GIS-based ecotourism potentiality mapping in the East Hararghe Zone, Ethiopia

Abstract

The East Hararghe Zone (EHZ) is one of the eastern Ethiopian zones most endowed with diverse landscapes and abundant resources to promote and use for ecotourism development. Potential ecotourism sites have, however, hardly ever been explored so far. The objective of this study was to model and identify potential ecotourism sites by combining Geographic Information System (GIS)-based Multi-Criteria Decision-Making (MCDM) and Analytical Hierarchy Process (AHP) methods. Six criteria, including landscape and naturalness, wildlife, topography, accessibility, geology, and climate, were established based on experts' preferences and literature, and the thematic factors for suitability modeling were derived from freely accessible satellite imagery and existing geospatial data and combined using a weighted linear combination (WLC) method. The results reveal that about 26.19% were highly suitable, 35.34% were moderately suitable, 25.28% were marginally suitable, and 13.17% were not suitable. Most areas with high to marginal suitability were found in the southeast, southwest, and uplands in the northern part, while the southernmost extent had the highest proportion of areas unsuitable for ecotourism development. The area under the curve (AUC) assessment has verified the model's performance, resulting in an overall AUC of 74.96%. This suggests that a model-driven map is a reliable spatial support tool for sustainable ecotourism development in areas with diverse landscapes and resources.

1. Introduction

Ecotourism blends environmental conservation, cultural diversity, and community outreach to generate financial incentives for nature conservation and scientific inquiry, highlight cultural and ecological concerns, and protect natural resources while sustaining ecosystems and maintaining development concepts [[1], [2], [3], [4]]. It also involves social responsibility and sustainable travel, making it an imperative industry that promotes cultural attractions, traditions, and social values, ultimately improving the well-being of local inhabitants [2,5,6]. Sustainable tourism and ecotourism are intertwined and are often attributed to ecological tourism, appealing to those who value environmental and social sensitivity [7,8]. The ecotourism industry, comprising human-nature interactions, has emerged as a nexus between tourism development and biodiversity protection, addressing the intrinsic flaws of conventional mass tourism practices that exceed the carrying capacity of the natural environment, leading to ecological ruin and impeding the workable aptitude of the tourism sector [9,10].

Ecotourism is gaining attention as one of global tourism's largest and fastest-growing service-based niche industries [2,11]. In 2019, the tourism sector, including ecotourism businesses, contributed around 10% of the global Gross Domestic Product (GDP) and was a primary source of employment, creating one out of every four new jobs worldwide [12]. Over the recent years, many developing countries in sub-Saharan Africa (SSA) [13,14], including Ethiopia, which has abundant natural resources and multiple attractions, have promoted the ecotourism industry [2,6,9,[15], [16], [17]]. Ethiopia has untapped ecotourism potential due to its diverse range of biological and physical reserves, including endemic mammal and bird species, as well as cultural and historical attractions that can attract tourists from all over the world [1,6,12]. These attractions include historical monuments, archaeological locations, religious structures, battlefields, museums, festivals, indigenous architecture, dress, handicrafts, artefacts, intangible cultural heritage, and expressions of physical and social attributes such as fantasy, nostalgia, pleasure, and pride [18]. Ecotourism plays a crucial role in diversifying livelihoods for communities and the country's economy while promoting ecological conservation, and cultural diversity, raising environmental education and awareness, and empowering local communities economically and socially. These are critical pathways to achieving sustainable development [9,15,[19], [20], [21], [22]]. Additionally, ecotourism can promote national identity development and authentic cultural preservation for future generations [23].

Indeed, the country recognizes the significant contributions that ecotourism can make to local communities and the national economy and, as a result, has targeted its development as an essential component of tourism [24]. Despite their documented benefits for environmental protection, cultural diversity, poverty reduction, and livelihood improvement for the community [15,25], the tourism sector and the ecotourism industry, however, are yet underdeveloped compared to other African countries. One of the most pressing contemporary challenges to ecotourism development is a lack of sectoral integration, involvement, promotion, training, and information [26]. Therefore, it is essential to note that human-induced influences related to the ecotourism industry can negatively impact the ecosystem and the living conditions of the local inhabitants if not well planned and handled [27]. To mitigate these potential impacts, an integrated and multifaceted approach must be strengthened, encompassing community involvement, government intervention, capacity building, stakeholder participation, biodiversity conservation, policy enforcement, scientific research and physical/environmental and tourism resource tracking. This can be a promising way to help identify the underlying and immediate challenges and convey strategies to address asymmetric interlinkages between ecotourism and ecological sustainability [6,9,19].

A growing body of studies emphasizes the importance of conducting a comprehensive land suitability analysis to identify and prioritize potential areas for sustainable ecotourism development [1,15,16,28]. However, determining suitable sites for ecotourism development can present a spatial decision-making challenge with various, sometimes conflicting, issues [29]. For example, Wight [30] highlights the need for the responsible promotion of ecotourism that does not deplete the resources on which it is based. This requires cooperation and understanding among all parties involved, including the government, non-governmental organizations (NGOs), travellers, tour guides, and communities. Additionally, ecotourism sites must meet tourist expectations and maximize visitor satisfaction while benefiting the environment, local communities, and the nation long term [30]. To address this challenge, the Geographical Information System (GIS) is a well-suited tool for conducting multi-criteria decision-making (MCDM) analysis to evaluate the criteria and relative importance of environmental, social, and economic factors [31,32]. GIS can handle a wide range of data from multiple geographical, temporal, and scale levels from various sources. It is a quick and low-cost tool for site identification, environmental impact assessment, spatial planning, and informed policy decisions for sustainable ecotourism development. MCDM aims to generate an assessment index using data from various criteria [33]. This approach has become the most promising tool and has been widely used as an effective solution for sustainable ecotourism development in regions with adequate resources, which involves adopting management measures to moderate the precarious impediments to ecosystem and biodiversity [10,16,34].

The East Hararghe Zone (EHZ) is situated in the eastern region of Ethiopia's Oromia Regional State and is considered a popular tourist destination owing to its diverse landscapes and natural resources suitable for ecotourism development. The most notable attractions in the EHZ comprise its beautiful natural lakes, including Adele, Finkile, and Haramaya, which had disappeared for over ten years but have now recovered and are and are home to aquatic bird species [35]. The Gara Muleta and Kundudo mountains are also home to natural forest reserves covered mainly by dense Afro-Alpine vegetation, which serves as a habitat for wild animals and birds. Furthermore, the Babile Elephant Sanctuary is another protected area home to a renowned African elephant species, “Loxodonta Africana Africana” [19]. Alongside these natural attractions, the EHZ also boasts several historical, cultural, and archaeological destinations [6].

