africa

land use

ecosystem monitoring

biodiversity index

In the last decades, there is a growing change in natural environments around the world, due to an alarming increase in the human population and its activity. Especially, the conversion of forests into agricultural fields and pastures is one of the main drivers of biodiversity losses globally [1, 2, 3]. These losses of biological diversity will negatively affect vital ecosystem functions that regulate the Earth system upon which humans ultimately depend.

Insecta represents the most species-rich taxon and provides a wide range of ecosystem processes and conditions to sustain human life [4]. In the meanwhile, they are among the taxa most severely and quickly affected by changes in an ecosystem and insect biodiversity losses can have negative effects on all ecosystem communities. The ecological role of entomological fauna makes it a useful bio-indicator to investigate the health status, the dynamics and functioning of ecosystems and the key factors related to their modification. Most importantly, the quantitative measures of insect species and individual abundances provide standardized indicators for the biodiversity assessment useful to compare various habitats, communities, and ecosystems.

The scientific understanding of insect biodiversity responses to human activity is a key element to evaluate the functional consequences of human disturbance on ecosystems [5]. Insect ecological monitoring has been shown to play a fundamental role in understanding the structural and functional biodiversity of ecosystems. In particular, it is the most effective tool for filling important knowledge gaps in many ecosystems and communities that are still poorly investigated. Moreover, information gathered from monitoring can be used as an early warning system for quick action against biodiversity loss, supporting policy intervention and the implementation of management practices to prevent, or at least delay, adverse changes. Monitoring also provides information useful to better understand the implications of a changing environment [6, 7]. Habitat modification and fragmentation have been considered the primary driver of Arthropod species losses worldwide [8]. Agriculture, deforestation, and urbanization habitats, and the practices associated with them, could account for two-thirds of observed insect declines [9]. It is therefore critical to understand how different land use practices may impact biodiversity and related ecosystem services in these managed habitats. Lepidoptera account for the second largest, most diverse, widespread, and widely recognized insect order in the class Insecta. Lepidopterans play an essential role in environmental indicators as they are sensitive and react quickly to subtle changes in environmental and habitat conditions [10, 11]. Due to their short life cycle, narrow niches and relatively low mobility, they are more sensitive to land-cover and land use changes than long-lived animals [11]. Most importantly, Lepidopterans are valuable indicators of environmental quality, considering their high degree of host-plant specialization and vulnerability to habitat deterioration [12]. Given their presence in a broad range of habitats, the loss of Lepidoptera may directly influence the provisioning of key ecosystem services, such as pollination and natural pest control [13].

Biodiversity and ecosystem services are the bases for the economic and social development of Ethiopia, the second largest population in Sub-Saharan Africa. Over 80% of the total estimated population of 120 million people lives in rural areas and depends on natural resources for their livelihoods, economic development and food security. Current and future economic, social and ecological contributions of biodiversity conservation and sustainable use of the resources are not only technical concerns but also political imperative as well.

This research work aimed to assess the impact of human interference in five different land uses, using lepidopteran biodiversity as bioindicators. The study was performed in the Choke Mountains (Ethiopia), an area historically covered by forests, natural grassland and shrub lands. Increasing human population, deforestation and land degradation broadly led to a change in land use, with the extension of cultivation and pasture area in a context of subsistence farming [14, 15]. We compared biodiversity assessment in natural forests, considered the benchmark, with the assessment in four different ecosystems altered by human activities. In particular, we considered the abundance and evenness of Lepidoptera biodiversity at the species level and the relationships with plant richness.

Overall, 4,459 moth individuals belonging to the selected 14 families and 339 species were recorded across the five land use types. Of the 339 taxa found, 67% were given a species name, while in 33% of cases the determination was made at the genus level. There were 41 singletons and 28 doubletons. The highest diversity was obtained in the Natural Forest (269 species) followed by Forest Fragment (174), then Pasture Land (148), Crop Fields (138) and Rural Settlement (136) (Fig. 1). The highest species-rich families were Geometridae (131 taxa), Erebidae (93 taxa), Tortricidae (40 taxa), and Sphingidae (17). NF showed by far the highest number of taxa compared to the total number of taxa found for 3 families (Geometridae, Erebidae, Sphingidae). For the other families, such as Tortricidae, the number of species was evenly distributed among two or more land use types (Table 1).

