Trends of Land Cover Change and Conservation Challenges of Grevy’s Zebra ( Equus grevyi , Oustalet 1882) in Hallaydeghe Asebot Proposed National Park, Southeast Ethiopia

,


Introduction
Globally, there has been a decline in biodiversity over the last four decades, attributed in large parts to habitat fragmentation and land conversion [1][2][3].Protected areas (PAs) are areas of biodiversity conservation and shaped by the land uses, species, and ecological process in the surrounding landscape and should not be viewed in isolation [4,5].External infuences can decrease the efective size of a PA, limiting their ability to protect biodiversity and ecosystem functions [6][7][8].Specifcally, there is a link between increases in anthropogenic activities such as land cover changes owing to changes in agricultural land use, urbanization and illegal extraction within and around PAs, and species extinction which may exacerbate impacts on PAs [9][10][11].Tese have afected functionality of the PAs.
Loss of functionality of PAs from surrounding land-use modifcation is a particularly daunting problem in developing nations.However, the resources in and around PAs are more critical to people living adjacent to PAs in developing nations because their livelihoods are often more directly dependent on the land [12].Natural disturbances, ecological processes, and human activity cause the landscapes of protected areas (National Parks and watersheds) to change over time [13].Terefore, to ensure the efectiveness of PAs in the developing world, it is necessary to understand changes driven by the surrounding landscape.
Protected areas in the majority of Eastern African countries frequently encountered land-use changes as a result of growing human and livestock populations [14].Te relationship between people and the environment where they live in is the main focus of land use and land cover change (LULCC) concerns [15].
In Ethiopia, protected areas have faced a range of conservation challenges.Te root causes of biodiversity conservation challenges in Ethiopia are associated with lack of adequate capacity, commitment, organizational set-up, and lack of monitoring of the implementation strategy on the status and trends of conservation challenges [16].Te land cover changes in and around the protected areas of the country are common as a result of anthropogenic impact such as increased heavy grazing, fre, agricultural expansion, and poor application of land-use policy [17].
Hallaydeghe Asebot Proposed National Park (HAPNP) was designated as a Wildlife Reserve with a primary objective to conserve, manage, and propagate wildlife within it.Te area was specifcally selected to serve as a corridor between Awash National Park and the surrounding plains so that the wildlife population of Awash National Park would be protected through freedom of migration [18].Te area is home to a number of wildlife species including Grevy's zebra, Soemmering's gazelle (Gazella soemmeringi), Beisa oryx (Oryx gazella), Gerenuk (Litocranius walleri), Salt's dikdik (Madoqua saltiana), African wolf (Canis lupaster), Spotted hyena (Crocuta crocuta), Aardwolf (Proteles cristatus), and Ostrich (Struthio camelus).It is also a remnant grazing land for the Afar and Issa pastoral communities, in which livestock numbers are a social indication of the owner's wealth status [19].Wildlife resources including Grevy's zebra that live in HAPNP are facing diferent forms of conservation challenges elsewhere that stem from expansion of invasive species, heavy grazing, and poaching and habitat degradation [20].
Consequently, over the last 30-40 years, Grevy's zebra has experienced dramatic reductions in range and numbers and are now found only in northern Kenya and in small pockets in Ethiopia.A recent study conducted by Tolera et al. [21] indicated as there are no more than 75 individuals of Grevy's zebra in the area.Grevy's zebra in HAPNP is mostly isolated and has unique population [22].Te major challenge facing conservationists is to identify and mitigate the causes of the decline in Grevy's zebra's range and population size.As already mentioned, the anticipated justifcations for Grevy's zebra population and range decline are habitat degradation, poaching, competition with livestock for critical resources, expansion of invasive species, and drought [23][24][25][26][27]. Te same factors might have afected the range and population size of Gravy's zebra in HAPNP of Ethiopia.
Tus, this study investigated the land cover change and conservation challenges for Grevy's zebra in HAPNP.As information relating to the land cover change and conservation challenges to these species is limited, assessing trends in land cover change in the stated protected area is important.So, assessing of changes in land cover using remote sensing and satellite imagery analysis and predicting is very important and could be useful to document land converted to other land-use types afecting the species.Remote sensing techniques provide an efective means of monitoring and measuring land cover change over spatial and temporal extents and may provide practitioners with insights into future land-use change processes [28].Satellite and aerial imagery analyses have versatile applications that allow us to measure spatial and temporal changes in and around PAs [29,30].It also uses globally available satellite imagery, making it a widely accessible research methodology for scientists in both developed and developing nations.
In a similar manner, we expected that anthropogenic activities would raise conservation challenges in the study area.In order to inform existing conservation eforts and ofer a foundation for reducing Grevy's zebra conservation issues in HAPNP, it was anticipated that the information gathered from this study would be able to explain the land cover change in the protected area and the conservation challenges in their natural range.