In addition to the aforementioned tourist destinations, the primary objective of this study is to identify other potential areas for ecotourism in the zones' landscapes. We used a combined modeling approach to achieve this, which involved conducting an automated GIS-based multi-criteria decision-making process using input parameters obtained from remotely sensed and field surveys. We also considered various environmental and socioeconomic criteria and factors. Each criterion's relative importance and weight were determined using the analytical hierarchy process (AHP). Subsequently, they were integrated into the weighted linear combination (WLC) system. Assessing potential ecotourism resources is advantageous for planning, policy decisions, and investors interested in finding prospective locations to establish ecotourism ventures. Such ventures can enhance the lives of communities by creating job opportunities, developing infrastructure, fostering regional integration, and contributing to the national economy through investment and foreign exchange revenue. In this regard, the findings of this study will help mitigate unsustainable tourism practices, promote eco-friendly tourism and maintain environmental sustainability in the study area.

2. Method and material

2.1. Study area

The EHZ, one of the administrative regions of the Oromia Regional State, is spatially located in eastern Ethiopia between 07⁰ 36′ N and 09⁰ 41′ N and 41⁰ 18′ E and 43⁰ 00′ E (Fig. 1a–c). It is located 400 km east of Addis Ababa, 300 km south of Djibouti, and 250 km west of Hargeisa. The study area consists of twenty administrative districts and has a total surface area of about 24,933 km^2, of which ca. 30–40% is classified as lowland (locally known as Kolla; 500–1500 m), while cool sub-humid (Woinadega, which has an altitude ranging from 1500 to 3200 m) and cool humid (Dega, with an altitude range of 2300–2300 m) agroecological zones roughly accounted for 35–45% and 15–20%, respectively. Kolla has annual rainfall averages of less than 700 mm, while Woinadega and Dega have more than 1,200 mm [36]. The zone's annual minimum and maximum temperatures are 10.34 °C and 26.7 °C, respectively, with a mean temperature of 20.85 °C. With an average elevation of 1505 m, the altitude varies greatly, from 511 m in the southern part to 3386 m at Gara Muleta, the highest peak in the northwest. There are two rainy seasons: Belg, which is limited in Dega, and Meher, a long-cycle cropping and production season for staple crops like sorghum and maize [37].

Fig. 1.

Location of the study area. (a) Location of Ethiopia in Africa, (b) location of the East Hararghe Zone in eastern Ethiopia, and (c) administrative districts.

2.2. Datasets and sources

Field surveys, questionnaires, and key informant interviews collected the required information for ecotourism potentiality mapping. Field visits were conducted, and data were gathered from various ecotourism sites with authorization from Babile Elephant Sanctuary headquarters and East Hararghe Zonal Agriculture and Tourism office. To ensure reliability and accuracy, a careful approach was employed to select expert informants for the study. Thirteen experts were chosen based on their demonstrated expertise, diverse professional backgrounds, and valuable contributions to relevant projects and publications. The availability and willingness to contribute to the research were also considered during the selection process. Additionally, referrals from previous informants were utilized to expand the pool of participants. It is worth noting that all selected participants provided written consent and received appropriate compensation. This careful selection significantly enhances the quality and reliability of the study, leading to a comprehensive and robust analysis as well as an informed decision-making process. Secondary data were obtained from various open sources and organizations, (as shown in Table 1)

Table 1.

Geospatial data and their sources.

Factor	Resolution or scale	Source
Administrative boundary	Shapefiles (1:25000)	UN Humanitarian [38]
Elevations, viewshed, slope, and water streams	30 × 30 m	ASTER† [39]
Roads, cultural areas, towns	Shapefiles (1:25000)	OSMΩ [40]
Temperature and rainfall	0.93 × 0.93 km; resampled to 30 × 30 m	Worldclim [41]
Land-use/land-cover	30 × 30 m	Landsat 8 OLI-TIRSα satellite imagery [42]
Protected areas	Shapefiles (1:25000)	World Database on Protected Areas [43]
Fault areas	Shapefiles (1:25000)	GSE±

†ASTER–Advanced Spaceborne Thermal Emission and Reflection Radiometer.

ΩOSM: Open Street map.

αOLI-TIRS: Operational Land Imager/Thermal Infrared Sensor.

±GSE: Geological Survey of Ethiopia.

2.3. Variables development

Based on expert judgments and information from various sources, we establish the following factors as indicators of appropriateness within the terrestrial ecosystems of the EHZ: landscape/naturalness, wildlife, topography, accessibility, geology, and climate. Ecotourism locations were evaluated using rainfall, temperature, elevation, slope, scenic attractiveness (viewshed), distance to the road, distance to the stream, distance to cultural areas, distance to the town, and distance to the fault line, land use, and cover, and protected areas. Euclidean distances and buffers were computed using several guidelines to discover characteristics and indicators of land suitable for ecotourism. The elevation, slope, visibility, and stream factors were calculated using a 30 m DEM resolution. The LULC factor was classified and reclassified using a 2021 Landsat image based on biophysical characteristics of potential ecotourism resources [45]. Protected areas were converted from vector maps to raster layers and evaluated for their potential as ecotourism sites. Then, a proximity analysis using Euclidean distance classifiable distances was followed. As a result, all the criteria were standardized. Standardization is a procedure to translate the original standards into similar units and scale them [46]. Before obtaining assessment competency for suitable weighting elements, an Analytic Hierarchy Process (AHP) as an MCDM method was used to organize the identified criteria into a hierarchy structure Table 2. Data from selected criteria were maintained, displayed, and manipulated independently. This study used ArcGIS 10.8 and Terraset 2020 software. GIS software is the most powerful technology to store, retrieve, integrate, analyze, and display information according to user-defined parameters. It has an essential impact on decision-making and planning [47]. Each processing detail is shown in Fig. 2.

Table 2.

The scale of pairwise comparison.

Intensity of importance	Definition
1	Equal importance
3	Moderate importance of the first element over the second element in the pair
5	The strong or essential importance of the first element over the second element in the pair
7	Very strong importance of the first element over the second element in the pair
9	The extreme importance of the first element over the second element in the pair
2,4,6,8	An intermediate value between two adjacent judgments
Reciprocals of the above non-zero numbers	The inverse of the importance

Source: Saaty [44]

Fig. 2.