Table 1

Distribution of the number of Lepidopteran species and individuals (in square brackets) by family and land use types collected in the Choke Mountains. Families are listed per descending overall abundance.
Family	Land use types					Overall
Family	CF	PL	FF	RS	NF	Overall
Geometridae	46 [172]	43 [128]	59 [207]	43 [186]	109 [692]	131 [1,385]
Tortricidae	24 [114]	20 [117]	26 [146]	29 [187]	32 [376]	40 [940]
Erebidae	41 [201]	42 [226]	52 [235]	40 [113]	72 [383]	92 [775]
Sphingidae	7 [16]	11 [79]	11 [32]	7 [18]	14 [73]	17 [218]
Saturniidae	0 [0]	6 [12]	1 [3]	1 [3]	7 [100]	8 [118]
Yponomeutidae	3 [24]	2 [10]	3 [15]	3 [40]	3 [27]	3 [116]
Crambidae	3 [5]	5 [22]	3 [5]	3 [13]	8 [51]	10 [96]
Lasiocampidae	6 [11]	5 [5]	6 [19]	4 [14]	8 [30]	10 [79]
Limacodidae	1 [5]	2 [15]	3 [17]	2 [13]	4 [25]	4 [75]
Nolidae	2 [4]	4 [17]	3 [16]	2 [9]	3 [14]	5 [60]
Uraniidae	2 [6]	2 [9]	4 [11]	3 [5]	2 [12]	5 [43]
Cossidae	2 [3]	2 [3]	2 [3]	1 [1]	5 [10]	7 [20]
Notodontidae	1 [1]	2 [2]	1 [1]	1 [1]	3 [14]	5 [19]
Metarbelidae	0 [0]	1 [2]	1 [4]	1 [1]	0 [0]	1 [7]
Total						339 [4,559]
CF = Crop Field; PL = Pasture Land; FF = Forest Fragment; RS = Rural Settlement; NF = Natural Forest

Looking at the number of specimens sampled in the five land use types during the 1-year research period, captures in Natural Forest were by far greater than in the other 4 land use types, the latter showed small differences between them (Fig. 2).

The overall most abundant species was Digama meridionalis deliae Berio,1939 (Family Erebiidae) (4.93% of all individuals), which was found to be most abundant in PL, FF and CF. Following this species, we have Clepsis paragongyla Razowski & Trematerra, 2019 (Tortricidae) (4.88%), most abundant in NF, FF and CF; Chiasmia procidata (Guenée, 1858) (Geometridae) (4.57%) in NF, FF and RS; Thylacandra delimana Razowski & Trematerra, 2019 (Tortricidae) (2.80%) in FF, NF and CF; Xylopteryx emunctaria (Guenée, 1858) (Geometridae) (2.58%) in FF, RS and NF (Table 2).

Table 2

List of the most common species and their BP, separately calculated in each land use type
Land use	Rank	List of the most common species	BP
NATURAL FOREST	1	Chiasmia procidata	5.88%
	2	Clepsis paragongyla	5.12%
	3	Gonimbrasia birbiri	4.52%
	4	Aroterosia punctineata	2.61%
	5	Xanthorhoe sp.1	2.40%
FOREST FRAGMENT	1	Digama meridionalis deliae	7.28%
	2	Clepsis paragongyla	5.43%
	3	Xylopteryx emunctaria	5.03%
	4	Aroterosia punctineata	3.97%
	5	Metarctia carmel	3.44%
PASTURE LAND	1	Digama meridionalis deliae	10.44%
	2	Clepsis paragongyla	5.29%
	3	Chiasmia procidata	4.84%
	4	Epichoristos fekensae	4.39%
	5	Cyligramma latona	4.08%
RURAL SETTLEMENT	1	Thylacandra delimana	7.81%
	2	Digama meridionalis deliae	5.58%
	3	Xylopteryx emunctaria	5.10%
	4	Epichoristos fekensae	4.15%
	5	Chiasmia procidata	3.83%
CROP FIELDS	1	Alpenus geminipuncta	9.49%
	2	Clepsis paragongyla	6.44%
	3	Digama meridionalis deliae	6.44%
	4	Orthonama obstipata	5.08%
	5	Yponomeuta fumigata	3.39%

Unique species, i.e. species collected, in one or more specimens, in one land use only, were observed in all land use types. As expected, they were higher in NF (76 taxa), followed by FF (16), RS (10), CF (9) and PL (7). Other 34 taxa were collected in human-impacted land use types, i.e. CF, PL, RS and FF, but not in NF.

Values calculated for the Shannon index, in general, were high and ranged between 4.2 in CF and 4.8 in NF (Fig. 2). Simpson index indicated very high values of evenness, above 0.97 in all land use types (Fig. 2). The highest values of dominance were observed in CF and PL, with the percentage of the dominant species around 10% (Fig. 2). Using Chao-1 index, the highest species richness was found in NF (365) and the lowest in CF (186) (Fig. 2). If compared to the sampled number of species, the highest increase was observed in PL (+ 73%) and RS (+ 71%).

Biodiversity indices calculated on a monthly basis showed extremely variable trends (Fig. 3). Overall, NF possesses the highest species diversity followed by FF, PL, and RS. The crop field showed the lowest species diversity. In the case of the Shannon index, the most diverse sample (4.29) was recorded in CF in August and the lowest (1.57) in PL in December. All land uses showed significant temporal fluctuations. For example, NF had the maximum in September (4.10), after the rainy season, and the lowest in April (1.88), during the dry season.

The Simpson index was high (above 0.85) for most months, some sharp decreases occurred in RS in November (0.67), in PL in December (0.65) and in NF in April (0.77). The Berger-Parker index had a trend opposite to Simpson’s value. BP values were in most cases below 30%, with some high peaks corresponding to November in RS (56%), December in PL (57%) and April in NF (42%).