Te Study Area Description.
Hallaydeghe Asebot Proposed National Park is located about 280 km away from Addis Ababa in the Ethiopian Rift Valley between latitude of 8 °92′ and 9 °48′ and longitude 40 °25′ and 40 °63′.Its former name was Alledeghi Wildlife Reserve, and the primary objective was to "conserve, manage, and propagate the wildlife within it."HAPNP has a semiarid climatic condition.Te site is dominated by a large alluvial plain with mountains rising along the eastern border of the proposed park [31].Te site was selected to serve as corridors between Awash National Park and the surrounding plains so that through migration, the wildlife population of Awash National Park would be protected [18].Currently, there are 13 kebeles belonging to three Woredas, namely, Amibara, Meiso, and Haruka in which the protected area occurs (Figure 1).Te community surrounding this protected area practice pastoral mode of life.Te area is home to a number of wildlife species including Grevy's zebra (Equus grevyi), lion (Panthera leo), Soemmering's gazelle (Gazella soemmeringi), Beisa oryx (Oryx gazella), Gerenuk (Litocranius walleri), Salt's dikdik (Madoqua saltiana), African wolf (Canis lupaster), Spotted hyena (Crocuta crocuta), Aardwolf (Proteles cristatus), and Ostrich (Struthio camelus) [32].
Te major vegetation types in and around the HAPNP include grassland, bushland, shrubland, wooded grassland, riverine forest, and highland forest [33].Chrysopogon plumulosus and Sporobolus iocladus contribute a relatively substantial percentage of the herbaceous vegetation on the plains [33].Te southern, northern, and western edges of the protected area are bush-grassland or shrubland, with Senegalia senegal being the dominant species here, as well as in some parts of the grassland.Te woody plant species in the plains of this study site include Prosopis julifora, Senegalia For 1990 and 2006 LCC map, GTPs were collected from the false color composite of 1990 and 2006 satellite images with the help of elders and knowledgeable people [39].GTPs were taken by well-trained stafs of the proposed park.A minimum of 50 points for each major land cover category were collected following [40].Te land cover of HAPNP was categorized into fve major diferent classes: grassland, woodland, bushland, forest, and settlement.Te vegetation description was based on the classifcation provided by [41].

Conservation Challenges in the Data Collection Method
2.3.1.General Approach.Te list of species conservation challenges was frst prepared in English by researchers and translated into Afarigna language by native language speakers to minimize confusion.Moreover, during data collection, the native language speakers read the developed questions for the respondents and probed them for the response they provided to minimize error made during response provision.Te list of Grevy's zebra conservation challenges presented for respondents were habitat degradation, poaching, drought, disease, predation, roadside killing, weak law enforcement, and unintegrated development to research participants.Based on the list provided to them, they asked to rank from top to down.Te interview average length time had been 90 minutes.

Sampling Techniques and Sample Size Determination.
Te preliminary survey was made to estimate the distance of the surrounding kebeles from head quarter of the park.Based on the preliminary survey, kebeles were classifed into three domains (less than 5 km, 6-10 km, and >10 km).After International Journal of Zoology stratifying kebeles into three strata based on their distance and accessibility from headquarter of the park, one kebele was selected using random sampling techniques from each strata and a total of three kebeles were considered.Ten, household sample sizes were calculated using the Yamane [42] formula.Te formula is expressed as where n is the sample size, N is the population size/sampling frame, and e is the error of prediction which was 0.05 (95%).
Te selected kebeles had a total of 996 households (N).Ten, the total respondents were 285.Te distribution of sample respondents to each kebele was made based on the proportion of the population size of the selected kebele.Finally, the study household was selected from each selected kebele by using systematic random sampling.Te frst household was randomly selected, and thereafter, every K th household were taken until the desired sample size is reached (where K � N/n).Tose family member individuals who had a better understanding about study issues were purposively selected.