Methodological frameworks of GIS-based MCDM analysis. Source: Developed by authors.

2.3.1. Accessibility

• (a).Road Proximity

The distance from the road is an essential factor that influences ecotourism. For example, the closer the area is to the road network in a specified area of interest, the greater the likelihood the site is suitable to be considered for developing ecotourism, as the road proximity increases the chance of tourists finding it easier to reach a specific tourist destination [2,48]. They can allow visitors to learn, appreciate, and enjoy themselves [49]. To model potential ecotourism locations for the present study area, the proximity of highways to attractions was classified at different intervals: 0–7 km, highly suitable (S1); 7–14 km, moderately suitable (S2); 14–24 km, marginally suitable(S3); and >24 km, not suitable (N) (Fig. 3b and Table 5) [50].

• (b).River Proximity

Fig. 3.

Suitability index of each factor: (a) Cultural site; (b) Road; (c) Land use land cover; (d) Stream; (e) Town; (f) Protected areas; (g) Elevation; (h) Slope; (i) Temperature; (j) Fault line; (k) Viewshed; (l) Rainfall.

Fig. 5.

The ROC curve for the ecotourism potentiality by the AHP approach.

Rivers are essential resources for the growth of ecotourism [51,52] and play an important role in ecological function as they are home to various wildlife and flora [2]. A network of streams in a landscape can produce visually appealing geomorphologic elements, variation, and diversity that appeal to a wide range of visitors. Water must be readily available for swimming, recreation, drinking, cooking, and sanitation [53]. River proximity was determined using ArcGIS Euclidean distance from natural attraction sites. According to Suryabhagavan & Balakrishnan [50], buffers within (0–3 km) are highly suitable for ecotourism development, (3–6 km) are moderately acceptable, (6–9 km) are marginally suitable, and (>9 km) are not ideal for ecotourism development (Fig. 3d and Table 5).

• (C)Proximity to Cultural Sites

Table 5.

Areal and proportion of ecotourism criteria and sub-factors.

Criteria	Factor	Suitability Rate	Suitability Class	Area (km2)	Area (%)
Accessibility	C1	0–30	S1	11,284.35	45.26
		30–60	S2	6,472.93	25.96
		60–100	S3	4,998.12	20.05
		>100	N	2,291.95	9.19
	C2	0–7	S1	10,297.56	41.30
		7–14	S2	7,339.22	29.44
		14–24	S3	5,194.22	20.83
		>24	N	2,216.34	8.89
	C4	0–10	S1	6,536.62	26.22
		10–22	S2	11,440.09	45.88
		22–48	S3	6,473.35	25.96
		>48	N	597.29	2.40
	C5	0–3	S1	10,721.81	43.00
		3–6	S2	8,485.17	34.03
		6–9	S3	3,257.21	13.06
		>9	N	2,583.15	10.36
Geology	C6	>23	S1	2,042.59	8.19
		13–23	S2	5,336.53	21.40
		6–13	S3	6,169.03	24.74
		0–6	N	11,499.19	46.12
Topography	C7	2,668–3,388	S1	110.53	0.44
		1,948–2,668	S2	3,782.71	15.17
		1,228–1,948	S3	15,882.64	63.70
		508- 1,228	N	5,271.35	21.14
	C8	2–5	S1	6,041.41	24.23
		5–25	S2	11,280.03	45.24
		25–35	S3	1,115.97	4.48
		>35 & < 2	N	6,570.49	26.35
Landscape/Naturalness	C3	Waterbody & Forest	S1	2,101.86	8.43
		Grassland& Shrubland	S2	8,814.76	35.35
		Bareland & Farmland	S3	13,963.34	56.00
		Built-Up	N	139.00	0.56
	C11	6–9	S1	376.81	1.51
		3–6	S2	236.40	0.95
		1–3	S3	391.28	1.57
		0	N	24,038.92	96.42
Wildlife	C10	Sanctuary	S1	626.80	2.51
		National Park	S2	7,287.52	29.23
		Non-forest	S3	520.20	2.09
		Outside protected area	N	16,612.87	66.63
Climate	C12	1,071–1,273	S1	380.97	1.53
		867 - 1,071	S2	2,348.44	9.42
		662–867	S3	10,756.06	43.14
		460–662	N	11,561.87	46.37
	C9	10–14	S1	527.38	2.12
		15–19	S2	4,336.23	17.39
		20–23	S3	14,322.81	57.45

Cultural sites are defined in this study as tourist sites, including historical monasteries, palaces, historical buildings, bridges, museums, cultural celebration areas, and other cultural assets [53]. ArcGIS Euclidean distance functions were utilized to investigate the proximity of cultural sites. Euclidean analysis, based on the proximity of cultural places, was employed to classify the proximity to cultural sites. The most convenient cultural attraction site, which is the closest, has been designated as highly suitable. According to Bunruamkaew & Murayama [45], proximity to cultural and historical attraction sites within (0–15 km) is highly suitable, (15–30 km) is moderately suitable, (30–45 km) is marginally suitable, and (>45 km) is not suitable for ecotourism (Fig. 3a and Table 5).

• (d)Proximity to Town

The role of a location's proximity to a town in determining its suitability for ecotourism development is crucial. Investors and entrepreneurs can exploit existing infrastructure, labor markets, and services by being close to towns, boosting sustainable ecotourism efforts [54]. The proximity to towns also benefits local economies by promoting job creation, local entrepreneurship, and increasing demand for locally produced goods and services [55]. To maintain the integrity of the area, four buffer zones were created around the town at distances of 0–10 m, 10–22 m, 22–48 m, and >48 m. These buffers were subsequently evaluated based on their significance compared to other layers (Fig. 3e and Table 5).

2.3.2. Landscape/naturalness

• (a)Land use and land cover (LULC)

A geo-rectified Landsat 8 OLI/TIRS (Operational Land Imager/Thermal Infrared Sensor) image was used to create a land use and land cover (LULC) map with seven classes, namely farmland, forest, built-up (residential and commercial), shrubland, grassland, water bodies, and bare land. Preprocessing methods, such as band extraction, layer stacking, and image enhancement and restoration, were applied before the classification [35,56]. A supervised image classification method based on a support vector machine (SVM) algorithm [35] was performed to generate the LULC map. The accuracy of the map was verified using point data collected from the Google Earth image. Then, the biophysical vegetation characteristics derived from the LULC data were used as a model variable for potential ecotourism site suitability mapping [1,5]. According to Ambecha et al. [2], forests and water bodies are highly suitable, while shrubland and grassland are moderately suitable; bare land and farmland were also marginally suitable for ecotourism development, built-up land was deemed unsuitable (Fig. 3c and Table 5).