Chao-1 estimates were particularly high for NF in September (167) and October (176), and then decreased to the minimum in April (14). CF showed the highest values between September (90) and November (79), keeping at lower values for the rest of the year. For PL a very high peak was observed in February (180), whereas for the rest of the year, estimates were constantly below 80. FF had two peaks, in September (139) and May (131), and values below 80 in the remaining months. RS had the peak of species in May (95) and lower levels in the other months.

A total of 254 vascular plant species were sampled: 178 in NF (Ankober, Bradi and Delima Natural Forests), 62 in FF (Melit Church Forest), 43 in RS, 35 in PL, and 10 in CF. A high positive linear relationship $\left({R}^{2}=0.86\right)$ between the number of vascular plants and Lepidoptera sampled species was observed (Fig. 4). The relationship with estimated lepidopteran species by the Chao-1 index was lower $\left({R}^{2}=0.56\right)$ (Fig. 4).

Biological resources and biodiversity in African tropical forests remain largely unknown. This makes even more difficult the estimation and prediction of the consequences of habitat fragmentation and modification as they change the total area of suitable habitats and their connectivity, the climate, and the biological community composition. The effects of habitat perturbation and losses are reflected in a decline in populations and species losses [16]. Those impacts affect not only rare and specialized organisms but often interest widely distributed species as well [17]. Chapman et al. [18] assessed the conservation challenges facing the biodiversity of sub-Saharan Africa, examining the current threats, and their changes under drivers of pressure such as human population growth and movement, economic expansion, and climate change. The present study focused on a tropical mountain area, the Choke Mountains in Ethiopia, highly fragmented and with very high pressure due to grazing and extensive crop cultivation. The main driver of this pressure is the growth of the human population which causes an increase in the demand for food and natural resources, generating conflicts between different land use and making the preservation of biodiversity in the area more difficult [19].

As expected, in our study we observed a higher lepidopteran diversity in the Natural Forest then in other land use types. It must be emphasized here that even the three forests included in the study are not exempt from impacts caused by the activities of local people, such as logging for wood use and charcoal manufacturing. A gradient of diversity decrease was observed for Forest Fragment, Pasture Land, Rural Settlement and Crop Fields, and the pattern of change is correlated to the level of perturbation affecting the different land use types.

In our data, floristic and faunistic abundance appear clearly related, both considering the observed and estimated Lepidoptera species richness. This result was not surprising, considering that Lepidopterans are phytophagous during the larval stage, as reported by Siemann et al. [20], Axmacher et al. [21], particularly focused on the Afro-tropical Geometridae, and Kittle et al. [22]. The lower relationship obtained by comparing flora with lepidopteran species abundance estimated by the Chao-1 index can be explained by the high number of singleton and doubleton species that could have markedly incremented the estimated biodiversity of some sampling sites, such as in the case of NF1.

In all land use types, the Shannon index was high (> 4) even for habitats with a high anthropic disturbance, such as Rural Settlement, or structure simplification, such as Crop Fields. The estimates of the Simpson index follow the same trend as the Shannon index, highlighting a high evenness structure of all investigated Lepidoptera communities. The Berger–Parker index calculated for the first dominant species increased from undisturbed to disturbed areas, with Natural Forest values very similar to those calculated from Forest Fragment and Rural Settlement.

The estimated species richness calculated with the Chao-1 index is much higher than observed values for all investigated land uses (e.g., in Natural Forest, Chao-1 estimated 365 species versus 269 sampled taxa), suggesting that further samplings are needed to keep a more complete picture of the hosted fauna. In percentage, the highest increases in species richness estimates were observed in Pasture Land (256 species estimated versus 148 sampled) and Rural Settlement (233 species estimated versus 136 sampled), confirming the value, in terms of biodiversity, of habitats where the impact of human activities is significant.

The monthly trends of the four biodiversity indices showed strong differences among the five land use types. The peaks of greatest biodiversity occur at different times of the year depending on the type of land use. There are months when the NF is not the land use with the greatest biodiversity, e.g. in April the Natural Forest showed the lowest Shannon index; in February-March and in June and July the highest Shannon index values were observed in Rural Settlement and Forest Fragment, respectively. Simpson index had two very low values calculated for Rural Settlements in November and Pasture Land in December, highlighting a structure dominated by few dominant species, as reflected also by the high Berger-Parker index in the same months. In the case of the Chao-1 index, the highest estimates were obtained in October for Natural Forest and in February for Pasture Land.

Our results showed that a significant proportion of Lepidoptera biodiversity in the Choke Mountains is hosted also in habitats where human modifications are more evident and the tree component is reduced. It is the case of the Rural Settlements, where, among the tukul in which the local human population lives, a more abundant lepidopteran fauna has been sampled than might have been expected, with as many as 10 species found in this land use alone. For several species, the population density will be highest in optimal habitats, but individuals will disperse into low-quality habitats creating a gradient from high-performing ‘source’ to low-density ‘sink’ populations that are maintained by immigration from source populations [23].