In-Depth Interviews.
In-depth interviews were carried out using structured and semistructured questions.In doing so, the participants for the in-depth interviews were selected purposively based on the responsibilities they have, experience, and relevance to issues under study.Accordingly, from nearby Woredas animal science, plant science, and natural resource management experts and police offcers had participated.Chief Warden and senior staf members of HAPNP had also participated.Te issues discussed were the current conservation challenges of Grevy's zebra and possible solutions for sustainable conservation of species.Te protected area feld personnel who were interviewed through the method were considered knowledgeable in view of their involvement in protected area management over time.

Observation Checklist.
A digital camera was used during fled observation to take the photos of conservation challenges' efects.Moreover, it was undertaken to augment the reliability of information collected through diferent sources.To undertake feld observation checklist, a walk along transect line was conducted.A total of 15 transect lines of varying lengths were established, and they were purposefully placed using GPS (Figure 1).At least 1500 meters separated the adjacent transects.Te ends of each transect were not less than 1000 m from the habitat margin.Transects were laid north to south direction in order to reduce impact of the sunlight.While the longest transect was 17 km, the shortest transect has 5 km length.Te main road that connects Ethiopia with Djibouti crosses the HAPNP.So, the data on impact of road side killing were collected via a mix of roadside walk and drive [43,44].Te length of the road in that data collection held was 23.5 km.Te data were collected by driving 4WD vehicles at a speed of 20-30 km/h, once a day.Road kill locations were recorded using a GPS (Garmin GPS72H).Once recorded, road kill was removed to avoid recounts.In addition, desk review was held to make stronger the fndings of the study.

Data Analysis 2.6.1. Data Analysis for LCC.
To generate land cover change maps, the satellite image for each period was classifed using supervised classifcation.Te methodology adopted for land use/cover change analysis is summarized in Figure 2. Overall, it involves radiometric normalization, land cover classifcation, accuracy assessment, and postclassifcation analysis using ArcGIS 10.8 and QGIS 3.26.3.Radiometric corrections of satellite images were carried out by converting DN values into radiance values using derived calibration coefcients.Radiometric correction is a technique for improving the brightness magnitude of a satellite picture for better visibility and analysis.By adjusting the bias and gain levels for each band, radiometric correction may be used to translate digital numbers (DNs) into Top of Atmosphere (TOA) radiance measurements [45].

LU/LC types
Working defnition/description Forest Landscape dominated by land spanning at least 0.5 ha covered by trees, dense woodlands, and dense bamboo, attaining a height of at least 2 m and a canopy cover of at least 20% or trees with the potential to reach these thresholds which could be natural/manmade [35] Bushland/shrubland Areas covered by densely growing woody vegetation of shrubby habit, generally more than 0.5 m and less than 7 m in height at maturity and without a defnite crown [35] Woodland/savannah woodland An area consists of trees that is branched, Acacia-Commiphora, deciduous and ranges from 8 to 20 m in height, crowns may touch grasses, and herbs are present [36] Openland/grass Land covered by grasses and other herbs, either without woody plants [37] Settlement Permanent residential areas (mainly urban built-ups), refugee camps, and rural villages which are clearly visible and identifed on satellite image and having extent of at least one hectare in the study area [38] 4 International Journal of Zoology Following the supervised classifcation of imagery, the postclassifcation change detection algorithm was performed to determine changes in land cover [46,47].Tis postclassifcation approach provides statistical evidence on how land cover has changed and is used to calculate and produce map of land cover changes over time.According to Han et al. [48], the most common methods for the detection of land change are image overlay, principal component analysis, change vector analysis, image rationing, and the normalized diference vegetation index (NDVI).In this study, classifcation comparisons of land cover statistics had been used.Each land cover type's covered areas over time have been compared.Ten, the directions of the changes (positive or negative) in each land cover type had been determined.