• (b)Visibility

Other attractions in the research area include unique landscapes and mountains. The landscape is the ideal raw material for tourism, and it can be analyzed according to terrain analysis for tourism and recreation [57]. The visibility factor, or scenic attractiveness, was calculated using a view-shed analysis from the DEM dataset integrated with the location of natural uniqueness [53]. OSM features were used to locate naturally distinct and culturally appealing locations. The landscape properties suitable for ecotourism were evaluated and classified as tourist attractions. Chhetri et al. [58] used viewshed analysis to determine scenic attractiveness. According to Bunruamkaew & Murayama [45] and Suryabhagavan & Balakrishnan [50], high visibility values between 6 and 9 are highly suitable, middle visibility values between 3 and 6 are moderate, low visibility values between 1 and 3 are marginal, and visibility values between 0 and 1 are unsuitable for ecotourism development (Fig. 3k and Table 5).

2.3.3. Topography

The topography describes the surface shape and relief of the land, encompassing various physical features that represent the external shape of the Earth [59]. When selecting a location for an ecotourism construction project, it is crucial to consider the elevation and slope. Evaluating the nature and elements of an area, including its position, angle, and stage, is necessary to determine which areas are appropriate for tourism [60,61].

• (a)Slope

The slope is one of the most important criteria for identifying potential ecotourism sites. The slope of the land is critical [45,62]. A low pitch of the land is required to develop ecotourism sites, and an increasing slope reduces the possibility of developing ecotourism sites [63]. Moreover, Geremew and Hailemeriam [64] state that a steeper slope with hanging and cliffs is more attractive to tourists than a gentler slope with hanging and cliffs. Bunruamkaew and Murayama [45] classify suitable slopes, with slopes of (2–5°) being highly suitable, (5–25°) moderately suitable, (25–35°) marginally suitable, and (>35° and 2°) unsuitable for ecotourism. Each class is prioritized based on importance, with a moderate slope indicating a tremendous potential for developing ecotourism sites (Fig. 3h and Table 5) [2,8].

• (b)Elevation

Moderate altitude is positively correlated with the potential for ecotourism [65,66], making it one of the essential variables in identifying tourist attractions [5,67] in the study area. The elevation in the area ranges from 508 to 3388 m, and the elevation classes are evaluated based on their landscape or topographic attractiveness for the tourist feature. The elevation value of (2,668–3,388 m) was rated as highly suitable, (1,948–2,668 m) as moderately suitable, (1,228–1,948 m) as marginally suitable, and (508–1,228 m) as not suitable for ecotourism (Fig. 3g and Table 5) [2].

2.3.4. Wild life (protected areas)

Wildlife areas are concerned with reserving and protecting areas where habitats and species can be found. Ecotourism and wildlife areas are inextricably linked [[68], [69], [70], [71]]. Polygon features of wildlife areas were obtained from the world database on protected areas and then converted to raster format. These areas were then categorized according to their ratings, with Wildlife Sanctuary (WS) ranked as high, National Park (NP) classified as moderate, and Non-Forest Reserve (NFR) classed as medium. Conversely, areas outside the protected area were deemed unsuitable for ecotourism (Fig. 3f and Table 5) [45].

2.3.5. Geology

• (a)Proximity to Fault Lines

Represents a linear function of ecotourism suitability, with a considerable distance away from geological faults indicating highly suitable locations [72]. The fault line is the nearest and has been deemed unsuitable. According to Mansour et al. [72], proximity to a fault line within a (0–6 km) distance was rated as unsuitable, (6–13 km) as marginally suitable, (13–23 km) as moderately suitable, and (>23 km) as highly suitable for ecotourism (Fig. 3j and Table 5).

2.3.6. Climate

Climate conditions are also necessary when selecting a site for ecotourism development [8]. The Worldclim raster image was used (precipitation and temperature). The Inverse Distance Weighted (IDW) toolset in ArcGIS Spatial Analyst tools interpolated annual mean rainfall and annual mean temperature [1].

• (a)Rainfall

In the region under study, rainfall variations range from 459 mm to 1273 mm. The area in the northwest receives the highest amount of precipitation (>1072 mm). Assignments are made for each classification based on the importance of rainfall availability in enhancing ecotourism potential [73,74]. A rainfall value within the range of (1,071–1,273 mm) is rated as highly suitable, (867–1,071 mm) as moderately suitable, (662–867 mm) as marginally suitable, and (460–662 mm) as unsuitable for ecotourism (Fig. 3l and Table 5) [1].

• (b)Temperature

The temperature of an area determines whether it is suitable for human recreation [53]. Temperatures ranging from 10 to 14 °C are considered very suitable, whereas temperatures between 15 and 19 °C are moderately suitable. Temperatures between 20 and 23 °C are marginally suitable, and temperatures between 24 and 27 °C are considered unsuitable for recreational activities (Fig. 3i and Table 5). The lower temperature range is preferred for tourism as most humans prefer a cooler environment for survival and entertainment. Additionally, most living things can survive better in lower temperatures versus higher temperatures [8,53].

2.4. Ecotourism suitability modeling

2.4.1. Estimating the weights and consistency: the AHP method

The AHP, developed by Saaty [44,75], is designed to solve problems and aid decision-making by considering various variables [76]. This method effectively manages qualitative and quantitative data [76,77], especially in complex and subjective decision-making scenarios. AHP helps decision-makers understand the decision-making model's structure by comparing each level's pair of items to the level above based on their importance [76,77]. The comparisons can be expressed In Eq. (1) as multiple square matrices, as shown below: C = ^[Cij^]nxn

$$\lfloor \begin{array}{c}{C}_{11}{C}_{12}{C}_{13}\\ C_{21}{C}_{22}{C}_{23}\\ C_{31}{C}_{32}{C}_{33}\end{array}\rfloor$$ (1)

The representation of matrices Eq. (2) that have reciprocal properties, $R={\left|\frac{1}{Cij}\right|}_{nxn}$ is;

$$\left[\begin{array}{c}\frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{11}}}\frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{12}}}\frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{13}}}\\ \frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{21}}}\frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{22}}}\frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{23}}}\\ \frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{31}}}\frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{32}}}\frac{\mathbf{1}}{{\mathit{C}}_{\mathbf{33}}}\end{array}\right]$$ (2)

The Saaty Scale, established by Saaty in 1980 [44], makes subjective pairwise comparisons. The scale is represented in Table 3. Once all pairwise comparison matrices are formed, a vector of weights, W = [W₁ W₂ …, W_n], is generated using Saaty's eigenvector approach [44].

$${\mathit{X}}_{\mathit{I}\mathit{J}}=\frac{C_{ij}}{\sum _{i=1}^n{C}_{ij}}\lceil \begin{array}{c}{X}_{11}{X}_{12}{X}_{13}\\ X_{21}{X}_{22}{X}_{23}\\ X_{31}{X}_{32}{X}_{33}\end{array}\rceil$$ (3)

Table 3.