Forest fragments, often of a few hectares, play a very important role in maintain biodiversity in areas (such as the one under investigation) where larger forests have almost disappeared. The observed species richness in this land use, second only to natural forest, could be due to a disproportionate number of moth species that possess larvae feeding on herbaceous, as observed by Summerville and Crist [24] in smaller forests. This aspect should be considered in conservation programs, where special attention should be given to these fragments, making them subject of specific protection measures. In any case, even environments subject to significant anthropic actions and with a reduced or absent tree cover have a non-marginal role in hosting species linked to shrubby or herbaceous plants.

The results of this study indicated that the richness, abundance, and composition of lepidoptera fauna differed among land uses investigated in the Choke Mountains, Ethiopia. Natural forest showed the highest level of biodiversity; other land uses, at different levels and times of sampling, contributed in maintaining biodiversity in areas where undisturbed environments have almost disappeared.

Information on how biodiversity is represented in habitats with different human impacts can give an important contribution in improving the management of the environment and minimising conflicts among local communities, regional government, and wildlife protection Authorities.

Study area

This research was conducted in the Choke Mountains, in the East Gojjam Zone of Amhara Regional State, Northern Ethiopia, between 6^o00’–10^o00’N and 36^o00’– 40^o00’E (Fig. 5). The Choke Mountains are part of the Blue Nile basin range and are located on a plateau covering about 52,139 ha that rises from a block of meadows and valleys and has an elevation ranging from 800 to 4080 m above sea level [25]. During the hottest months (January–May), the temperature reaches about 27°C. The amount of annual rainfall varies from 900 to 1800 mm, with the mean annual rainfall in the period 2010–2019 being 1377 mm. The area experiences unimodal rainfall: from October to February is the dry season whereas from March to September is the wet season. Heavy rain is prominent in June, July and August [26]. The agricultural production systems in the area vary according to the agroecological context and are mainly characterized by subsistent mixed crop-livestock production systems [27].

For the aim of this study, we focused on an area with an altitude ranging from 2200 m to 2900 m above sea level, characterized by the same agroecosystem zonation, Midland Sloping Lands, as defined by Simane et al. [15]. We investigated four land use types characterized by increasing levels of human impact: Forest Fragments (FF), Pasture Land (PL), Crop Fields (CF), and Rural Settlements (RS). From each type, four sampling sites were selected at a distance of at least 300m from each other. In order to have a representative picture of Lepidoptera communities in climax conditions at the same altitudes, we selected three sampling sites in relatively undisturbed Natural Forests (NF) in the territory of the Blue Nile basin: Wof Washa Natural Forest in Ankober (North Shewa Zone), Delima Natural Forest (Michakel district in East Gojjam Zone), and Bradi Natural Forest (Guangua district in Awi Zone). All selected NF sites are classified as Dry Evergreen Afromontane vegetation [28]. The map of the research area and sampling sites is presented in Fig. 1. The locations of sampling sites and the corresponding sampling efforts are reported in Table 3. The altitude and geographical coordinates of sampling sites were measured with the Global Positioning System (GPS), a Garmin GPS III instrument (Garmin Instruments Inc., Olathe, KS, USA).

Table 3

Locations of sampling sites and number of sampling nights
Land use types		Latitude (N)	Longitude (E)	Altitude (m/a.s.l)	Number of sampling sites	Number of sampling nights
Natural Forest (NF)	Wof Washa	9°44'38.79"	39°44'50.30"	2900	1	9
	Delima	10°37'26.24"	37°40'50.60"	2377	1	9
	Bradi	10°51'4.35"	36°37'15.86"	2182	1	9
Forest Fragment (FF)		10°26'28.22"	37°43'47.78"	2600	4	12
Pasture Land (PL)		10°25'19.57"	37°43'15.72"	2539	4	12
Rural settlement (RS)		10°26'32.12"	37°44'54.82"	2467	4	12
Crop Field (CF)		10°27'1.33"	37°44'38.37"	2531	4	12

A brief description of each land use type is reported hereafter. Unless differently indicated, the vegetation description was derived from personal observations.

Bradi Natural Forest (NF)

Bradi natural forest is located in the Guangua District, Awi Zone (Amhara National Regional State). It is part of the Gojjam Floristic Region, western Ethiopian highlands [29]. The area encompasses about 458 hectares of land (Fig. 6a). The forest exhibits a declining trend in its area coverage due to ongoing human-driven land use and land cover changes, mainly caused by expansions in agriculture and settlement. The area is dominated by Rothmania urcelliformis, Vepris dainellii, Rytigynia neglecta, Albizia schimperiana and Croton macrostachyus. Samplings were performed at the center of the core zonation of the forest (Table 3).

Wof Washa Natural Forest (NF)

Wof Washa Natural Forest is located in the Ankober District, North Shewa Zone of Amhara National Regional State (Fig. 6b). The site is perched on the eastern escarpment of the Ethiopian highlands that is regarded as Dry ever green Afromontane Forest and grass land complex. Annual rainfall ranges from 1000 to 1400 mm and cold temperatures are prominent for most of the year. The agroecological zone composition of Wof-Washa forest consists of Weyna Dega (cool sub- humid), Dega (cold and humid) and Wurch (cold and moist) regions [30]. There were widespread human interferences in the Wof Washa Natural Forest due to the increasing demand for construction, firewood and livestock feed. Samplings were performed at 2900 m.a.s.l (Table 3).