Percentage of land use land cover change �
Area final year − Area initial year Area initial year x100. ( Te accuracy of the land cover classifcation was assessed using a confusion matrix to compare the 1990, 2006, and 2021 classifcation results to the ground observations.For each land cover class, a contingency matrix was generated and the overall accuracy, the Kappa statistic, and the producer and user accuracy for each class were calculated [49].Te total accuracy was calculated by dividing the number of correctly classifed elements (i.e., the sum of the diagonal elements in an error matrix) by the total number of pixels included in the evaluation process.Te Kappa statistic is an alternative measure of the classifcation accuracy that subtracts the efect of random accuracy, and it quantifes how much better a particular classifcation is compared to a random classifcation.International Journal of Zoology 2.7.Data Analysis for Conservation Challenges.Content analysis was used to analysis the conservation challenges.Furthermore, statistical package software for social science (SPSS) version 25.0 was used to compare and rank the conservation challenges of the species.2).

3
In 1990, the large areas were covered by woodland and grassland, respectively, while settlement area coverage was very small (Figure 3).Te details of land converted from one type to the other during 1990-2006 is provided in (Table 3) that is found under appendex section.
Between 2006 and 2021, period bushland increased by 35,229.32ha (19.23%).In contrast, grassland decreased by 8.12% (Table 4).Te details of land converted from one type to the other during 2006-2021 is provided in (Table 5) that is found under appendex section.
During 2006, the large area coverage was woodland and bushland, respectively, while settlement area coverage was minimal (Figure 4).
Te land cover of HAPNP in 2021 was forest land 7,534.61ha, shrubland 89,307.96ha, grassland 37,458.72 ha, woodland 48,149.44ha, and settlement 749.23 ha (Figure 5).Te details of land converted from one type to the other during 1990-2021 is provided in (Table 6) that is found under appendex section.
Te conversion of woodland to shrubland 9,213.31ha was seen to be the largest land cover change that occurred between the years of 1990-2006 in HAPNP.Te smallest change was from woodland to settlement 0.5 ha.Between 2006 and 2021, substantial land cover change from grassland to shrubland 14,324.37ha had occurred.During the same period, woodland to forest 1.34 ha was the smallest land cover change that occurred.In general, from 1990 to 2021, the signifcant land cover change was undergone from grassland to shrub/bushland 16,790.32ha.
Te accuracy of land cover map of the 1990, 2006, and 2021 was assessed using GTPs collected during feld mission and high resolution images available at Google Earth Engine as a supplementary tool.For the 2021 land cover map, aerial photographs taken in 2021 and detailed ground survey conducted in 2021/22 were used as reference.Overall, land use/cover classifcation accuracy levels for the three dates range from 89 to 94 percent with Kappa statistics ranging from 0.84 to 0.91 (Table 7).Tese accuracy levels satisfy the minimal accuracy required by [50] for satellite-derived land use/cover maps; hence, they were sufcient for the analysis of the research area.

Conservation Challenges of Grevy's Zebra/Equus grevyi/ 3.2.1. Sociodemographic Information of Respondents.
When we look at the respondent's profle, 90% of the respondents from the society who participated in ranking of Grevy's zebra conservation challenges were men and 69% of the respondents lack formal education and are illiterate.Out of the total respondents, 41.3% were within the age bracket of 30-40.Respondents in this area are agropastoralists, with goats and sheep being the most common livestock types.

Grevy's Zebra Conservation Challenges.
Based on respondent's survey, feld observation checklist, and indepth interviews with staf members and other relevant stakeholders, the conservation challenges of Grevy's zebra were prioritized.Accordingly, habitat degradation was identifed as a top conservation challenge for Grevy's zebra.Unintegrated development and week law enforcement, respectively, were identifed as the second and third conservation challenges for Grevy's zebra in HAPNP (Table 8).