Average random inconsistency indices (RI).

n	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
RI	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45	1.49	1.51	1.48	1.56	1.57	1.59

n = number of factors; RI = random index [44].

The weights are determined in two stages: The first involves normalizing the pairwise matrix, and the second involves constructing a weighted matrix [78]. The pairwise comparison matrix, C = [C_ij]_nxn, will be normalized using Eq. (3) as presented in Chen [78]. Each element in the matrix will be divided by the sum of its respective column [79], for all j = 1, 2, …, n. To obtain a weighted matrix [80], the division of the sum of the matrix columns by the number of elements used, n, is done for every instance of i using Eq. (4).

$$W_{ij}=\frac{\sum _{j=1}^n{X}_{ij}}{n}\lceil \begin{array}{c}{W}_{11}{W}_{12}{W}_{13}\\ W_{21}{W}_{22}{W}_{23}\\ W_{31}{W}_{32}{W}_{33}\end{array}\rceil$$ (4)

The consistency vector is then calculated by dividing the weighted sum vector by the weight of the element, Cv [80].

$${Cv}_{11}=\frac{1}{W_{11}}\lceil C_{11}{W}_{11}+C_{12}{W}_{21}+C_{13}{W}_{31}\rceil$$

$${Cv}_{21}=\frac{1}{W_{21}}\lceil C_{21}{W}_{11}+C_{22}{W}_{21}+C_{23}{W}_{31}\rceil$$ (5)

$${Cv}_{31}=\frac{1}{W_{31}}\lceil C_{31}{W}_{11}+C_{32}{W}_{21}+C_{33}{W}_{31}\rceil$$

According to Chen [78], there is a relationship between the weight vector W and C's pairwise comparison matrix. This relationship is shown in Eq. (6).

$$C_W={\lambda }_{\mathit{max}}W$$ (6)

${\mathrm{\lambda }}_{\max }$ represents the maximum eigenvalue of the comparison matrix [81]. In the AHP technique, it is a critical variable for validation, serving as a reference index during consistency checks by calculating the consistency index (CI). To obtain it, average the values of the consistency vector, Cv, as indicated in Eq. (7) [5].

$$\lambda =\sum _{i=1}^n\left({CV}_{ij}\right)$$ (7)

The consistency index, CI, for each matrix of order n is obtained using Eq. (8) [78].

$$CI=\frac{{\lambda }_{\mathit{max}}-n}{n-1}$$ (8)

The consistency ratio, CR, is the ratio of CI and RI, as shown in the formula Eq. (9).

$$CR=\frac{CI}{RI}$$ (9)

The random consistency index (RI) is calculated using a randomly generated pairwise comparison matrix. If the CR value is greater than 0.1 (CR > 0.1), the pairwise comparisons must be re-examined [5]. Table 3 displays the RI values for matrices ranging from order 1 to 15 (i.e., 1 to 15 elements in one level) [44].

The criteria and sub-criteria were selected based on prior studies, natural conditions within the study area, and expert opinions [82]. Terraset 2020's Super Decision program was utilized to ascertain the weights of each layer (sub-criteria) and the inconsistency rate for each judgment. During the evaluation, the consistency rate of each decision ought to be less than 0.1 [32,83]. Eigenvectors and consistency ratios were obtained for each criterion as weights [84]. Table 4 displays the ratings on a nine-point continuous scale designed by Saaty [44] for the AHP decision-making process [75,78].

Table 4.

Pairwise comparison of twelve factors in nine-point continuous scales.

	C1	C2	C3	C4	C5	C6	C7	C8	C9	C10	C11	C12
C1	1
C2	1/3	1
C3	1/3	1/3	1
C4	1/3	1/2	3	1
C5	1/2	3	3	2	1
C6	1/2	2	2	3	2	1
C7	1/4	1/2	2	1/2	1/3	1/3	1
C8	1/4	1/3	2	1/2	1/3	1/3	1/2	1
C9	1/3	1/4	1/3	1/4	1/5	1/5	1/3	1/3	1
C10	1/2	3	3	3	2	2	2	4	5	1
C11	1/3	1/3	1/2	1/3	1/4	1/4	1/2	1/2	2	1/4	1
C12	1/3	1/4	1/2	1/4	1/5	1/5	1/3	1/3	2	1/2	2	1

Source: Computed by authors. Criteria terminology: C1: Cultural sites; C2: Road; C3: Land-use and land-cover; C4: Town; C5: Stream; C6: Fault; C7: Elevation; C8: Slope; C9: Temperature; C10: Protected area; C11: Visibility; C12: Rainfall.

2.4.2. Land suitability assessment

GIS-based MCDM techniques are widely used for land suitability analysis in the map overlay process [83,85]. The primary benefit of utilizing MCDM is its ability to incorporate information from multiple criteria to generate a single evaluation index by assessing all components at various scales [84]. For the present study, the selection and prioritization of influence criteria were accomplished by reviewing relevant literature and complementing the perspectives of the experts in the field based on their preliminary knowledge and judgment. Furthermore, it enables researchers to merge data from multiple criteria based on their relative weights, guided by expert knowledge, to produce a single output map [31,83,84,[86], [87], [88]]. Then, each thematic layer was combined with its weight using the weighted linear combination (WLC) method in the ArcGIS software environment [[89], [90], [91]]. The WLC method (Eq. (10)), a straightforward approach to overlaying the raster datasets on a standard rating scale and weighting each based on its priority, multiplies the suitability factor (Vi) by the corresponding weight of the individual factor (Wi) [92,93]. This results in an ecotourism map of the study area with four suitability classes, namely, highly suitable (S1), moderately suitable (S2), marginally suitable (S2), and not suitable (N), following the Food and Agriculture Organization (FAO) framework for land evaluation [94]. Finally, we applied a majority filter function to refine the salt and pepper effect of the suitability map.

$$\mathit{{\rm E}}=\sum _{\mathbf{i}=\mathbf{1}}^{\mathit{n}}{\mathit{W}}_{\mathit{i}}{\mathit{V}}_{\mathbf{i}}$$ (10)

where W_i represents the relative importance (weight) of parameter i, V_i represents the relative weight of parameter i, and n represents the total number of parameters.