Delima Natural Forest (NF)

It is a dry evergreen montane forest in the Michakel district, East Gojjam Zone. The average annual rain ranges from 900mm to 1800 mm and the average annual temperature is between 18ºC and 25ºC. Michakel district is experiencing high levels of environmental degradation, with high rates of deforestation and only 7% of the forest remaining. Characteristic trees and shrubs of the site include Acacia abyssinica Hochst. ex Benth., Croton macrostachyus Hochst. ex Carissa spinarum, Celtis africana, Juniperus procera, Mimusops kummel, Olea europaea subsp. cuspidata, Podocarpus falcatus and Prunus africana (Fig. 6c).

Rural Settlement (RS)

We defined as RS the areas characterized by rural human settlements surrounded by small plots of forest remainings, farmyards, dirt roads bordered by hedges, small eucalyptus plantations, and strips of herbaceous plants, forming a mosaic of heterogeneous plots. Several food and cash crops are grown in the home gardens. Traditional rooted part of plant biodiversity conservation and utilization is found in home gardens. In RS, land use systems involve the management of multipurpose trees and shrubs in strong association with annual and perennial crops. Livestock is raised within the compounds of individual houses, and the whole tree-crop-animal unit is intensively managed by family labour. Such systems are essentially man-made and reflect the wisdom of the traditional culture and ecological knowledge that evolved over the years. The high diversity of plant species in home gardens plays multiple socioeconomic and ecological roles because it is related to the production of food and other products such as firewood, fodders, spices, medicinal plants and ornamentals. (Fig. 6d).

Forest Fragment (FF)

It is a patchy forest remaining around the Melit church, located in the area of the Ennerata district of East Gojjam zone (Fig. 6e). The forest fragments have an extension of 0.5 km², surrounded by farmland, communal grazing lands and villages. In the surroundings of the area, human activities, such as crop cultivation and livestock husbandry, have been most intense for years. Even though the forest coverage of the area is small, because of the proximity of the church, the vegetation of the core site is untouched and is in good condition. The most dominant vegetation in the area includes Acacia abyssinica Hochst, Euphorbia abyssinica Gmel., Rosa abyssincia Lindley., Olea europaea L. subsp. Cuspidate, Croton macrostachyus Del. and Solanum anguivi Lam. The sampling was made at the peak of the plateau where most of the vegetation is found (Table 3).

Pasture Land (PL)

The site is dominated by small to medium-sized perennial bushes (i.e., plants that have persistent woody multiple stems above the ground with shorter height, less than 6–10 m) interspersed with grasses (Fig. 6f). The local community commonly uses this kind of land as communal grazing land. Unlikely, the site is currently used by youth associations as an apiculture training centre as a means of area rehabilitation. The most common plant species include Rosa abyssinica Lindley., Myrsine africana L., Vernonia myriantha Hook.f. and Acacia mearnsii De Wild.

Crop Field (CF)

The area is characterized by a mixed crop-livestock production system, mainly subsistent farming. The dominant crops are tef (Eragrostis tef), maize (Zea mays) and wheat (Triticum aestivum and T. durum) (Fig. 6g). Barley (Hordeum vulgare), potato (Solanum tuberosum), fava beans (Phaseolus vulgaris), chickpeas (Cicer arietinum), oat (Avena sativa), and sorghum (Sorghum bicolor) are also grown to a lesser extent, based on their agro-ecological preferences. Since agriculture activity fully relies on rain, the sowing season for almost all crops is in the rainy season (June - August). Strips of grasses and verges land are very common between croplands.

Lepidoptera Sampling, Preparation And Identification

The majority of moths are night flying and attracted by artificial light sources. For Lepidoptera sampling, we used light traps because of their high effectiveness in attracting and preserving specimens in relatively good condition [31]. In the FF, PL, CF, and RS sites, specimens were collected using battery-operated light traps. The light traps used in this study were equipped with 12 Volt, 15 Ampere batteries, 20W UV led strips (385–400 nm band) with a length of 50 cm, a plastic cover, four transparent vanes and a collection box underneath containing the battery powering the device [32]. Moths, attracted to the light, hit the transparent vanes and then slipped through a plastic funnel into the collection box. When sampling was completed, moths resting on the outside of the trap were also manually collected. A piece of Diethyl ether or Trichloroethylene-dipped tissue was dropped inside the trap to stun specimens. In the NF sites, the collection was made using an illuminated white sheet with the same light source and battery used in the previously described traps. Since light trapping is highly dependent on weather conditions (e.g., temperature, wind and moonlight), sampling activity was restricted to periods without strong moonlight, windy days and heavy rains.

From August 2017 to July 2018, Lepidopteran sample collection was performed once a month. In the FF, PL, CF, and RS sites, traps were placed in sequence in all sites, for four consecutive nights. The traps were active for over 4 hours, from 19:00 to 23:00. Likewise, in the three NF sites, the manual collection was carried out three consecutive nights, over 4 hours per night, from 19:00 to 23:00.