Grevy's Zebra Conservation Challenges and Teir Drivers in HAPNP
Habitat degradation: the expansion of invasive plant species, Ruthenium Parthenium hysterophorus, Abutilon fgarianum, and Mesquite Prosopis julifora, was in and around the proposed park.Furthermore, exotic plant species include Eucalyptus Viminal's, Eucalyptus citriodora, Eucalyptus globules, Eucalyptus saligna, Euclea racemosa, Grevillea robusta, Jacaranda mimosifolia, Jatropha curcas, Leucaena leucocephala, Schinus molle, and Senna didymobotrya which are another pressing problem.Moreover, the practice of heavy grazing by pastoralist livestock, human-induced fre, and the local communities' harvesting of park resources for diferent purposes were considered as the main causes for Grevy's zebra habitat degradation.Moreover, there are diminishing sizes of grassland habitat due to land cover change.Most of the corridors through species crosses from one side to the other side are disconnected.Poaching: although not confrmed during this study period, the information gained through key informant interviews and other relevant people indicates that Grevy's zebra is illegally hunted by local communities for medicinal purpose.Te local communities believe that the foot of Grevy's zebra is used to heal from leprosy diseases.Drought: HAPNP is found in a harsh environment where there is not enough rainfall and prolonged dry season aggravated by climate change impacts.Diseases: although the local communities practice the pastoral mode of lifestyle, the culture of vaccinating domestic livestock is very poor.Grevy's zebra and 6 International Journal of Zoology   International Journal of Zoology domestic cattle graze together from Hallaydeghe grassland plain.From the respondent's views and feld observations, it was possible to understand that the condition facilitated the zoonotic disease transmission.Predation: in HAPNP, there are large carnivores like lion and hyena that can prey on Grevy's zebra.In contrast, the species move to water points at night when the site is free from human activities.Te condition facilitates the predation process over species.
Road side killing: the main road that connects Ethiopia with Djibouti crosses through HAPNP.Despite the presence of many vehicles that use the road, there is no speed breaker, not enough sign posts, and awareness raising posters for drivers.In views of respondents and the researcher, feld observations made roadside killing to be considered as one of the conservation threats for Grevy' zebra.Weak law enforcement: there was severe shortage of manpower, especially scouts, who can enforce the rule  International Journal of Zoology of law.Te other pressing problem observed was lack of integration between wildlife stakeholders and inadequate awareness made for local communities regarding Grevy's zebra conservation.Unintegrated development: the urbanization is expanded by disconnecting the corridors that wildlife uses.Investment expansion that did not align with wildlife conservation is undergoing.Tese investments disconnected Grevy's corridors.

Discussion
Te results of the classifcation produced were acceptable accuracies (Table 7) and are generally compatible with prior studies involving Landsat image classifcation [51].We identifed various land cover patterns in HAPNP after analyzing the change in land cover between 1990 and 2021.Tese patterns are likely to afect how well the protected area functions.In 1990, when the area was a wildlife reserve, the land cover of the area was 13,527.[52].Tis declining of grassland cover but increasing of bushland coverage observed was attributed to the efects of intensity of livestock grazing.Increased grazing intensity causes reduced grass biomass [53][54][55][56] that, in turn, reduces fre intensity, that benefts woody plants to regenerate, thus facilitating the shift from grassland to bushland.Tis result is in agreement with the previous study reports [57,58].In addition to the growth in human and livestock populations, the conversion of grazing lands to cultivation areas outside the HAPNP has caused grazing land scarcity.Tis, in turn, can force people to enter the park boundaries for livestock grazing and other forms of resource use.
Our result confrms with the previous study of the authors in [51] that LCC from forest to shrub conversion was the main form of land cover transition observed in the Kasungu National Park of Malawi.In contrast to the current fndings, the study was conducted by the authors in [59] in Awash National Park and the grassland was expanded by 14.2% between 1972 and 1986 as well as by 10.5% of the study area during the entire study period .Te land cover change observed in HAPNP was mostly facilitated by expansion of exotic and invassive plant species and this study confrms with the fndings of [9] where nonnative or alien species posed a signifcant threat to protected areas by their both direct and indirect impacts to native species.Te posed impacts had efects on broader scale ecological patterns and then contribute for habitat degradation.
Habitat degradation in HAPNP has become the top conservation challenges for Grevy's zebra.One of the main reasons for habitat degradation in the area is livestock overgrazing by nomads (the number of livestock is more than the capacity and potential of pastures), which has decreased high-density pasture volume and increased lowdensity pasture in this area.Tese fndings have also been confrmed with previous studies, and overgrazing was reported as the most important reason for habitat degradation [60,61].Furthermore, our study goes in line with the fndings of the study of Kenya's Grevy's zebra Technical Committee assessed and ranked habitat degradation frst rank as conservation challenges [62,63].Another conservation challenge for the species is the uncontrolled movements of local community for harvesting of grasses, frewood collection, and keeping livestock in this area.Furthermore, there is continual expansion of temporary and permanent settlements in and around the park for the need of animal fodder and water sources.Insufcient monitoring is one main reason for the uncontrolled entry of nomads and livestock into the area, and they have many negative impacts on habitat degradation.Tese fndings have also been confrmed in literature reviews [64,65].
Te other conservation challenge of the species in HAPNP is the practices of unintegrated development around the proposed park.Accordingly, in the HAPNP, the plain of the study area and surroundings have been extensively destroyed due to multiple factors, particularly the interference of diferent institutions with the management of the PA through the development of physical and economic activities.Trough overwhelming infuence, different surrounding parts of this PA have been assigned to various organizations or investors for fnancial exploitation, and they could quickly destroy the ecosystem and habitats of the area.
Moreover, another important issue increasing the conservation challenges for the species in this area is the growth of human activities, particularly urban development and road.
Te inability of the protected area to monitor and manage the area due to (staf shortage, limited skill to utilize modern technology to track Grevy's zebra ecological trait, and severe fnancial shortage), along with the infuence of some governmental stakeholders, has led to extensive human activities in the area.Tese fndings were confrmed by the previous study [66].
In addition, the lack of cooperation among diferent organizations and the protected area staf in protecting this area has caused extensive conservation challenges.Other studies have confrmed these fndings [67,68].
According to the current study report, there is no signifcant evidence of predation on Grevy's zebra in Hallaydeghe Asebot Proposed National Park.Possibly, the presence of numerous large and medium sized mammals in the proposed park reduced the rate of predation over Grevy's zebra.Te fndings of this study difer from the study conducted in Kenya that revealed predation as a potential limiting factor in the growth of the population of Grevy's zebra within the range of Grevy's zebra [69].Lions and possibly hyenas are a major threat to the growth of Grevy's population on Lewa Wildlife Conservancy [70].