2.4.3. Model validation

The map assessing the area's suitability for ecotourism was validated using various tourist attraction sites. For this purpose, 94 Ground Control Points (GCPs) were identified using portable Global Positioning System (GPS) data on areas with the potential to support ecotourism, such as important animal habitats, cultural sites, historic mosques and churches, wildlife attractions, and vistas. Out of all the GCPs from tourist destinations, a random split of 70% (65 sample points) was utilized for model training, while the remaining 30% (29 points) was used for model validation by measuring the statistical value of the area under the curve (AUC) by plotting the receiver operating characteristic (ROC) [92,95]. The ROC exhibits all potential threshold values by presenting the false-positive and false-negative values on the Y and X axes. By relating the ability of systems to predict predetermined “events,” the AUC measures the prediction accuracy of both absence and presence. The AUC value and prediction accuracy associations could be classified as follows: poor (0.5–0.6); average (0.6–0.7); good (0.7–0.8); very good (0.8–0.9); and excellent (0.9–1) [96].

3. Results

3.1. Ecotourism suitability factors

In order to build a suitability model for identifying and mapping ecotourism potentiality in the EHZ (ecotourism high potential zone) of the Oromia Regional State in Ethiopia, six criteria were carefully considered. These criteria include accessibility, geology, topography, landscape/naturalness, wildlife, and climate. Twelve corresponding input factors were integrated into the model to account for these criteria. To evaluate the suitability of each factor, a four-scale classification system was applied: S1, representing areas with a high-capacity to meet the defined criterion requirements; S2, representing areas that moderately meet the necessary criteria; S3, signifying areas that meet only some of the established standards, and N, representing areas that do not meet the required criteria conditions. Table 5 outlines the classified criteria and their corresponding attributes for each suitability factor. The resulting maps derived from this analysis consist of thematic layers ranging from unsuitable to highly suitable for ecotourism potentiality (as shown in Fig. 3).

3.2. Analytical hierarchy process (AHP)

Table 6 provides the input variables pairwise comparison for relative preferences based on expert opinions and weighted by AHP, along with the CR and weight determined for each factor. The CR was computed to be 0.06, which is acceptable. The importance of the protected area comes in second place with a weight of 15.49%, followed by the distance from the fault, which comes in third place with a weight of 13.65%, and proximity to cultural sites, which is given a higher weight of 17.97%, according to the paired comparison in the AHP model. Following roads (9.02%) and the proximity to towns (7.69%), the proximity to a stream is the fourth most important factor (12.59%). The weights assigned to the remaining variables, including elevation, slope, LULC, rainfall, and visibility, were 5.87%, 4.86%, 4.37%, 3.29%, and 2.98%, respectively. As opposed to the stated criteria condition, the temperature is the least preferred factor, with a weight of 2.23% (Table 6).

Table 6.

Pairwise comparison of twelve factors in nine-point continuous scales.

Factors	C1	C2	C3	C4	C5	C6	C7	C8	C9	C10	C11	C12	∑
Weights	0.18	0.09	0.04	0.08	0.13	0.14	0.06	0.05	0.02	0.15	0.03	0.03	1
Percentage (%)	17.97	9.02	4.37	7.69	12.59	13.65	5.87	4.86	2.23	15.49	2.98	3.29	100

CI = 0.06 (acceptable).

3.3. Potential ecotourism sites

The suitability map for ecotourism in the EHZ was created by combining thematic maps using GIS-based MCDM frameworks, as depicted in Fig. 2. Our findings indicate that the high and moderately suitable classes were the major contributors to the overall potential areas in the EHZ, constituting approximately 26.19% (ca. 6,531 km²) and 35.34% (8,812 km²) respectively (Table 7). Additionally, the landscape was classified as marginally suitable for ecotourism (25.28%), while the remaining area, which accounted for a smaller proportion of the study area, was deemed unsuitable for ecotourism (13.17%). Our modelling results demonstrate that the potential regions for ecotourism development were unevenly distributed geographically across different parts of the zone, which could be attributed to variations and impacts associated with the model factors considered, such as climate, biophysical features like topography, geology, and land cover, and proximate attributes and infrastructure accessibility across the zone's landscape (Fig. 4).

Table 7.

Area coverage of suitable class.

Level of suitability	Area coverage
	km2	%
Highly suitable (S1)	6,530.95	26.19
Moderately suitable (S2)	8,812.15	35.34
Marginally suitable (S3)	6,303.65	25.28
Not suitable (N)	3,285.79	13.17

Fig. 4.

Potential ecotourism Suitability map for the EHZ, Ethiopia.

Table 8 provides a classification of the proportion of district-level ecotourism potential that can be used as a reference for developing a database and inventory of areas with ecological tourism potential. This classification helps minimize planning fragmentation and redundancy. For instance, in the southeast, a substantial part of the highly suitable classes was primarily concentrated in Babile (1,548.91 km², which is 6.17%, attributable to protected areas with biological resources and wildlife reserves, such as the Babile Elephant Sanctuary), Fedis (which covered 527.52 km², or 2.11% of the area) in the southeast, Gola Oda (800 km², or 3.20%), Bedeno (578.73 km², or 2.32%), Girawa (238.72 km²), and Melka Balo (201.62 km²) in the southwest of the study area. The remaining potential areas belonging to this suitability class, which has significant potential for ecotourism, were identified in administrative districts, such as Haromaya (which has freshwater lakes such as Lake Haromaya, Lake Adele, and Lake Finkle, with a variety of birds and aquatic species), Kersa, Meta, Kombolcha, Gursum, and Jarso in the north and northeast, and splendid landscapes in the northwest, such as Deder, Goro Gutu, Goro Muti, and Meta (Table 8).

Table 8.

Classification of potential ecotourism areas at the district level.