All the collected sample specimens were placed in boxes with cotton and transported to the laboratory, pinned with entomological pins, relaxed in humid chambers and prepared with open wings on spreading boards for taxonomy study.

Comparative morphological methods were used for the identification of sampled material. Morphological characters of adults, like genitalia and wing venation, were used for sorting. When wing patterns alone did not allow the attribution of the specimens to a given species, the genitalia were prepared using the method of Hardwick [33]. The identification of the specimens was done using all available publication reference collections, as far as possible to species level.

We focused the analysis on the collected material belonging to the following 14 families: Yponomeutidae, Limacodidae, Metarbelidae, Tortricidae, Uraniidae, Cossidae, Crambidae, Geometridae, Lasiocampidae, Notodontidae, Nolidae, Saturniidae, Sphingidae, and Erebidae. Based on the actual fauna knowledge [34], the considered families form approximately 55% of the Lepidoptera species known to date in Ethiopia and can therefore be considered as a representative sample of Lepidoptera fauna. These families were selected based on the authors' better knowledge and because of the help provided by some specialists in those families (see Acknowledgements), which facilitate the identification of each specimen at the species level, even in cases where it was not possible to assign a name, probably because belonging to undescribed taxa.

Vegetation Sampling

Plant species were sampled using a standard procedure selecting a square plot of 400 m² (20 m x 20 m) at each of the 19 sampling points (Table 3). The distance between two consecutive plots along a line transect was at least 300 m. Identification of plant specimens was performed both in the field and in the laboratory (for more ambiguous specimens), and it was based on the use of taxonomic keys, descriptions and illustrations in the various volumes of the Flora of Ethiopia and Eritrea [35, 36, 37, 38], comparison with authenticated herbarium specimens, and consulting with the experts of plant taxa.

A list of all sampled vascular plants was prepared, with the presence/absence indication for each of the five land use types.

Biodiversity Analysis

All specimens of Lepidoptera were assigned to an Operative Taxonomic Unit (OUT) at the species level, used for biodiversity analysis. Statistical analyses of the OUTs were conducted using various diversity indices developed to measure and compare biodiversity [39]. Specifically, the following indices were calculated: Shannon’s diversity (H’), Simpson’s evenness, Berger–Parker dominance (BP) and Chao-1 species richness.

The Shannon index (H’) is the most commonly used measure for diversity and it is defined as:

$$H{\prime }=-\sum _{i}\frac{{n}_{i}}{n} ln\frac{{n}_{i}}{n}$$

where ${n}_{i}$ is the number of individuals of taxon i and $n$ is the overall number of found individuals. It varies from 0, for communities with only a single taxon, to high values for communities with many taxa, each with few individuals. Therefore, high values of H’ are associated with a greater degree of biodiversity. Usually, the Shannon’s index value varies from 1.5 to 3.5 and rarely exceeds 4.5 [40].

Simpson diversity index is a measure of the evenness of a community quantifying how close each species is in numbers to others in the same environment. It ranges from 0 to 1, with evenness increasing when approaching 1. It is calculated as 1-D, with D (dominance) calculated as:

$$D= \frac{\sum _{i}{n}_{i} ({n}_{i}-1)}{n(n-1)} .$$

Berger-Parker index (BP) measures the proportion of species in a sample and usually it ranges between 0 and 1. According to Buschini and Woiski [41], the index can also be calculated as a percentage:

$${BP\%}_{i}=\frac{abundance of the i-th species}{total abundance}*100$$

If ${BP\%}_{i}$ is higher than 5%, the species was termed a dominant species; if ${BP\%}_{i}$ ranges between 2.5% and 5%, the species was termed an accessory species or species of intermediate abundance, the species with ${BP\%}_{i}$ lower than 2.5% was termed as an incidental species.

Chao_1 species richness is an estimate of total species richness, where rare species (i.e., species occurring with one individual: singleton, or with two individuals: doubleton) were noted. It is calculated as:

$$Chao\_1 = S +\frac{F1 (F1 - 1)}{2 (F2 +1)}$$

where S is the number of sampled taxa, F1 is the number of singleton species and F2 is the number of doubleton species.

Biodiversity indices were calculated for each land use type (NF, PL, CF, FF, RS). Confidence intervals for all these indices were computed with a bootstrap procedure. We extracted 9999 random samples from the original sample, each with the same total number of individuals as in the original sample. For each individual in the random sample, the taxon is chosen with probabilities proportional to the original abundances. A 95% confidence interval is then calculated [42].

Biodiversity and cluster analyses were carried out using the software Past version 3.25 [42].

A regression analysis was carried out to assess the linear relationship between plants and Lepidoptera. For each sampling site, the number of lepidopteran species, both sampled or estimated by the Chao-1 index, was compared with the observed number of plant species.