Conclusion and Recommendations
Te study is important because it has revealed trends in land cover changes in Hallaydeghe Asebot Proposed National Park, one of the habitats of the remaining population of Grevy's zebra in the World.Tere has been reduction in grassland habitat on which the zebra thrives and produce young but increase in bushland, settlements, and woodland coverage from 1990 to 2006 indicating loss of critical habitat of the species.A similar situation has been shown by the study that is decline in grassland habitat being encroached by bushland and other land-use types that has also been increased from 2006 to 2021 possibly leading to the decline in the abundance of zebra in the protected area.Tese have been compounded by the fact that habitat degradation is the main cause of change in land cover in the area.Furthermore, it has been revealed that habitat degradation and unintegrated developments have been the major challenges to the survival of Grevy's zebra in and around the HAPNP.Generally, the results demonstrated that there are signifcant conservation challenges posed on the ecosystem components, including key grazing plain of HAPNP.Terefore, urgent conservation eforts that could reduce land use land cover changes are needed.In addition, restoration and International Journal of Zoology rehabilitation of the grassland habitat for the survival of the ruminant Grevy's zebra in the protected area are highly required.Monitoring and use of remotely sensed data could allow for better decision-making when developing policies and practices to mitigate the impacts of land use change in HAPNP.Furthermore, the conservation challenges identifed need attention from wildlife managers and other stakeholders.
User's accuracy � Number of Correctly Classified Pixels in each Category Total Number of Reference Pixels in that Category (The Row Total) x100, Producer accuracy � Number of Correctly Classified Pixels in each Category Total Number of Reference Pixels in that Category (The Column Total) x100, Total (overall) accuracy � Total Number of Correctly Classified Pixels (Diagonal) Total Number of Reference Pixels x100, Kappa Coefficient (T) � (TSxTCS) − (Column Total x Row Total) TS 2 − (Column Total − Row Total) .

Figure 2 :
Figure2: Study approach adopted for the analysis of land use/cover changes of HAPNP.

Table 1 :
Defnition of land cover category across its defnition.

Table 2 :
Area and proportion of LCC in HAPNP between 1990 and2006.

Table 4 :
Te LCC of HAPNP between the years of 2006-2021.