District	Highly suitable (S1)		Moderately suitable (S2)		Marginally suitable (S3)		Not suitable (N)
	Km2	%	Km2	%	Km2	%	Km2	%
Babile	1,548.91	6.17	1056.88	4.24	1669.35	6.69	589.94	2.36
Bedeno	578.73	2.32	243.98	0.97	0.02	0.00	0.00	0.00
Chinaksen	9.57	0.03	269.67	1.08	696.81	2.79	568.11	2.27
Deder	385.75	1.54	244.36	0.98	2.10	0.08	0.00	0.00
Fedis	527.52	2.11	173.86	0.69	0.00	0.00	0.00	0.00
Girawa	238.72	0.95	630.10	2.52	241.02	0.96	0.06	2.47
Golo Oda	800.02	3.20	662.01	2.65	652.16	2.61	7.21	0.02
Goro Gutu	172.86	0.69	253.37	1.01	66.50	0.26	0.05	0.02
Goro Muti	163.23	0.65	54.44	0.21	0.00	0.00	0.00	0.00
Gursum	317.75	1.27	550.18	2.20	36.20	0.14	0.00	0.00
Haromaya	160.50	0.64	379.41	1.52	33.55	0.13	0.00	0.00
Jarso	90.72	0.36	272.64	1.09	208.90	0.83	1.29	0.05
Kersa	276.83	1.11	198.03	0.79	0.97	0.03	0.00	0.00
Kombolcha	164.23	0.65	129.26	0.51	0.24	0.09	0.00	0.00
Kumbi	263.01	1.05	786.35	3.15	881.14	3.53	1503.5	6.03
Kurfa Chele	41.77	0.16	191.52	0.76	10.23	0.04	0.00	0.00
Melka Balo	201.62	0.80	749.62	3.00	311.34	1.24	0.21	0.08
Meta	264.35	1.06	174.60	0.70	20.53	0.08	0.00	0.00
Meyu Muleke	91.56	0.36	667.05	2.67	1281.04	5.13	614.98	2.46
Midhaga Tola	233.02	0.93	1127.45	4.52	189.39	0.75	0.00	0.00
Total	6,530.67	26.16	8814.88	35.36	6301.57	25.28	3285.3	13.18

The primary attributes of these districts are that they receive high rainfall and have good vegetation cover, which results in the highly alluring potential for ecotourism, including spectacular terrain, wildlife, and diverse vegetation types. These districts can attract eco-tourists and encourage ecotourism activities by providing appropriate infrastructure and services consistent with the local natural character. However, ecotourism development in these areas must address environmental issues that can be addressed through sustainable planning in ecotourism-friendly regions, such as resource conservation, environmental management, and tourism development. On the contrary, most districts in the lower altitudes in the northeast (e.g., Chinaksen), southernmost extent (Kumbi and Meyu Muleke), southeastern (some parts of the south of Babile), central highlands, and the northwestern part of the zone contributed the largest and most significant proportions of marginally to not suitable areas for ecotourism development.

3.4. Model validation

The accuracy of the modeled potential ecotourism map was validated based on comparative references from existing ecotourism locations (Fig. 5). Fig. 6(a–q) provides an overview of existing ecotourism destinations from which references for this study's model validation were collected. An overall accuracy of 74.96%, which indicates the model's good performance, was obtained by applying statistical accuracy assessments of the AUC by plotting the ROC curve to verify the accuracy of the ecotourism suitability model.

Fig. 6.

Existing ecotourism and landscape features in the EHZ, Ethiopia.

4. Discussion

Potential ecotourism location modeling is a complex procedure that requires the simultaneous examination of numerous factors, including environmental and socioeconomic criteria [97]. Previous studies have emphasized the importance of considering these variables when planning a strategy for future ecotourism development (e.g., Refs. [54,67]. In the present study, we modeled ecotourism potential in the EHZ, Ethiopia, using a GIS-based MCDA approach to determine the extent to which the zone areas could be suitable for ecotourism. We developed an ecotourism suitability model using six different criteria: accessibility, geology, topography, landscape/naturalness, wildlife, and climate, consisting of twelve factors including rainfall, temperature, elevation, slope, viewshed, proximity to roads, streams, cultural areas, towns, and fault lines, land use and cover, and protected areas. The criteria for evaluating potential ecotourism sites were established based on expert opinion to reduce biases in judgment. The model factors adopted from prior studies [1,2,43,47,49,52,71] were weighted using the AHP method. This method has been widely used as a practical tool for evaluating the thematic layers (model factors) and identifying suitable locations for ecotourism [8,62,[97], [98], [99], [100]].

The overall computed CR, which is 0.06, falls below the maximum acceptable value of 0.1 when all factors are considered [32,83]. This suggests that the experts' preferences were plausible, resulting in a positive outcome. Accordingly, the higher AHP weight was assigned to proximity to cultural sites (17.97%), while the least weight was given to temperature (2.23%), suggesting their importance in determining the final suitability map (Table 6). The results show that the essential elements of ecotourism are the proximity to cultural sites, protected area, and the distance from faults. It is imperative to recognize that any ecotourism activity must not degrade the environment, or else the purpose of ecotourism is lost, and natural resources are destroyed, leading to unsustainable development. To ensure sustainability, it is essential to foster a connection between ecotourism and local communities. This will help locals generate income and support ecosystem protection, thus contributing to sustainable development.

Based on the WLC, the suitability map was produced and categorized following the established suitability classification method used in previous studies [1,50]; the modeled potential ecotourism map of the study area was divided into four levels of suitability: S1, S2, S3, and S4. Accordingly, the most suitable ecotourism option was identified: S1 covered 26.19% of the study area. A more significant proportion of this class (53%) is confined within Babile, Fedis, Golo Oda, and Bedeno, which are spatially located in the southeast and southwest of the study area.

Given their proximity to protected areas (e.g., Babile Elephant Sanctuary), cultural/historical sites (e.g., Gara Muleta Royal Prison and Mesgida Biyo Guda; caves like Abdulahi Ibro, Gole Gaya, and Rakober Zala) and other attractions, these areas are ideal for ecotourism. These areas were also categorized as ecotourism hotspots due to their proximity to waterbodies and major towns, accessibility to infrastructures like roads, ecological and biological richness, and diversity that support various tourism activities. It is, however, worth mentioning that maintaining natural ecosystems and implementing specific ecotourism safety rules would be critical to sustainable ecotourism. Providing environmental protection education to tourists and residents is important to improve this activity further. These activities comprising camping, trekking, bird-watching, sightseeing, and other activities centered on the visual features of the landscape, can be established. This activity is beneficial regarding its environmental, economic, and sociocultural impacts, focusing on the sociocultural aspects.