Crop Field

Pasture Land

Forest Fragment

Rural Settlement

Natural Forest

Acknowledgements

We want to thank Ethiopian Biodiversity Institute (Ethiopia) for partial sponsorship during the study leave of the first Author and dr. Getnet Atenafu (Debre Markos University, Ethiopia) for providing logistic support for the field work. The authors are grateful to the specialists who provided helps in verify part of the collected material, namely: Dr. Axel Hausmann (Bavarian State Collection of Zoology, Munich, Germany), Prof. David Agassiz (Natural History Museum of London, UK), Dr. Graziano Bassi (Natural History Museum of Geneva), Dr. Herman H. Hacker (Bamberg, Germany) and Prof. Pasquale Trematerra (University of Molise, Campobasso, Italy) for the families Geometridae, Ypomoneutidae, Crambidae, Erebidae/Nolidae and Tortricidae, respectively.

Authors contributions

All authors contributed to the study conception and design; TT carried out the field work; TT and AS studied the collected material, analyzed data and prepared the first draft of the manuscript; AS and GG commented on previous versions of the manuscript and have substantively revised it. All authors read and approved the final manuscript.

Funding

The present work was carried out under the International PhD program in Agricultural Technology and Biotechnology of the University of Molise (Italy), with a scholarship given to the first author. The fieldwork was partially funded by the Belmont Forum Collaborative Research Project NILE-NEXUS (Grant number: 1624335).

Availability of data and materials

The datasets analyzed during the current study are available at https://doi.org/10.5281/zenodo.7064866

Ethics approval and consent to participate

We required no specific permits since collection localities correspond to neither private properties nor protected areas. Also, the field studies did not involve endangered or protected species.

Consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

Author details

¹Animal Biodiversity Directorate, Ethiopian Biodiversity Institute, Comoros street, Addis Ababa, Ethiopia; ²Department of Agriculture, Environment and Food Sciences, University of Molise, Campobasso, Italy; ³Department of Civil, Environmental, Architectural Engineering and Mathematics, University of Brescia, Italy

Foley F, Fries D, Asner P, Barford C, Bonan G, Carpenter R, et al. Global consequences of land use. Science 2005, 309:570–574.
Millennium Ecosystem Assessment Ecosystems and human well-being: biodiversity synthesis. Washington: World Resources Institute, 2005. 137pp.
Fugère V, Kasangaki A, Chapman J. Land use changes in an afrotropical biodiversity hotspot affect stream alpha and beta diversity. Ecosphere 2016,7(6):1-18.
Chivian E, Bernstein A. How our health depends on biodiversity. https://www.bu.edu/sph/files/2012/12/Chivian_and_Bernstein_2010_How_our_Health_Depends_on_Biodiversity.pdf Accessed on 8 Sep 2022
Nichols E, Larsen T, Spector S, Davis AL, Escobar F, Favila M, Vulinec K. Global dung beetle response to tropical forest modification and fragmentation: A quantitative literature review and meta-analysis. Biol Conserv 2007,137:1-19.
An S, Choi W. Forest moth assemblages as indicators of biodiversity and environmental quality in a temperate deciduous forest. Eur J Entomol 2013;110(3):509–517.
Spellerberg I. Monitoring ecological change. Cambridge University Press, 2005.
Wilson EO. The Future of Life. New York: Vintage Books, 2002.
Murphy SM, Richards LA, Wimp GM. Editorial: Arthropod interactions and responses to disturbance in a changing world. Front Ecol Evol 2020,8(93), doi: 10.3389/fevo.2020.00093
Hill JK, Hamer KC. Determining impacts of habitat modification on diversity of tropical forest fauna: the importance of spatial scale. J App Ecol 2004,41:744-754.
Munyuli MB. Butterfly diversity from farmlands of central Uganda. Psyche 2012,481509:1-23.
Erhardt A, Thomas JA. Lepidoptera as indicators of changes in the semi-natural grasslands of lowland and upland Europe. In: Collins NM, Thomas JA, editors. The conservation of insects and their habitats. London: Academic Press; 1991. p.213–237.
Fox R. The decline of moths in Great Britain: a review of possible causes. Insect Conserv Diver 2013,6:5–19.
Teferi E, Uhlenbrook S, Bewket W, Wenninger J, Simane B. The use of remote sensing to quantify wetland loss in the Choke Mountain range, upper Blue Nile basin, Ethiopia. Hydrol Earth Syst Sci 2010,14:2415-2428.
Simane B., Zaitchik BF, Ozdogan M. Agroecosystem analysis of the Choke Mountain watersheds, Ethiopia. Sustainability 2013,5:592-616.
Pimm S.L, Jenkins CN, Abell R, Brooks TM, Gittleman JL, Joppa LN, et al. The biodiversity of species and their rates of extinction, distribution, and protection. Science, 2014,344(6187):1246752.
Gaston KJ, Fuller RA. Biodiversity and extinction: losing the common and the widespread. Progr Phys Geogr 2007,31(2):213-225.
Chapman CA, Abernathy K, Chapman LJ, Downs C, Effiom EO, Gogarten JF, et al. The future of sub-Saharan Africa’s biodiversity in the face of climate and societal change. Front Ecol Evol 2022,10:790552.
Simane B. Building community resilience to climate change. Lessons from Choke Mountain Agro-ecosystems, Abay/the Blue Nile Highlands. Ethiopia: Addis Ababa University Press, 2016. pp. 302.
Siemann E, Tilman D, Haarstad J, Ritchie M. Experimental tests of the dependence of arthropod diversity on plant diversity. Am Nat 1998,152(5):738-750.
Axmacher JC, Brehm G, Hemp A, Tünte H, Lyaruu HVM, Müller-Hohenstein K, Fiedler K. Determinants of diversity in afrotropical herbivorous insects (Lepidoptera: Geometridae): plant diversity, vegetation structure or abiotic factors? J Biogeogr 2009,36:337–349.
Kittle AM, Kumara PHSC, Pathirathna DG, Sanjeewani HKN, Seneviratne HTJ, Watson AC. A comparison of floral and faunal diversity between two small, disturbed forest patches in Sri Lanka’s central highlands. Wildlanka 2016,4(3):134–141.
Pulliam HR. Sources, sinks, and population regulation. Am Nat 1988,132:652–661.
Summerville KS, Crist TO. Contrasting effects of habitat quantity and quality on moth communities in fragmented landscapes. Ecography 2004,27:3-12.
Fetene A, Alem D, Mamo Y. Effects of Land use and Land Cover changes on the extent and Distribution of Afroalpine Vegetation of Northern Western Ethiopia: the case of Choke Mountains. Research J Environ Sci 2014;8(1):17-28.
Western Amhara Meteorological Service Center , 2020
Azmeraw A. Nature and determinants of participation in community based participatory watershed management in Choke Mountain: the case of Chemoga Watershed, East Gojjam Ethiopia, Ethiopia: Addis Ababa University Press, 2010.
Getachew T. Estimating willingness to pay for forest ecosystem conservation. The case of Wof-Washa Forest, North Shewa Zone, Amhara National Regional State, Ethiopia. J Resour Dev Man 2018,46:46-61.
Friis I, Demissew S, Breugel PV. Atlas of the Potential Vegetation of Ethiopia. Addis Ababa: The Royal Danish Academy of Sciences and Letters (Natural habitats), 2010. 307 pp.
Ministry of Agriculture Federal Democratic Republic of Ethiopia. Ethiopia’s Agricultural Sector Policy and Investment Framework 2010-2020 Draft Final Report, 2010.
Axmacher JC, Fiedler K. Manual versus automatic moth sampling at equal light sources- a comparison of catches from Mt. Kilimanjaro. J Lepidopterists’Society 2004,58:196–202.
Infusino M, Brehm G, Di Marco C, Scalercio S. Assessing the efficiency of UV LEDs as light sources for sampling the diversity of macro-moths (Lepidoptera). Eur J Entomol 2017,114: 25-33.
Hardwick DF. Preparation of slide mounts of lepidopterous genitalia. Can Entomol 1950,82:231–235.
Tujuba TF, Sciarretta A, Hausmann A, Abate GA. 2019 – Lepidopteran biodiversity of Ethiopia: current knowledge and future perspectives. Zookeys, 882: 87-125.
Hedberg I, Edwards S. Flora of Ethiopia (Vol. 3): Pittosporaceae to Araliaceae. Addis Ababa and Uppsala: The National Herbarium, 1989. Pp. 732.
Hedberg I, Kelbessa E, Edwards S, Demissew S, Persson E. Flora of Ethiopia and Eritrea (Vol. 5): Gentianaceae to Cyclocheilaceae. Addis Ababa and Uppsala: The National Herbarium, 2006. pp. 690.
Hedberg I, Friis I, Persson E. Flora of Ethiopia and Eritrea (vol. 1): Lycopodiaceae to Pinaceae. Addis Ababa and Uppsala, The National Herbarium, 2009a. Pp. 305.
Hedberg I, Friis I, Persson E. Flora of Ethiopia and Eritrea. General Part and Index to Volumes 1-7 (vol. 8). Addis Ababa and Uppsala: The National Herbarium, 2009b. Pp. 331.
Magurran AE. Ecological diversity and its measurement. Princeton: Princeton University Press, 1988. Pp. 185.
Kent M, Coker P. Vegetation Description and Analysis: A practical approach. London: Belhaven Press, 1992. pp. 263.
Buschini ML, Woiski TD. Alpha-beta diversity in trap-nesting wasps (Hymenoptera: Aculeata) in Southern Brazil. Acta Zool 2008,89(4):351–358.
Hammer Ø. PAST Paleontological Statistics. Reference Manual. Version 3.25. 2019. https://folk.uio.no/ohammer/past/past3manual.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Lepidoptera as a tool for human disturbance assessment on biodiversity in Choke Mountains, Ethiopia

Status:

Version 1

Abstract

Background:

Conclusions:

Figures

Background

Results

Discussion

Conclusions

Methods

Study area

Bradi Natural Forest (NF)

Wof Washa Natural Forest (NF)

Delima Natural Forest (NF)

Rural Settlement (RS)

Forest Fragment (FF)

Pasture Land (PL)

Crop Field (CF)

Lepidoptera Sampling, Preparation And Identification

Vegetation Sampling

Biodiversity Analysis

Abbreviations

Declarations

References

Status:

Version 1