The area of S2 covered 8,812 km² (35% of the total area), of which 57.28% (ca.5,050 km²) is found in Meyu Muleke and Kumbi in the south; Melka Balo and Golo Oda in the southeast; and Babile and Midhaga Tola in the southeast. In addition, the portion of S3 occupied 25% of the area. This class has fewer weights for ecotourism development. They can, however, be used as adequate public conveniences—services and facilities (e.g., communication, transportation, green hotels, restaurants, eco-lodges, highways, and health centers) are developed. On the contrary, a small part of the lowland areas in the northeastern, southern, and southeastern parts (which accounted for 13% of the site) was categorized as having unsuitable potential for ecotourism. This class is mainly found near geological faults; areas away from geological faults benefit ecotourism [72] and are far from towns, roads, water bodies, and cultural landmarks.

Various approaches are utilized for validating the suitability modeling of ecotourism, with the most popular being constructing a ROC-AUC, which has a value ranging from 0 to 1.0. A value of 0 denotes results that were lesser than random, while a value of 1.0 signifies absolute classification) [92,95,96]. The model's accuracy confirms the results' value [1,101]. Compared with the references from active ecotourism locations (Fg.5), the model accuracy calculated for the present study was 74.96%, indicating that the potential ecotourism map was accurately determined and that the designed model successfully identified potential sites with a high accuracy level. Consistent with our findings, various studies have demonstrated that GIS-based modeling and MCDA with AHP are cost-effective and time-efficient approaches for mapping prospective ecotourism sites [54,97,102]. For example, Sahani [8] modeled suitable locations for ecotourism development in Himachal Pradesh, India. Abrehe et al. [1] examined potential ecotourism site suitability for environmentally friendly natural resource management in Kafta Sheraro National Park, northwestern Ethiopia. Geremew and Hailemeriam [64] used the AHP method and GIS to evaluate land suitability in the Bench Maji Zone, southwest Ethiopia. Bunruamkaew and Murayama [5] investigated land suitability for sustainable ecotourism development. Pathmanandakumar et al. [103] used landscape, protected area, topography, accessibility, and community characteristics, assigned weights, and then integrated them into a GIS system with weighted overlay analysis to identify five possible ecotourism zones in Batticaloa District, Sri Lanka. Sobhani et al.

[104] employed an MCDA model, fuzzy set theory, and mathematical modeling to identify suitable zones for ecotourism development in Lar National Park and Kavdeh Wildlife Refuge, Iran. Furthermore, they evaluated landscape changes to assess ecological connectivity's ecological function and status [104] in these areas. Similarly, Mansour et al. [72] looked at seven spatial criteria: turtle beach, bird beach, sand beach, road network, mangrove trees, coral reefs, and built-up area, as well as six physical criteria: elevation, aspects, slopes, geology, fault lines, and soil types. Furthermore, Ambecha et al. [2] examined LULC, elevation, slope, rainfall, temperature, road network, and river to identify site suitability for ecotourism development in Ethiopia's Andiracha area. Unfortunately, although several studies have assessed ecotourism appropriateness, studies that statistically model the accuracy of modeled suitability maps are infrequently available [1,101]. This has prompted more research to bridge significant information gaps regarding the model's accuracy and sensitivity assessments, which show the model's performance and subsequent consistency maps.

This study determined the weights of the criteria through the opinions and preferences of experts. However, since the outcomes may vary depending on the participants, planners in any area can modify the model by adjusting the weights of the alternatives or criteria sets to suit new scenarios over time. While integrating AHP and GIS is a widely used approach that helps identify potential ecotourism sites for multifaceted decision-making processes requiring various factors, it is paramount to note their inherent limitations regarding the oversimplification of actual on-ground dynamics. This can lead to potential sites being overlooked, while subjectivity in assigning weights to criteria by AHP may introduce bias into the evaluation process. Both AHP and GIS heavily rely on data accuracy, which may not always be readily available, especially in underrepresented regions [29,[105], [106], [107], [108]]. Apart from these shortcomings, the findings of this study could serve as a reference for future studies that focus on land suitability modeling. Despite analyzing publicly available geospatial data, future studies regarding selecting potential ecotourism sites can extend the usage of higher spatial resolution datasets and suitability modeling techniques. This can be achieved by incorporating emerging artificial intelligence (AI) and machine learning algorithms to enhance the effectiveness of evaluating potential locations for ecotourism development.

5. Conclusions

The study was conducted in the East Hararghe Zone of Ethiopia to identify potential ecotourism sites using a GIS- based MCDA approach. The study analyzed twelve input parameters across six criteria, which included accessibility, geology, topography, landscape/naturalness, wildlife, and climate. The parameters were weighted using the AHP method and combined using the WLC method to map the spatial distribution of potential ecotourism sites effectively. The results showed that more than 60% of the zone's area had highly to moderately suitable classes for ecotourism development, primarily located in the southeast, southwest, and northern districts. However, over 38% of the area was identified as marginal to non-suitable for ecotourism development. Policy and decision-makers can use this study's findings to improve planning and investments, crucial for biodiversity conservation and sustainable ecotourism development. Moreover, ecotourism can potentially boost the local community's socioeconomic well-being in areas where it is feasible.

The research comprehensively analyzed the ecotourism potential of the East Hararghe Zone in Ethiopia using GIS-based MCDA. The study results showed that over 60% of the area was highly suitable for ecotourism development, mainly in the southeast, southwest, and northern districts. These areas have favorable accessibility, topography, geology, wildlife, naturalness, and climate parameters, making them attractive to tourists. The mapping of potential ecotourism sites is a valuable tool for policymakers and decision-makers in planning sustainable tourism development that benefits local communities and conserves biodiversity. The study provides information that can guide regional investments by identifying areas suitable for ecotourism development. This, in turn, can create employment opportunities and improve the economic well-being of the local population while conserving natural resources.

In conclusion, the study underscores the importance of ecotourism as a tool for promoting sustainable development in Ethiopia. The findings can guide policymaking and investment decisions for ecotourism development in the East Hararghe Zone and similar areas with ecotourism potential. Overall, the study contributes to the growing body of knowledge on ecotourism planning and management and highlights the need for further research in this area.

Declarations

Author contribution statement

Kalid Hassen Yasin: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Gezahegn Weldu Woldemariam: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Data availability statement

Data will be made available on request.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: