Modeling of the Economic Value of Ecosystem Services for Sustainable Resource Management in Northern Ethiopian Highlands

doi:10.21203/rs.3.rs-3341972/v1

Download PDF

Article

Modeling of the Economic Value of Ecosystem Services for Sustainable Resource Management in Northern Ethiopian Highlands

https://doi.org/10.21203/rs.3.rs-3341972/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Nature offers meaningful solutions to the climate crisis as we look to a world of greater risk and uncertainty with climate change. But while we know that nature is vital, we also know that we are losing nature at a rate unprecedented in human history. Global forest area is projected to decline gradually, which may affect existing ecosystem services (ESs) and the ecological and economic benefits of these services. We analyzed the ESs provision value of the Hugumburda State Forest (HSF) and the Southern Tigray-northern Ethiopian highlands, which was based on optical remote sensing, the Ecosystem Services Provision Index (ESPI), and the revised ESs valuation coefficients under different scenarios and the ecosystem services value function (ESVf). The forest coverage of HSF is anticipated to increase by 19.1 Km² (3.8%) until 2030 from its earlier coverage of 88.6 km² in the year 2020. As a result, ecosystem services value (ESV) provisions are expected to increase, sustain, and improve the local community's well–being by 1.88 million US$ ha–1 year–1. We also found a significant relationship between the ESV of forestland, shrubland, arable land, grassland, and water body and their area expansion (p < 0.01). The ESPI showed a higher value in the forest & grassland areas (0.38), whereas it was lower in water bodies (–0.51). HSF provides higher-level regulating services than provisioning, supporting, and cultural services. The combined use of optical remote sensing and ESPI is essential for developing a cost-effective ESs provision management framework regarding environmental and social benefits.

Earth and environmental sciences/Climate sciences

Earth and environmental sciences/Ecology

Earth and environmental sciences/Environmental sciences

Earth and environmental sciences/Environmental social sciences

ESs

ESV

ESVf

Optical remote sensing

GIS

HSF

We investigated ecosystem services provision sustainability under different scenarios from the period 1984–2030
The provision of ecosystem services increased from 3.7 million US$ ha^–1 year^–1 in 1984 to 10.6 million US$ ha^–1 year^–1 in 2030.
The relationship between ESV of forestland, shrubland, arable land, grassland, water body and their area expansion were significant (p < 0.01).
The study area provides higher regulating ecosystem services than provisioning, supporting and cultural.

Ecosystems are increasingly stressed, irreversible losses occur in many parts of the world, putting many natural systems' capacity for adaptation to the test, especially when human activity has already fragmented or damaged those ecosystems (Pede, 2020). More than ever, it is becoming obvious that these losses will only increase without risk mitigation and prevention strategies because of climate change, environmental degradation, population growth, and mass urbanization (Zari et al., 2019). One of the best choices we can make to secure our future and meet sustainable development objectives is to invest in ecosystem-based solutions to the planetary crisis (Koebele et al., 2020). One–third of the earth's land mass was covered by forest (~ 399.9 billion km²), providing numerous ecosystem services. However, its spatial coverage had slowly declined from 32.3% in 1990 to 31.3% in 2015 (Romijn et al. 2015; Miura et al. 2015). The global forest area designated for ecosystem services has been nearly 101.8 billion (25.45%) since 2015 (Miura et al. 2015). For instance, Russia's forest coverage ranked as one of the highest forest areas in the world, is estimated to be 81.5 billion km², while China's is 5.8 billion km². The ecosystem services in both countries have increased in the last two and three decades due to increased trends in forest coverage. In Africa, the total forest area is predicted to decrease by 2.8 million km² by 2030 from its earlier coverage of 67.4 million Km² in 2010 (d'Annunzio et al. 2015). This reduction may affect the provision of ecosystem services (ESs), the most important environmental element. However, the existing ESs research and published articles worldwide are limited, particularly in Africa. For example, globally, the total ESs–related research articles cover 38% of Europe, 28% of North America, and 15% of Asia; however, Africa, South America, and Australia cover 6–7% (Adhikari & Hartemink 2016).

The sustainability of human well–being strongly relies on the existing and future ecosystem functions and properties (Mouchet et al. 2017). According to the Millennium Ecosystem Assessment (MEA) (2005), ESs are defined as direct and indirect benefits people derive from various ecosystems for maintaining their well–being. Those benefits are categorized into provisioning (e.g., food production), regulating (e.g., climate regulation), supporting (e.g., soil formation), and cultural services (e.g., recreation) (Adhikari & Hartemink 2016; Wolff et al. 2015; Siebe & Pons 2015; Koschke et al. 2012; Martínez–Harms & Balvanera 2012; Raymond et al. 2009; MEA, 2005). Human health and well–being depend on ESs provision and are listed as one of the four ecosystem service categories (Mouchet et al. 2017; Koschke et al. 2012; Brown et al. 2007). This service provides the necessary amenities for continuing human well–being as well as sustaining the quality of human life because they are essential in adapting and mitigating the negative impacts of climate change by reducing emissions and managing forests in a sustainable way for livelihood support (Baral et al. 2018; Pandey et al. 2016; Euliss et al. 2010). Sustainable development is possible if scientific research and actions are taken with inclusive and cleverly designed solutions. The idea is based on the ecosystem approach, which seeks to holistically manage the environment to promote equitable conservation, restoration, and sustainable use. Forest ESs are crucial for providing many services and can improve if the local community administers them, as the welfare of ESs is linked with the local, regional, national, and international beneficiaries. However, human activities have been influencing the ecological environments and thus ESs by altering the patterns, density and spatial extent of forests and other vital land cover types (Lambert et al. 2017) due to a lack of knowledge on ecosystem functioning and provision (Koschke et al. 2012).

Land use and management decisions have been influenced by human demand for goods and services provided by the land (Wolff et al. 2015). The demands of human beings have been fulfilled by land and its potential resources (Birhane et al. 2019). Similarly, the demands for ESs worldwide have also increased at various spatial and temporal scales because the spatial distribution of ESs varies depending on their species composition, age, structure, and management (Rafik et al. 2018; FAO 2015; Tittensor et al. 2014). For instance, forest ESs no longer provide essential goods and services in Nepal due to forest degradation (Adhikari et al. 2018). The impacts of these services on the community's livelihood could be more extensive, mainly in Afromontane ecosystems (Solomon et al. 2019). Many failures, such as overexploitation, occurred because of a poor understanding and a lack of scientific knowledge on managing forest ecosystem services. For example, the original ecosystem services were not fully understood by the regional planning actors of the Regklam region, Germany (Koschke et al. 2012). This limitation shows little knowledge at broader planning levels about the importance and use of ESs at the landscape, community, and household levels (Adhikari et al. 2018; Akira et al. 2017; Euliss et al. 2010). ESs must be included in planning, managing, and analyzing land use policy (Paruelo et al. 2016; Koschke et al. 2012). There is no clear indication of the changes in ecosystem provision due to a lack of well–established criteria to identify ecosystem attributes & services (Paruelo et al. 2016). Monitoring forest lands to realize their ESs is essential because it was initially valued for providing wood but later used as a regulator of water resources and global carbon sinks (Yushanjiang et al. 2018; Parsa et al. 2016; Wolff et al. 2015).

The quantification and mapping of ESs based on various perspectives provide good insights into the current and future supply of goods to formulate policy (Wolff et al. 2015). Therefore, research on ESs should be supported with biophysical data sets and requires more accurate inputs and processes at applicable scales, resolution, and advanced analytical approaches for decision–making processes by enhancing the existing exploration of ESs (Paruelo et al. 2016; Koschke et al. 2012; Sherrouse et al. 2011). Consequently, remote sensing–based mapping of ESs has become increasingly applied (Vauhkonen 2018). ESs can be better derived using satellite imageries by generating land cover information. Human beings can easily alter land cover, which can influence the provision of ESs. In this regard, the earth observation data sets obtained from optical remote sensing (e.g., Sentinel, Landsat, & MODIS) contribute vital roles in modeling, analyzing, and mapping ESs (Caio et al. 2015). Analysis of ESs provision by integrating Remote Sensing, Geographical Information Systems, and ESPI data sets has received great attention worldwide because it models the spatial patterns of ESs provision across a landscape. The Cellular Automata–Markov Chain model is robust and can predict future ESs provision. The model has been incessantly used in various studies because it can show both the transition matrix and the probability of changes from one class to another in the future (e.g., Hishe et al. 2019; Gidey et al. 2017; Parsa et al. 2016).

The changes in land cover, climate and natural disasters are noted as some of the major drivers of ESs change (Wolff et al. (2015). The sustainability of ESs has not been assessed in the community forests in many developing countries due to, among other things, a lack of appropriate data, tools, and management framework (Paudyal et al. 2015). The Hugumburda State Forest (HSF) is one of the forest priority areas in Tigray–Ethiopia, that consists of unique biodiversity and endemic species. The HSF is a dry montane forest and sustains high biodiversity with a range of ESs in Tigray–Ethiopia (Kidane et al. 2010). This forest priority area has supported the livelihoods of local communities and downstream areas (Kidane 2015; Aynekulu 2011). However, there is a lack of documented studies that integrate a robust Ecosystem Services Provision Index (ESPI) in the study area. The study of ESs provision can be directly generated from land cover types of the study area, but it requires good–quality satellite data. For instance, using LiDAR and Synthetic Aperture Radar (SAR) products can help overcome the challenges of forest characterization and biomass estimation due to high cloud cover in the tropics (Romijn et al. 2015). The cloud-free optical remote sensing data sets are still valid to estimate ESs.

Investigating ESs supply capacity in major land use types at different periods is vital (Ncube et al. 2018). Numerous studies have also highlighted that the land cover change has highly influenced the provision of ESs. Egzampal Paudyal et al. (2015) in Nepal and Vauhkonen (2018) in Evo, Finland, conducted using biophysical and socioeconomic data sets to map ecosystem services and to predict the provision of potential ESs using airborne laser scanning (ALS) and forest resource data sets respectively. However, the authors lacked crucial information about the sustainability of the community–managed forest and spatial trends of the historical and predicted ESs provision in the area using the ALS and Light detection and ranging (LiDAR), respectively. A few ES–related studies were conducted using optical remote–sensing products in various parts of Ethiopia. For instance, Anley et al. (2022) in the Rib watershed of the Upper Blue Nile Basin; Woldeyohannes et al. (2020) in Abay–Chamo basin of Southern Ethiopia; Negash et al. (2020) in the Borena landscape of North–Central Ethiopia; Solomon et al. (2019) in Dry Afromontane Forest of Wujig Mahgo Waren Southern Tigray–Northern Ethiopia; Gashaw et al. (2018) in Andassa watershed of the Upper Blue Nile Basin; Tolessa et al. (2017) in Chillimo forest area of Central highlands of Ethiopia; Kindu et al. (2016) in Munessa–Shashemene landscape of the Ethiopian highlands. However, these studies did not show how the ecosystem service values would change in the future in response to the forest and other major land cover dynamics under various scenarios. The estimated ESV was limited because the studies were not supported with advanced ESs models (including ESPI and CA–Markov chain). Therefore, the model-based valuation of ESs provides comprehensive information on historical and future trends, thereby improving the current ESs valuation approaches. However, there is a gap in linking land cover and the value of ecosystem services under different scenarios in Ethiopia's mountainous and intermountain landscapes.

This study will contribute to adding a body of knowledge that fills the existing research gaps of ESs provision by integrating the advanced ESs model. It also predicts the sustainability of ESs provision in HSF based on its historical trends using the Cellular Automata–Markov Chain (CA–Markov) model under different scenarios, and it uses the revised ESs valuation coefficients, which makes a definitive study to depict the sustainability of ESs provision for better conservation and planning. The applied models provide crucial information on historical and future ESs but require many data sets to process and calibrate (Paruelo et al. 2016). This study explored the ecosystem services provision sustainability of HSF under different scenarios from 1984–2030 and Modeled the Ecosystem Services Provision Index in each agroclimatic zone of HSF. Also, this study estimated HSF's ecosystem service values and functions under each service category based on the TEEB (The Economics of Ecosystems and Biodiversity) valuation database and their revised monetary values from 1984–2030. Finally, this study analyzed the coefficient of sensitivity in each ecosystem service and found the main ecosystem service type of HSF. Therefore, this study is noteworthy because it provides comprehensive information on the sustainability of HSF and advances the existing studies carried out worldwide to improve policy in protecting and managing forests. It is also imperative to comprehend the ESs value for livelihood security.

The contribution of this study to the body of knowledge is, therefore: 1) to fill the existing research gaps of ESs provision by integrating advanced ESs models; 2) to predict the sustainability of ESs provision in HSF based on its historical trends using the Cellular Automata–Markov Chain (CA–Markov) model under different scenarios; 3) to use the revised ESs valuation coefficients, which makes a definitive study to depict the sustainability of ESs provision for better conservation and planning. The applied models provide crucial information on historical and future ESs but require many data sets to process and calibrate (Paruelo et al. 2016). The objectives of this study were to i) Explore the ecosystem services provision sustainability of HSF under different scenarios from the period 1984–2030; ii) Model the Ecosystem Services Provision Index in each agroclimatic zone of HSF; iii) Estimate the ecosystem service values and functions of HSF under each service category based on The Economics of Ecosystems and Biodiversity (TEEB) valuation database and their revised monetary values from the period 1984–2030; iv) Analyze the coefficient of sensitivity in each ecosystem service; e) Determine the primary ecosystem service type of HSF. This study is noteworthy because it provides comprehensive information on the sustainability of HSF and advances the existing studies carried out worldwide to improve policy in protecting and managing forests. It is also imperative to comprehend the ESs value for livelihood security.

2.1. Study area

HSF is one of Tigray– Ethiopia's distinctive and protected dry Afromontane forests. It is found about 160 km south of Mekelle, the capital city of Tigray –Ethiopia. The forest is located between 39°30'0'' to 39^°50'0'' E and 12^o30'0'' to 12^o45'0'' N (Fig. 1). The total estimated area of HSF is 502.9 Km² and extends within six different districts of Southern Tigray, namely, Endamehoni 56.3 km² (11.2%), Ofla 151.1 km² (30.1%), Raya Azebo 219.2 km² (43.6%), and Raya Alamata 76.3 km² (15.2%). The altitude of HSF was between 1434 and 3681 meters above sea level. The altitudes of HSF have increased from northeast to southwest directions (Birhane et al. 2019). HSF also has a bimodal rainfall pattern with low rainfall in the short season from February to May and high rainfall in the main rainy season from June to September. The study area receives up to 1128 mm of mean annual rainfall but is very erratic and short in the lowland area. The monthly minimum and maximum temperatures were 8.5°C in October and 6.5°C in December (Aynekulu, 2011). The same author has also reported that the temperature of the study area reaches up to 27.9°C in February and drops to 24.5°C in June. The study area comprises Qh (i.e., undifferentiated alluvial, lacustrine and beach sediments) that covers 133.1 km² (26.5%) and P2a (Ashangi formation with deeply weathered alkaline & transitional basalt flows with rare intercalations of tuff, often tilted) covers 351.1 km² (70.6%). Lake Hashenge's geological formation covered the smallest part of HSF within an area of 14.7 km² (2.9%). The major soil types of HSF include Chromic Vertisols 191.95 km² (38.17%), Eutric Cambisols 117.08 km² (23.28%), Eutric Nitisols 44.56 km² (8.86%), Eutric Regosols 71.84 km² (14.28%), Gleyic Cambisols 2.27 km² (0.45%), Leptosols 55.41 km² (11.02%), Pellic Vertisols 3.01 km² (0.6%), and the rest of the area is covered with Lake Hashenge 16.8 km² (3.34%). The uplands of the forest area have extremely shallow to medium soils, while relatively deep soils occur in gently sloping areas and valley bottoms. Moreover, the HSF has nine distinct dominant forest species that help to comprehend more ESs provision. In this regard, the exogenous eucalyptus is the dominant forest species, mainly in the midlands of the study area. However, Juniperus procera, Olea europaea, Ziziphus Spina Christi, Eucalyptus, Tamarindus indica, Carissa edulis, Cordia africana, Acacia spps, and Opuntia ficus Indica are widely distributed in the midlands and lowlands of HSF. The total population of the study area was 64,786 (CSA, 2007). The inhabitants are engaged in crop growing and livestock rearing. The common crops cultivated include teff, maize, barley, wheat, beans, peas, lentils, and sorghum. The farmers in the study area also commonly rear sheep, goats, and cattle.

2.2. Data acquisition

In this study, the Landsat Thematic Mapper (TM) and Operational Land Imager (OLI) images path 168 and row 051 were downloaded from the United States Geological Survey (USGS) http:/earthexplorer.usgs.gov archive during the dry season of Month(s) at a spatial resolution of 30×30 meter (Table 1). The images were cloud-free and acquired during the dry season to reduce the effects of atmospheric errors and improve the interpretation and classification processes of the images. Using a handheld GPS etrex 12–high channel sensitivity, 240 ground truth data from eight different forest and non–forest land cover types were gathered to verify the classified land cover classes for the ESs provision analysis.

Table 1

Characteristics of optical remote sensing products used for the analysis of ESs provision.
Sensor	Bands	Spectral (µm)	Spatial resolution (m)
Landsat 4–5 TM for the periods of 1984, 1995 &2002	Band 1 – Blue	0.45–0.52	30
	Band 2 – Green	0.52–0.60
	Band 3 – Red	0.63–0.69
	Band 4 – Near Infrared	0.77–0.90
	Band 5 – Short–wave Infrared	1.55–1.75
	Band 6 – Thermal Infrared	10.40 − 12.50
	Band 7 – Short–wave Infrared	2.09–2.35
	Band 8 – Panchromatic	0.52–0.90	15
Landsat 8 OLI for the periods of 2013 & 2020	Band 1 – Coastal Aerosol	0.43–0.45	30
	Band 2 – Blue	0.45–0.51
	Band 3 – Green	0.53–0.59
	Band 4 – Red	0.64–0.67
	Band 5 – Near Infrared (NIR)	0.85–0.88
	Band 6 – Short–wave Infrared (SWIR) 1	1.57–1.65
	Band 7 – Short–wave Infrared (SWIR) 2	2.11–2.29
	Band 8 – Panchromatic	0.50–0.68	15
	Band 9 – Cirrus	1.36–1.38	30
	Band 10 – TIRS 1	10.60–11.19	100
	Band 11 – TIRS 2	11.5–12.51

2.3. Data processing and analysis

2.3.1. The historical and future land cover types identification for ESs analysis

Applying– processing techniques is the most crucial step in analyzing any satellite image. This technique is applied before you use the raw images for any analysis because any image can be distorted during its acquisition by atmospheric, radiometric and geometric effects. Therefore, the raw images must be checked and corrected for any distortions pixel by pixel to improve the classification and analysis of the images. The satellite images were used after carefully applying the preprocessing techniques. For instance, we reduced the haze (the darkness of the row images) and enhanced the visibility of each pixel or object using the histogram equalization for visibility of the images and conducting better analysis on the historical and future forest, non-forest classes as shown clearly in Fig. 2. Analysis of forest and non–forest land use classes is crucial in any forest management practice (Birhane et al. 2019) and ESs provision. In detecting, monitoring, and assessing the ESs provision changes in the study area, a post-classification method was used by taking the initial year 1984. Five ESs provision changes were detected between 1984–1995; 1995–2002; 2002–2013; 2013–2020; and finally, computed the prediction changes between 2020–2030.

2.3.2. Quantification of the Ecosystem Services Provision Index (ESPI)

The Ecosystem Services Provision Index (ESPI) can be a feasible and complementary indicator for monitoring ESs because it is a proxy of ecosystem service supply (Jullian et al. 2021). Model-based estimation of ESs has been noted as a robust approach because it provides better information about ESs. We calculated the NDVI value of HSF using Landsat Thematic Mapper (TM) and Operational Land Imager (OLI) sensors to quantify ESPI. NDVI is measured using the visible and near-infrared portions of the electromagnetic spectra. Viña et al. (2011) reported that spectral vegetation indices are mathematical combinations of various spectral bands, typically in the visible and near-infrared portions of the electromagnetic spectrum, used to estimate NDVI and other important information from the remotely sensed data sets. The value of NDVI ranges between − 1 and + 1 (Table 2). The negative values show no–vegetation, while the positive values indicate abundant vegetation. In this study, we used the ESPI to analyze and map the direct and indirect provision of different ecosystem services in HSF pixel by pixel based on the Paruelo et al. (2016) equation in ArcGIS 10.8.1 (Eq. 2):

$$\text{E}\text{S}\text{P}\text{I}= {\text{N}\text{D}\text{V}\text{I}}_{\text{m}\text{e}\text{a}\text{n}}\times (1-{\text{N}\text{D}\text{V}\text{I}}_{\text{c}\text{v}})$$

Where ESPI is Ecosystem Services Provision Index; ${\text{N}\text{D}\text{V}\text{I}}_{\text{m}\text{e}\text{a}\text{n}}$= Mean annual NDVI value; ${\text{N}\text{D}\text{V}\text{I}}_{\text{c}\text{v}}$= the long–term annual NDVI coefficient of variation. The negative ESPI value shows a decrease, while the positive is an increase.

Besides, the mean annual, standard deviation and coefficient of variation (CV) of NDVI value for the entire study period were estimated in ArcGIS 10.8.1 as follows (Eq. 3–5):

$$NDVI\mu =\frac{{\sum }_{i}^{N}{x}_{i}}{N}$$

Where $NDVI\mu$ is the long-term mean annual NDVI value of the study period, ${x}_{i}=$ NDVI value for the study period, $N=$ the number of studies years.

$$\text{N}\text{D}\text{V}\text{I}{\sigma }=\sqrt{\frac{{\sum }_{i}^{N}({x}_{i}-NDVI\mu {)}^{2}}{N}}$$

Where $\text{N}\text{D}\text{V}\text{I}{\sigma }=$The standard deviation of NDVI value for the study period

$${\text{N}\text{D}\text{V}\text{I}}_{\text{c}\text{v}}=\text{N}\text{D}\text{V}\text{I}{\sigma }/NDVI\mu$$

Table 2

Vegetation conditions based on NDVI value range analysis.
NDVI value range	Vegetation condition
0.72–1.00	Very good
0.42–0.72	Good
0.22–0.42	Normal
0.12–0.22	Poor
–0.12–0.12	Very poor

2.3.3. Estimation of Ecosystem Services Provision Value (ESV) and its annual changes

The current ecosystem service value was estimated based on the study area's forest and other land cover types using the modified ecosystem services coefficient suggested by Kindu et al. (2016) (Tables 3 & 4 and Eq. 6).

$$\text{E}\text{S}\text{V}=\sum \left({A}_{K}\times {VC}_{K}\right)$$

Where $\text{E}\text{S}\text{P}=$ The total estimated ecosystem service provision value, ${A}_{K}=$ Area coverage of the forest and other important land cover types in ha, ${VC}_{K}=$ The coefficient values of forest and other land cover types (US$ ha^–1 year^–1).

Table 3

The revised coefficient values of ecosystem service based on the Economics of Ecosystem and Biodiversity valuation

(Source: Kindu et al. 2016)
Land cover	Equivalent biome	Ecosystem Services Coefficient (US$ ha^–1 year^–1)
Forest land (Fl)	Tropical Forest	986.69
Shrub land (Sl)	Grass/rangelands	293.25
Arable land (Al)	Cropland	225.56
Barren land (Bl)	Desert	000.00
Built–up Area (BuA)	Urban/rural	000.00
Grassland (Gl)	Grass/rangelands	293.25
Road (R)	Desert	000.00
Water body (Wb)	Lakes/rivers	8103.50

Table 4

Annual value coefficients of ecosystem service functions for each land cover type based on the Economic value of the Ecosystem and Biodiversity valuation database modified by Kindu et al. (2016)
Ecosystem services functions	Land cover types of the study area based on the optical remote sensing
Ecosystem services functions	Fl	Sl	Al	Bl	BuA	Gl	R	Wb
Provisioning services
Water supply	8	117.45						2117
Food production	32		187.56			117.45		41
Raw material	51.24
Genetic resources	41
Regulating services
Water regulation	6	3				3		5445
Water treatment	136	87				87		431.5
Erosion control	245	29				29
Climate regulation	223	0
Biological control		23	24			23
Gas regulation	13.68	7				7
Disturbance regulation	5
Supporting services
Nutrient cycling	184.4
Pollination	7.27	25	14			25
Soil formation	10	1				1
Habitat/refugia	17.3
Cultural services
Recreation	4.8	0.8				0.8		69
Cultural	2
Total	986.69	293.25	225.56	0	0	293.25	0	8103.5

We also analyzed the annual changes in ecosystem services value as follows (Eq. 7):

$$\%ESV= \frac{{ESV}_{fy}-{ESV}_{iy}}{{ESV}_{iy}} \times 100$$

Where $\%ESV=$ Ecosystem services value in percent, ${ESV}_{fy} and {ESV}_{iy}=$ Ecosystem services value during the final and initial year. The positive $\%ESV$ shows increases, whereas negative values indicate a decrease in forest ecosystem services value.

In addition, we quantified the values of each service given by each ecosystem function to comprehend the impacts of ESV change as follows (Eq. 8):

$${ESV}_{f}=\sum \left({A}_{K}\times {VC}_{fK}\right)$$

A sensitivity analysis was also conducted to analyze the percentage of ESV change in the valuation coefficient as follows (Eq. 9):

$$\text{C}\text{S}= \frac{{(ESV}_{j}-{ESV}_{i})/{ESV}_{i}}{{(VC}_{jk}-{VC}_{ik})/{VC}_{ik}}$$

Where, ${ESV}_{j}$ and ${ESV}_{i}$ = The adjusted and initial total estimated ecosystem services provision value of the forest, ${VC}_{jk}and {VC}_{ik}$ = The adjusted and initial coefficient values of forest and other land cover types (US$ ha^–1 year^–1). According to Kindu et al. (2016), if the CS value is greater than one, the quantified ESV is considered elastic relative to that coefficient or else inelastic (CS value is less than one).

3.1. Ecosystem services of HSF

Forest land (Fl), shrubland (Sl), Arable land (Al), Barren land (Bl), Built-Area (BuA), Grassland (Gl), Road (R), and Water bodies (Wb) are the eight major land cover types that have been identified in this study (Figs. 3a–f). Our findings indicated that the area of HSF has been increasing from 1984–2020 and is expected to increase from 2020–2030. As a result, the community of the study area used HSF for housing construction, farm implements, fuel wood, woody toothbrushes and non–timber products (traditional medicine, honey, grass for livestock feeding, water for human consumption, trough, and conveyance bridge for irrigating farmlands), among others. Even though it was against the law to collect forest products in the study area, illegal wood collecting was frequent, especially at night. Birch et al. (2014) reported similar findings, such as timber for local housing construction, firewood, food, and fodder as the main ESs used by the local people from the Himalayan Forest in Nepal. The forest cover of the area was around 37.1 Km² (7.4%) in the early time of HSF (i.e., 1984), and it later increased to 88.6 Km² (17.6%) in the year 2020 (Figs. 3a & e). This trend is also anticipated to proceed positively from 2020–2030 if the existing community–based forest management activities are in place. The size of the forest will be about 107.6 Km² (21.4%) in 2030, associated with a spatial growth of 19.1 Km² (3.8%) from 2020. In corroboration, Gidey et al. (2017) reported an increase in the forest size by 6 Km² (2.47%) annually, and Birhane et al. (2019) an increase in the size of 71.4 Km² from the year 1985 to 2015. One of the possible reasons for the increase in forest coverage is the intervention that has been widely implemented in the area. Exclosure, land rehabilitation, and participatory forest management programs were vital in sustaining and increasing the forest size. It is also crucial to enrich the livelihood of rural inhabitants because the local communities are protecting and managing forest resources to increase forest cover (Paudyal et al. 2015). However, the global forest lands are projected to decrease annually from 0.13–0.06% until 2030 (Annunzio et al. 2015), while the size of HSF is increasing.

The changes in the extent and composition of forest ecosystems have large impacts on the biophysical conditions, further affecting the supply of forest ecosystem services (Rafik et al. 2018). In 1984, Al, Sl, and Fl controlled the most extensive coverage: 259.2 km², 154.0 km², and 37.1 km², respectively. The Fl, Gl, and Sl have increased by 9.0 km², 3.8 km², and 2.1 km² between 1984 and 1995 (Fig. 3a & b). Conversely, the Al, Bl, and Wb have declined by 13.1 km², 1.8 km² and 0.1 km², respectively. The increased trends in land cover classes were Gl, Fl, and the declining Wb land cover classes agreed with the study of Hishe et al. (2019), who found a similar trend in the Middle Suluh Valley, northern Ethiopia, from 1985–2000. BuA did not show any change for the period 1984 to 1995. In this study period, the HSF expanded substantially to the east of Lake Hashenge, and the shrubland area spatially expanded on the suburb of the state above the forest and the eastern edge of the study area (Fig. 3b). The Al, Fl, R, BuA and Wb showed an increased trend in the period 1995–2002 (Fig. 3b & c). Al and Fl were the dominant land cover classes that increased by 12.4 km² and 11 km² from 1995–2002, respectively. On the contrary, Sl, Gl, and Bl showed a declining trend of 12.9 km², 9.2 Km², and 2.9 km² between 1995 and 2002, respectively. Similar declining trends of Sl and Gl were reported by Gashaw et al. (2017) in the Andassa watershed of Blue Nile Basin, Ethiopia, in the period 2000–2015 and by Hishe et al. (2019) in the Middle Suluh Valley, northern Ethiopia in the period of 2000–2015. Fl, Al, BuA, Wb, and R land covers showed an increasing trend from 2002–2013. Fl has increased by 12.6 km², Al by 2.4 km², BuA by 3.3 km², Wb by 0.4 km² and R by 0.2 km² (Fig. 3c & d). Similar increasing trends of Fl and Al were reported by Sewnet (2016) in the Infraz watershed study of northwestern Ethiopia. HSF's protection and management improved the forestland density and coverage. On the other hand, Sl, Bl, and Gl have declined by 17.3 Km², 1.3 Km², and 0.3 Km², respectively, from 2002 to 2013. In this period, the Fl has increased from 69.7 Km² to 88.6 Km², i.e., by 18.9 Km². This was due to the attention given to the HSF conservation. The forestland substantially expanded from the south to the north tip along the state forest belt (Fig. 3e). Al and Gl cover classes slightly increased by 1.7 km² and 0.9 km², respectively. Small patches of BuA (2.4 km²) have also increased in the period 2013–2020 as shown in the map by the yellow color (Fig. 3e). On the contrary, Sl class has significantly declined by 23.2 km² and this trend was in agreement with the study of Gebrelibanos & Assen (2015) where mainly non–forest land cover improved into forest category.

3.2. Ecological Sustainability of ESs in HSF from the year 2020–2030

Based on Table 5, we assessed the sustainability of ESs provision by predicting the 2030 forest cover and its expansion status (Fig. 3f). The Fl and BuA are predicted to increase by 19.0 km² and 3.2 km², respectively. BuA will concentrate around Lake Hashenge in both the south and north direction, and some are distributed inside the arable land (Fig. 3f). The Wb remains unchanged in its area (14.3 km²). The computed transitional probability matrix indicated how the study area's forests and other important cover types would be altered. Table 5 shows the probability of forest and non–forest transitional probability matrix from the period 2013–2020 for predicting the future ESs provision sustainability. Fl of the study area remained as Fl by 84% due to implementing a forest management program, but the remaining 16% was altered to Sl and Gl. Birhane et al. (2019) found that forest land expanded at the expense of shrublands between 1985 and 2015 in the HSF due to afforestation. The Sl also persisted as Sl by 52%, but the remaining 48% was changed into Al (19%), Fl (14%), and R (7%). Al was continued as Al by 82% next to Fl. This is probably due to an increased population who need land for agricultural cultivation as a matter of survival. The remaining 18% of Al was changed into BuA (7%), Sl (6%), Gl & R (2% each), and Bl (1%). The transition probability matrix also shows Bl had a lesser probability (13%) to persist as Bl because the community of the study area was aggressively working on diminishing Bl and increasing land productivity during the last few decades. As a result, 87% of the Bl was converted into Sl (32%), Al (27%), Gl (14%), BuA (8%), and R (6%). The BuA had also continued as BuA by 18% due to illegal settlement and other related cases. The study area has scattered BuA constructed by the local community. However, the government was discouraged from settling inside the forest land for security and sustainability reasons of the forest. The remaining 82% of BuA shifted into Bl (53%), Sl (12%), Al (11%), Gl and R (3% each). Moreover, the Gl continued as Gl by 67%. The remaining 33% were shifted into Bl (17%), Al (8%), Sl (3%), R (3%), BuA (1%) and Wb (1%). The study area's Wb had a high probability of remaining as Wb by 100%. The R also persisted as R by 86%. The rest of the 14% shifted to other non–forest cover types such as BuA (11%) and Gl (3%).

Table 5

Forest and non–forest transitional probability matrix from the period 2013–2020
LC	Fl	Sl	Al	Bl	BuA	Gl	R	Wb
Fl	0.84	0.07	0.00	0.00	0.00	0.09	0.00	0.00
Sl	0.14	0.52	0.19	0.02	0.03	0.03	0.07	0.00
Al	0.00	0.06	0.82	0.01	0.07	0.02	0.02	0.00
Bl	0.00	0.32	0.27	0.13	0.08	0.14	0.06	0.00
BuA	0.00	0.12	0.11	0.53	0.18	0.03	0.03	0.00
Gl	0.00	0.03	0.08	0.17	0.01	0.67	0.03	0.01
R	0.00	0.00	0.00	0.00	0.11	0.03	0.86	0.00
Wb	0.00	0.00	0.00	0.00	0.00	0.00	0.00	1.00

3.3. Validation and verification of the predicted ecosystem services

We used the actual and simulated forest and non–forest cover types in 2020 to verify and ensure our model's correctness in this study area (Table 6). Our results indicated that Fl has aligned 93.1% with the actual Fl area coverage. This coverage shows a reduction of 6.1 km² (1.21%) in the model. In addition, BuA and Al were aligned almost 100% with their actual area coverage (up to 0.1 km² increase). Bl and Gl were also aligned with their actual area coverage by 78.9 and 79.1%, with a reduction of 0.4 km² and 4.5 km², respectively. The model also estimated an increase of 5.4 km² (1.07%) in Sl and 0.2 km² (0.04%) in R. The model generally simulated 96.3% agreement with HSF's actual forest and non–forest area coverage.

Table 6

Model validation and verification are based on the study's actual and simulated 2020 forest and non–forest cover types.
LC	Actual area coverage of 2020		Simulated area coverage of 2020		Change of area coverage between the actual & simulated (±)
LC	km²	%	km²	%	km²	%
Fl	88.6	17.6	82.5	16.4	–6.1	1.21
Sl	102.7	20.4	108.1	21.5	+ 5.4	1.07
Al	261.8	52.1	261.9	52.1	+ 0.1	0.02
Bl	1.9	0.38	1.5	0.3	–0.4	0.08
BuA	9.9	1.97	9.9	1.97	0.0	0.00
Gl	22.4	4.45	17.9	3.56	–4.5	0.89
R	1.3	0.26	1.5	0.3	+ 0.2	0.04
Wb	14.3	2.84	14	2.78	–0.3	0.06
Total	502.9	100	502.9	100	–	–

3.4. Ecosystem Services Provision Index (ESPI) across each ACZ

Figure 4 shows the ESPI of the study area ranging from − 0.51 to 0.38. The lowlands (Kolla) of the study area, which covered about 51 km² (10.14%) and characterized by a lower slope gradient, have shown a decreased ESPI value than the midlands (Weyna Dega) 267.8 km² (53.25%), highlands (Dega) 180.7 km² (35.93%) and subalpine (Wurch) 3.4 km² (0.68%), respectively. The minimum, maximum, mean, and standard deviation of ESPI in the lowland ACZ were − 0.3, 0.18, − 0.20, and 0.06, respectively. In the midlands, the ESPI value was − 0.46, 0.38, − 0.14, and 0.09, respectively. Similarly, we found ESPI values of − 0.51, 0.37, − 0.15 and 0.14 in the highlands. Moreover, we explored ESPI values of − 0.26, 0.29, − 0.05, and 0.07 for the subalpine ACZ of the study area. ESPI was better in the study area's midlands and subalpine agroclimatic zones.

3.5. Ecosystem Services Valuation (ESV)

The total ecosystem services value showed an increased trend over the different periods. Initially, it was a total of US$ 26.27 million ha^–1year^–1 in 1984 and later increased to US$ 26.92 million ha^–1year^–1 in the second period of 1995. Similarly, in the third period (2002), it increased to US$ 27.91 million ha^–1year^–1. The same is true, there was a slight increment in the fourth period (in 2013) to US$ 29.10 million ha^–1year^–1, and the trend continues with a slight increase to US$ 30.69 million ha^–1year^–1 and finally predicted to slightly increase to US$ 31.96 million ha^–1year^–1 by the year 2030 (Fig. 5 and Tables 7 & 8). ESV of HSF was contrary to the trends found by Kindu et al. (2016), who reported a decrease in the total ecosystem services value observed from 1973–2012 in the Munessa–Shashemene landscape of the Ethiopian highlands. Besides, Solomon et al. (2019) found an increased trend in the total ESV in the Wujig Mahgo Waren forest from 16.6 million US$ ha^–1 year^–1 in 1985 to 18.1 million US$ ha^–1 year^–1 in 2016. In the year 2000, the ESV of Wujig Mahgo Waren forest was 19 million US$ ha^–1year^–1. However, a slight reduction was observed in the area. Therefore, the increased trend of ESV in Wujig Mahgo Waren forest is inconsistent with HSF due to the intervention and management activities.

In HSF, during the initial period of 1984, the contributions of land cover classes to the total ESV ranked with the highest proportion by Wb (40.7%) followed by Al, Sl, Fl, and Gl. They accounted for about 25.2%, 17.2%, 13.9%, and 3.0% of the total ESV. In the final period of 2020, the contribution patterns changed, and Fl (28.5%) became the second next to Wb (37.8%), while Al, Sl, and Gl contributed 21.8%, 9.8%, and 2.1% of the total ESV, respectively. The main reason for becoming the Fl as the second contributor was the expansion of forest area cover from 3710 ha in 1984 to 8860 ha in 2020. The sequences of land cover classes contributing to the ESV in the prediction year of 2030 were like that of the final year of 2020. The shares were 36.3% for Wb, 33.2% for Fl, 20.8% for Al, 7.9% for Sl and 1.8% for Gl of the total ESV. In general, our findings revealed that the ESV for the forestland cover class has consistently increased from 13.9% in 1984 to 28.5% in 2020 and is still to increase to 33.2% in 2030. This trend was contrary to the result found by Kindu et al. (2016) that natural forests declined in size by half from about 21% in 1973 to 9% in 2012 in their study area. One possible factor for reducing natural forests could be poor community–based forest land management in their area. Therefore, the ESV of HSF is better if the existing community–based forest management practices continue.

Table 7

Ecosystem services value from the period 1984–2030 in a million (US$ ha^–1year^–1)
LC	ESV* 1984	%	ESV* 1995	%	ESV* 2002	%	ESV* 2013	%	ESV* 2020	%	ESV* 2030	%
Fl	3.7	13.9	4.6	16.9	5.6	20.2	6.9	23.6	8.7	28.5	10.6	33.2
Sl	4.5	17.2	4.6	17.0	4.2	15.0	3.7	12.7	3.0	9.8	2.5	7.9
Al	6.6	25.2	6.3	23.4	6.6	23.7	6.7	22.9	6.7	21.8	6.6	20.8
Bl	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
BuA	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Gl	0.8	3.0	0.9	3.3	0.6	2.2	0.6	2.1	0.7	2.1	0.6	1.8
R	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
Wb	10.7	40.7	10.6	39.4	10.9	38.9	11.3	38.7	11.6	37.8	11.6	36.3
Total	26.27	100	26.92	100	27.91	100	29.10	100	30.69	100	32.0	100
ESV* = Ecosystem services value

The relationship between ESV of Fl and Fl area expansion was positive (R² = 0.996) and statistically significant (p < 0.01). The positive correlation indicates that when Fl increased, the ESV of Fl tended to increase. Besides, the relationship between ESV of Sl and Sl area was positive (R² = 0.998) and statistically significant (p < 0.01). The positive correlation indicates that when Sl increases, the ESV of Sl tends to increase. The relationship between the ESV of Al and Al area was positive (R² = 0.966) and statistically significant (p < 0.01). The positive correlation indicated that when the Al trend increases, the ESV of Al tends to increase. The relationship between ESV of Gl and Gl area is positive (R² = 0.969) and statistically significant (p < 0.01). The positive correlation indicates that when Gl increases, the ESV of Gl tends to increase. Furthermore, the relationship between ESV of Wb and Wb area was positive (R² = 0.997) and statistically significant (p < 0.01). The positive correlation indicated that when Wb increases, the ESV of Wb tends to increase.

The ESV and area coverage over different periods are depicted in Fig. 5a–e. The ESV of Sl slightly increased from US$ 4.5 million in 1984 to US$ 4.6 million in 1995 (Fig. 5b). On the contrary, the trend of ESV for Sl declined in the proceeding study periods. For example, from 1995 to 2002, US$ 0.4 million was reduced. Similarly, US$ 0.5 million and US$ 0.7 million of ESV decreased from 2002 to 2013 and 2013 to 2020, respectively. For the same land cover type, 2030, it is expected to decrease from US$ 3.0 million to US$ 2.5 million. Among the other land cover types, Sl has shown a very slight decrement (US$ 0.3 million) of ESV from the period 1984 to 1995 and in the preceding periods with no change and an increment of US$ 0.1 million (Fig. 2c). The smallest ESV was observed for the Gl land cover type in each study period (Table 7 and Fig. 2d). In the first period from 1984 to 1995, Gl increased by US$ 0.1 million and again dropped by US$ 0.3 million from the year 1995 to 2002. It again increased by US$ 0.1 million by 2020, and there was no change for 2030 (Fig. 5d). The slight fluctuation in Lake Hashenge in the study area also showed a slight variation in EVS. Accordingly, from 1984 to 1995, US$ 0.1 million of ESV declined, and US$ 0.3 million and US$ 0.4 million increased in 2002 and 2013, respectively (Fig. 5e).

3.6. Ecosystem services value change

The ecosystem services value of HSF has been consistently increasing at different amounts from 1984 to 2030 (Table 8). For instance, total ESV has increased by 0.6 million US$ ha^–1year^–1 between 1984 and 1995. This value also increased to 1.0 million US$ ha^–1year^–1 in 1995–2002. Similarly, from 2002–2013, the ESV escalated to 1.2 million US$ ha^–1 year^–1. The highest ESV increase (1.6 million US$ ha^–1 year^–1 ) was also observed from 2013–2020. The predicted ESV of 2020–2030 in HSF is 1.3 million US$ ha^–1 year^–1. This has shown a slight decrease from the years 2013–2020. Kindu et al. (2016) reported a decrease of 19.4 million US$ ha^–1 year^–1 total ESV between 1973–2012 in the Munessa–Shashemene landscape of the Ethiopian highlands. For ESs of HSF, the Fl increased from 0.9 million US$ ha^–1year^–1 in the period 1984–1995 to 1.1 million US$ ha^–1year^–1 in 1995–2002, 1.2 million US$ ha^–1year^–1 in the year 2002–2013, 1.9 million US$ ha^–1year^–1 between the period 2013–2020 & 2020–2030, respectively. However, a decrease of 0.92 million US$ ha^–1year^–1 in Fl was reported by Biedemariam et al. (2022) from the period 1985–2015 in the Andassa watershed of the Upper Blue Nile Basin.

In HSF, the Sl increased by 0.1 million US$ ha^–1year^–1 in 1984–1995. However, it reduced to − 0.4 million US$ ha^–1 year^–1 in 1995–2002, − 0.5 million US$ ha^–1 year^–1 in 2002–2013, − 0.7 million US$ ha^–1 year^–1 between 2013–2020. The reduction of Sl has been common in most parts of the country due to the high demand for arable land expansion, charcoal production, fuelwood collection, livestock grazing and other uses by the local community. For instance, Biedemariam et al. (2022) reported a reduction of 6.3 million US$ ha^–1 year^–1 between 1985–2015 in the Andassa watershed of the Upper Blue Nile Basin. Similarly, the ESV of Sl in HSF is predicted to reduce by − 0.5 million US$ ha^–1 year^–1 from 2020–2030. The Al declined by − 0.3 million US$ ha^–1year^–1 in 1984–1995. Later, the ESV increased by 0.3 million US$ ha^–1 year^–1 between 1995–2002, 0.1 million US$ ha–1 year–1 in 2002–2013, and no change between 2013–2020. Biedemariam et al. (2022) found an increase of 1.86 million US$ ha^–1year^–1 in AL between 1985–2015 in the Andassa watershed of the Upper Blue Nile Basin. The ESV of Al is also predicted to reduce by − 0.1 million US$ ha^–1 year^–1 from 2020–2030. In addition, the GL increased by 0.1 million US$ ha^–1 year^–1 in 1984–1995. However, it declined by − 0.3 million US$ ha^–1year^–1 between 1995–2002. Similarly, Biedemariam et al. (2022) reported a loss of 4.6 million US$ ha^–1 year^–1 between 1985–2015. We have also observed no change in ESV between 2002–2013 and 2013–2020. The ESV of Gl is also predicted to reduce by − 0.1 million US$ ha^–1 year^–1 from 2020–2030. Besides, the Wb was diminished by − 0.1 million US$ ha^–1year^–1 in 1984–1995. However, it increased by 0.2 million US$ ha^–1year^–1 between 1995–2002. Later, the ESV of Wb was escalated by 0.4 million US$ ha^–1 year^–1 and 0.3 million US$ ha^–1 year^–1 between 2002–2013 and 2013–2020, respectively. We have also observed no change in ESV between 2020 and 2030.

Table 8

Change of ecosystem services provision value from the period 1984–2030 in a million US$ ha^–1 year^–1
LC	ESV* 1984–1995	%	ESV* 1995–2002	%	ESV* 2002–2013	%	ESV* 2013–2020	%	ESV* 2020–2030	%
Fl	0.9	24.3	1.1	23.8	1.2	22.0	1.9	27	1.9	21
Sl	0.1	1.4	–0.4	–8.3	–0.5	–12.1	–0.7	–18	–0.5	–16
Al	–0.3	–5.0	0.3	5.0	0.1	0.9	0.0	0	–0.1	–1
Bl	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0	0.0	0
BuA	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0	0.0	0
Gl	0.1	14.0	–0.3	–30.5	0.0	–1.4	0.0	8	–0.1	–10
R	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0	0.0	0
Wb	–0.1	–0.8	0.2	2.3	0.4	3.7	0.3	3	0.0	0
Total	0.6	33.9	1.0	–7.7	1.2	13.1	1.6	20	1.3	–6
ESV* = Ecosystem Services Value

3.7. Ecosystem service sensitivity analysis of HSF

In this study, we found that the estimated coefficient of sensitivity (CS) value in all ecosystem services of HSF was less than one, which is an inelastic and reliable result (Table 9). This finding also agrees with other studies, e.g., Solomon et al. (2019) and Kindu et al. (2016). The CS value of HSF ranged from 0.00 to 0.40. A lower CS value of 0.00 was observed in GL but higher in Wb (0.40) after their value coefficients were adjusted by 50%. The CS values of Fl and Wb were high because of their highest ESs value and area coverage but less in Sl, Al, and Gl despite their area coverage. Similarly, Kindu et al. (2016) reported the highest CS value in natural forests and water bodies. The CS revealed that the ESV estimation was robust despite uncertainties on the modified value coefficients (Kindu et al. 2016).

Moreover, the increasing trend of the Fl coefficient adjusted by 50% affected the ESV of 1984, 1995, 2002, 2013, and 2022 by ± 7.0%, ± 8.0%, ± 10.1%, ± 11.8%, and ± 14.2%, respectively. The Fl is also estimated to be affected by ± 16.6% until 2030. However, the decreasing trend of the Sl value coefficient modified by 50% affected the ESV of 1984, 1995, 2002, 2013, and 2022 by ± 8.6%, ± 8.5%, ± 7.5%, ± 6.3%, and ± 4.9%, respectively. It is also predicted to be affected by ± 4% by 2030. Besides, the shrinking and sometimes increasing trend of the Al value coefficient adjusted by 50% affected the ESV of 1984, 1995, 2002, 2013, and 2022 by ± 12.6%, ± 11.7%, ± 11.8%, ± 11.5%, and ± 10.9%, respectively. It is also modeled to be affected by ± 10.4% until 2030. The increasing and decreasing trends of the Gl value coefficient amended by 50% affected the ESV of 1984, 1995, 2002, 2013, and 2022 by ± 1.5%, ± 1.6%, ± 1.1%, ± 1%, and ± 1.1%, respectively. It is also modeled to be affected by ± 0.9% until 2030. The decreasing trend of the Wb value coefficient altered by 50% affected the ESV of 1984, 1995, 2002, 2013, and 2022 by ± 20.4%, ± 19.7%, ± 19.5%, ± 19.3%, and ± 18.9%, respectively. It is also demonstrated to be affected by ± 18.1% until 2030.

Table 9

Percentage change in the total quantified ecosystem service value and coefficient of sensitivity (CS) after a 50% adjustment of the revised ESV coefficients.
Change of Value Coefficients (VCs)	1984		1995		2002		2013		2020		2030
Change of Value Coefficients (VCs)	%	CS	%	CS	%	CS	%	CS	%	CS	%	CS
Fl VC ± 50%	± 7.0	± 0.1	± 8.5	± 0.2	± 10.1	± 0.2	± 11.8	± 0.2	± 14.2	± 0.3	± 16.6	± 0.3
Sl VC ± 50%	± 8.6	± 0.2	± 8.5	± 0.2	± 7.5	± 0.2	± 6.3	± 0.1	± 4.9	± 0.1	± 4.0	± 0.1
Al VC ± 50%	± 12.6	± 0.3	± 11.7	± 0.2	± 11.8	± 0.2	± 11.5	± 0.2	± 10.9	± 0.2	± 10.4	± 0.2
Gl VC ± 50%	± 1.5	± 0.0	± 1.6	± 0.0	± 1.1	± 0.0	± 1.0	± 0.0	± 1.1	± 0.0	± 0.9	± 0.0
Wb VC ± 50%	± 20.4	± 0.4	± 19.7	± 0.4	± 19.5	± 0.4	± 19.3	± 0.4	± 18.9	± 0.4	± 18.1	± 0.4

3.8. Ecosystem service functions under each service category

We investigated the value of HSF's ecosystem service functions (ESVf) in all categories, and ESVf has increased from 25.5 million US$ ha^–1 year^–1 in 1984 to 29.9 million US$ ha^–1 year^–1 in 2020 (Table 10). This value is projected to increase to 31.18 million US$ ha^–1 year^–1 in 2030 (Table 10). When we look at each service category, the provisioning of ESV_f of HSF was higher than regulating, supporting and cultural services. The mean annual ESV_f of provisioning was estimated at 2.60 million US$ ha^–1 year^–1 during the year 1984 to 2020, regulating 2.12 million US$ ha^–1 year^–1, supporting 0.5 million US$ ha^–1 year^–1, and cultural 0.6 million US$ ha^–1 year^–1. However, the ESV_f of each service type will be expected to increase to 2.61 million US$ ha^–1 year^–1 provisioning, 2.18 million US$ ha^–1 year^–1 regulating, 0.54 million US$ ha^–1 year^–1 supporting, and 0.07 million US$ ha^–1 year^–1 cultural until the period 2030.

In addition, the mean annual ESV_f of food production, listed under the ESs provisioning, is estimated at 4.6 million US$ ha^–1 year^–1. This value was higher than other ESs provisioning categories, followed by water supply, which was 3.75 million US$ ha^–1 year^–1, then raw material 0.35 million US$ ha^–1 year^–1 and genetic resource 0.28 million US$ ha^–1 year^–1. Therefore, we can easily understand that Ess's provision of HSF was improving the availability of food (for both humans and livestock) in the community of the study area due to the implementation of integrated and community–based forest management practices. Similarly, the mean annual ESV_f of water regulation grouped under the regulating ESs was 7.6 million US$ ha^–1 year^–1, water treatment 2.83 million US$ ha^–1 year^–1, erosion control 2.1 million US$ ha^–1 year^–1, climate regulation 1.51 million US$ ha^–1 year^–1, biological control 0.97 million US$ ha^–1 year^–1, gas regulation 0.2 million US$ ha^–1 year^–1, and disturbance regulation 0.03 million US$ ha^–1 year^–1. Moreover, the mean annual ESV_f of nutrient cycling clustered under the supporting ESs was estimated at 1.25 million US$ ha^–1 year^–1, pollination 0.72 million US$ ha^–1 year^–1, soil formation 0.08 million US$ ha^–1 year^–1, habitat 0.12 million US$ ha^–1 year^–1. The mean annual ESV_f of recreation grouped under the cultural ESs was quantified as 0.12 million US$ ha^–1 year^–1, and cultural 0.01 million US$ ha^–1 year^–1.

Table 10

The annual estimated value of ecosystem functions under each service category from 1984–2030 in a million US$ ha^–1 year^–1.
Ecosystem services	ESV_f ₁₉₈₄	ESV_f ₁₉₉₅	ESV_f ₂₀₀₂	ESV_f ₂₀₁₃	ESV_f ₂₀₂₀	ESV_f ₂₀₃₀	Overall ESV_f statistics from 1984–2030
Ecosystem services	ESV_f ₁₉₈₄	ESV_f ₁₉₉₅	ESV_f ₂₀₀₂	ESV_f ₂₀₁₃	ESV_f ₂₀₂₀	ESV_f ₂₀₃₀	Df	Min	Max	Mean	Stdev	CV
Provisioning services
Food production	5.35	5.17	5.33	5.19	5.52	5.51	0.16	5.17	5.52	4.60	0.15	0.03
Water supply	4.63	4.64	4.56	4.47	4.30	4.13	–0.51	4.13	4.64	3.75	0.20	0.05
Raw material	0.19	0.24	0.29	0.36	0.45	0.55	0.36	0.19	0.55	0.35	0.14	0.39
Genetic resources	0.15	0.19	0.23	0.29	0.36	0.44	0.29	0.15	0.44	0.28	0.11	0.39
Regulating services
Water regulation	7.26	7.22	7.72	7.65	7.88	7.88	0.62	7.22	7.88	7.60	0.29	0.04
Water treatment	2.64	2.81	2.78	2.82	2.91	3.01	0.36	2.64	3.01	2.83	0.12	0.04
Erosion control	1.43	1.67	1.88	2.13	2.53	2.95	1.51	1.43	2.95	2.10	0.56	0.27
Climate regulation	0.83	1.03	1.27	1.55	1.98	2.40	1.57	0.83	2.40	1.51	0.59	0.39
Biological control	1.04	1.02	1.00	0.96	0.92	0.87	–0.17	0.87	1.04	0.97	0.06	0.07
Gas regulation	0.18	0.19	0.19	0.20	0.21	0.22	0.04	0.18	0.22	0.20	0.02	0.08
Disturbance regulation	0.02	0.02	0.03	0.03	0.04	0.05	0.03	0.02	0.05	0.03	0.01	0.36
Supporting services
Nutrient cycling	0.68	0.85	1.05	1.28	1.63	1.98	1.30	0.68	1.98	1.25	0.49	0.39
Pollination	0.84	0.84	0.81	0.39	0.74	0.71	–0.13	0.39	0.84	0.72	0.17	0.24
Soil formation	0.06	0.06	0.07	0.08	0.10	0.12	0.06	0.06	0.12	0.08	0.02	0.28
Habitat	0.06	0.08	0.10	0.12	0.15	0.19	0.12	0.06	0.19	0.12	0.05	0.39
Cultural services
Recreation	0.12	0.04	0.13	0.14	0.15	0.16	0.04	0.04	0.16	0.12	0.04	0.36
Cultural	0.01	0.01	0.01	0.01	0.02	0.02	0.01	0.01	0.02	0.01	0.01	0.39
Total	25.50	26.09	27.48	27.70	29.90	31.18	5.68	24.07	32.01	26.53	3.04	4.17
Df = Ecosystem service functions difference from 1984–2030

The HSF generally provides higher ecosystem services in regulating than provisioning, supporting, and cultural (Fig. 6) due to its significant roles in improving and treating water, climate, and erosion control from 1984 to 2020. The regulating ESs of HSF would still contribute a better service until 2030 than others.

Forests have been providing multiple ecosystem services globally. Conversely, its spatial coverage has been diminishing from time–to–time due to natural and anthropogenic factors. It is, therefore, crucial to investigate its sustainability to ensure the well–being of local communities. The size of the Hugumburda dry Afromontane forest increased from 1984–2020 and is expected to increase in 2030, unlike the global reduction in the forest area. HSF area coverage increased from 37.1 km² in 1984 to 88.6 km² in 2020 and is predicted to increase to 107.6 km² by 2030 if the existing forest management practices are in place. Therefore, the provision of forest ecosystem services increased from 3.7 million US$ ha^–1 year^–1 in 1984 to 10.6 million US$ ha^–1 year^–1 in 2030. Thus, it is expected to improve the livelihood of the local community. This research is notable for providing comprehensive information on the valuation of forest ecosystem service in the study area and enriching existing activities in protecting and managing forest resources.

Acknowledgments

Mekelle University supported this research, NORHED project "Steps toward Sustainable Forest Management with the Local communities in Tigray, Northern Ethiopia (ETH 13/0018) and NORAD IV, TRECCAfrica II, Open Society Foundation–Africa Climate Change Adaptation Initiative (OSF–ACCAI) projects of Mekelle University. We thank the University of Witwatersrand and the United States Geological Survey (USGS). We are also indebted to the community of the study area for their invaluable support during our fieldwork activities.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose.

Author Contributions Statement

E.G wrote the main manuscript text. M.G prepared Fig. 3a–f supported by E.G. E.G prepared the remaining Figures. E.G., P. M., E.B., S.H., A.Z., T.T., M.M.R., A.G., S.T., O.D., R.L., T.W., M.G., and J.N have critically reviewed the manuscript.

Adhikari, K., & Hartemink, A. E (2016) Linking soils to ecosystem services –A global review. Geoderma 262:101–111. http://dx.doi.org/10.1016/j.geoderma.2015.08.009
Adhikari S., Baral H., Nitschke C (2018) Adaptation to Climate Change in Pan chases Mountain Ecological Regions of Nepal. Environments 5:42.
Akira M., Kenneth P., Lena G (2017) Forest biodiversity and ecosystem services Biodiversity and ecosystem services in forest ecosystems. Applied Ecology 54: 12–27.
Anley, M.A.; Minale, A.S.; Haregeweyn, N.; Gashaw, T. 2022 Assessing the impacts of land use/cover changes on ecosystem service values in Rib watershed, Upper Blue Nile Basin, Ethiopia. Trees For People 7:100212.
Aynekulu, E., Aerts, R., Moonen, P., Denich, M., Gebrehiwot, K., Vågen, T–G., Mekuria, W., Boehmer, H.J (2012) Altitudinal variation and conservation priorities of vegetation along the Great Rift Valley escarpment, Northern Ethiopia. Biodivers Conserv 21: 2691–2707.
Baral H., Jaung W., Bhatta L., Phuntsho S., Sharma, S., Paudyal K., Zarandian A., Sears R., Sharma R., Dorji T (2018) Approaches and Tools for Assessing Mountain Forest Ecosystem Services. Centre for International Forestry Research: Bogor, Indonesia 17.
Biedemariam, M., Birhane, E., Demissie, B., Tadesse, T., Gebresamuel, G., Habtu, S (2022) Ecosystem Service Values as Related to Land Use and Land Cover Changes in Ethiopia: A Review. Land 11:2212. https://doi.org/10.3390/land11122212
Birch, J.C., Thapa, I., Balmford, A., Bradbury, R.B., Brown, C., Butchart, S.H.M., Gurung, H., Hughes, F.M.R., Mulligan, M., Pandeya, B., Peh, KS–H., Stattersfield, A.J., Walpole, M., Thomas, D.H.L 2014) What benefits do community forests provide, and to whom? A rapid assessment of ecosystem services from a Himalayan Forest, Nepal. Ecosyst.Serv 8:118–127.
Birhane, E., Ashfare, H, Fenta, A. A, Hishe, H., Gebremedhin, M.A, G. wahed, H., Solomon, N (2019) Land use land cover changes along topographic gradients in Hugumburda national forest priority area, Northern Ethiopia, Remote Sensing Applications: Society and Environment 13:61–68. https://doi.org/10.1016/j.rsase.2018.10.017
Brown, T. C., Bergstrom, J. C., & Loomis, J. B (2007) Defining, valuing, and providing ecosystem goods and services. Natural Resources 47: 329–376.
Caio C., Araujo B., Peter A., John D (2015) A Global Environmental Change and Earth Observation Research Group. Ecological Indicators 52:430–443.
Central Statistical Agency (CSA) (2007) Population and Housing Census Report–Country–2007, Central Statistical Agency, 2010–07, Ethiopia. http://www.csa.gov.et/index.php/2013-02-20–14-51-51/2013-04-01–11-53-00/census–2007
d’Annunzio, R., Sandker, M., Finegold, Y., Min, Z (2015). Projecting global forest area towards 2030. Forest Ecology and Management 352:124–133. http://dx.doi.org/10.1016/j.foreco.2015.03.014
Euliss N., Smith L., Liu S (2010) The need for simultaneous evaluation of ecosystem services and land use change. Environmental Science and Technology 44 (20):7761–7763.
FAO (2015) Terms and definitions, Forest Resources Assessment Working Paper (http://www.fao.org/docrep/017/ap862e/ap862e00.pdf).
Gashaw, T., Tulu, T., Argaw, M., Worqlul, A.W., Tolessa, T., Kindu, M (2018) Estimating the impacts of land use/land cover changes on Ecosystem Service Values: The case of the Andassa watershed in the Upper Blue Nile basin of Ethiopia. Ecosyst. Serv 31:219–228.
Gashaw, T., Tulu, T., Argaw, M., & Worqlul, A. W (2017) Evaluation and prediction of land use/land cover changes in the Andassa watershed, Blue Nile Basin, Ethiopia. Environmental Systems Research 6(1): 1–17. https://doi.org/10.1186/s40068-017-0094–5
Gebrelibanos, T., & Assen, M (2015) Land use/land cover dynamics and their driving forces in the Hirmi watershed and its adjacent agro–ecosystem, highlands of Northern Ethiopia. Journal of Land Use Science 10(1): 81–94. https://doi.org/10.1080/1747423X.2013.845614
Gidey, E., Dikinya, O., Sebego, R., Segosebe, E., & Zenebe, A (2017) Cellular Automata and Markov Chain (CA_Markov) model–based predictions of future land use and land cover scenarios (2015–2033) in Raya, northern Ethiopia. Modeling Earth Systems and Environment 3(4):1245–1262. https://doi.org/10.1007/s40808-017-0397–6
Hishe, S., Bewket, W., Nyssen, J., & Lyimo, J (2019) Analysing past land use land cover change and CA–Markov–based future modelling in the Middle Analysing past land use land cover change and CA–. Geocarto International 1–31. https://doi.org/10.1080/10106049.2018.1516241
Jullian, C., Nahuelhual, L., Laterra, P (2021) The Ecosystem Service Provision Index as a generic indicator of ecosystem service supply for monitoring conservation targets, Ecological Indicators 129: 107–855. https://doi.org/10.1016/j.ecolind.2021.107855
Kidane L., Tamrat B., Sileshi N (2010) Vegetation Composition in Hugumbirda–Gratkhassu National Forest Priority Area, South Tigray. Momona Ethiopian Journal of Science 2: 27–48.
Kidane, L (2015) Vegetation ecology and plant diversity of Hugumburda–Gratkhassu National Forest Priority Area, Northern Ethiopia: habitat fragmentation in time and space, Ph.D. Dissertation Addis Ababa University.
Kindu, M., Schneider, T., Teketay, D., Knoke, T (2016) Changes of ecosystem service values in response to land use/land cover dynamics in Munessa–Shashemene landscape of the Ethiopian highlands. Science of the Total Environment 547:137–147, http://dx.doi.org/10.1016/j.scitotenv.2015.12.127
Koebele, E. A., Crow, D. A., & Albright, E. A (2020) Building Resilience during Recovery: Lessons from Colorado' s Watershed Resilience Pilot Program. Environmental
Koschke, L., Fürst, C., Frank, S., Makeschin, F (2012) A multi–criteria approach for an integrated land–cover–based assessment of ecosystem services provision to support landscape planning. Ecological Indicators 21 (2012):54–66. doi: 10.1016/j.ecolind.2011.12.010
Lambert, T., Bouillon S., Darchambeau F., Morana C., Roland F., Descy J (2017) Effects of human land use on the terrestrial and aquatic sources of fluvial organic matter in a temperate river basin (the Meuse River, Belgium). Biogeochemistry 136 (2): 1–21.
Martínez–Harms, M. J. & Balvanera, P (2012) Methods for mapping ecosystem service supply: a review, International Journal of Biodiversity Science, Ecosystem Services & Management 8, (1–2):17–25. DOI: 10.1080/21513732.2012.663792
Millennium Ecosystem Assessment (MEA) (2005) Ecosystems and human well–being: synthesis. Washington DC: Island Press.
Miura, S., Amacher, M., Hofer, T., San–Miguel–Ayanz, J., Ernawati, Thackway, R (2015) Protective functions and ecosystem services of global forests in the past quarter–century, Forest Ecology and Management 352:35–46. http://dx.doi.org/10.1016/j.foreco.2015.03.039
Mouchet, M. A., Rega, C., Lasseur, R., Georges, D., Paracchini, M.–L, Renaud, J., Stürck, J., Schulp, C. J. E., Verburg, P. H., Verkerk, P. J. & Lavorel, S (2017) Ecosystem service supply by European landscapes under alternative land–use and environmental policies, International Journal of Biodiversity Science, Ecosystem Services & Management 13:1, 342–354. DOI: 10.1080/21513732.2017.1381167
Ncube, S, Spray, C., & Geddes, A (2018) Assessment of changes in ecosystem service delivery – a historical perspective on catchment Landscapes.International Journal of Biodiversity Science, Ecosystem Services & Management 14:145–163. DOI: 10.1080/21513732.2018.1489306
Negash, E. Getachew, T., Birhane, E., Gebrewahed, H (2020) Ecosystem Service Value Distribution Along the Agro–ecological Gradient in North–Central Ethiopia. Earth Systems and Environment. https://doi.org/10.1007/s41748-020-00149–7
Pandey, S., Cockfield G., Maraseni T (2016) Assessing the roles of community forestry in climate change mitigation and adaptation; Nepal. Ecological Management 360:400–407.
Parsa,V., Yavari A., Nejadi A (2016) Spatio–temporal analysis of land use/land cover pattern changes in Aras Baran Biosphere Reserve; Iran. Modeling Earth Systems and Environment 2 (4): 178.
Paruelo, J. M., Texeira, M., Staiano, L., Mastrángelo, M., Amdan, L., Gallego, F (2016) An integrative index of Ecosystem Services provision based on remotely sensed data, Ecological Indicators 71: 145–154. http://dx.doi.org/10.1016/j.ecolind.2016.06.054
Paudyal, K., Baral, H., Burkhard, B., Bhandari, S. P., Keenan, R. J (2015) Participatory assessment and mapping of ecosystem services in a data–poor region: Case study of community–managed Forests in central Nepal. Ecosystem Services 13:81–92. http://dx.doi.org/10.1016/j.ecoser.2015.01.007
Pede, E. (2020) Planning for Resilience New Paths for Managing Uncertainty.
Rafik B., Toufik B.,Abdmalik T., Abbas C (2018) A new spectral index for the extraction of built–up land features from Landsat 8 satellite imagery.
Raymond, C. M., Bryan, B. A., MacDonald, D. H., Cast, A., Strathearn, S., Grandgirard, A., Kalivas, T (2009) Mapping community values for natural capital and ecosystem services, Ecological Economics 68:1301–1315. doi:10.1016/j.ecolecon.2008.12.0067
Romijn, E., Lantican, C. B., Herold, M., Lindquist, E., Ochieng, R., Wijaya, A., Murdiyarso, D., Verchot, L (2015) Assessing change in national forest monitoring capacities of 99 tropical countries, Forest Ecology and Management 352: 109–123. http://dx.doi.org/10.1016/j.foreco.2015.06.003
Solomon, N., Segnon, A. C., & Birhane, E (2019) Ecosystem Service Values Changes in Response to Land–Use/Land–Cover Dynamics in Dry Afromontane Forest in Northern Ethiopia, Int. J. Environ. Res. Public Health 19: 16–4653.doi:10.3390/ijerph16234653
Sewnet, A (2016) Land use/cover change at Infraz watershed by using GIS and remote sensing techniques, northwestern Ethiopia. International Journal of River Basin Management 14(2):133–142. https://doi.org/10.1080/15715124.2015.1095199
Sherrouse, B. C., Clement, J. M., Semmens, D. J. (2011). A GIS application for assessing, mapping, and quantifying the social values of ecosystem services, Applied Geography 31:748–760. doi:10.1016/j.apgeog.2010.08.002
Siebe, J., Pons, M. (2015). Assessment of Urban Ecosystem Services using Ecosystem Services Reviews and GIS–based Tools. Procedia Engineering 115:53–60. doi: 10.1016/j.proeng.2015.07.354
Tittensor, D., Walpole M., Hill S., Boyce D., Britten G., Burgess, N (2014) A mid–term analysis of progress toward international biodiversity targets. Science 346:241–244.
Tolessa, T., Senbeta, F., Kidane, M (2017) The impact of land use/land cover change on ecosystem services in the central highlands of Ethiopia. Ecosyst. Serv 23: 47–54.
Vauhkonen, J (2018) Predicting the provisioning potential of forest ecosystem services using airborne laser scanning data and forest resource maps. Forest Ecosystems 5:24. https://doi.org/10.1186/s40663-018-0143–1
Viña, A., Gitelson, A. A., Nguy–Robertson, A. L., Peng, Yi (2011) Comparison of different vegetation indices for the remote assessment of green leaf area index of crops, Remote Sensing of Environment 115: 3468–3478. doi: 10.1016/j.rse.2011.08.010
Woldeyohannes, A.; Cotter, M.; Biru, W.; Kelboro, G (2020) Assessing changes in ecosystem service values over 1985–2050 in response to land use and land cover dynamics in Abaya–Chamo Basin, Southern Ethiopia. Land 9:37.
Wolff, S., Schulp, C.J.E., & Verburg, P.H (2015) Mapping ecosystem services demand: A review of current research and future perspectives, Ecological Indicators 55: 159–171. http://dx.doi.org/10.1016/j.ecolind.2015.03.016
Yushanjiang, A., Zhang F., Yu H., Kung H (2018) Quantifying the spatial correlations between landscape pattern and ecosystem service value: a case study in Ebinur Lake basin, China. Ecological Engineering 113:94–104.
Zari, M. P., Kiddle, G. L., Blaschke, P., Gawler, S., & Loubser, D (2019) Utilising nature-based solutions to increase resilience in Pacific Ocean Cities. Ecosystem Services, 38(July), 100968. https://doi.org/10.1016/j.ecoser.2019.100968

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Modeling of the Economic Value of Ecosystem Services for Sustainable Resource Management in Northern Ethiopian Highlands

Status:

Version 1

Abstract

Figures

Highlight

1. Introduction

2. Methods

2.1. Study area

2.2. Data acquisition

2.3. Data processing and analysis

2.3.1. The historical and future land cover types identification for ESs analysis

2.3.2. Quantification of the Ecosystem Services Provision Index (ESPI)

2.3.3. Estimation of Ecosystem Services Provision Value (ESV) and its annual changes

3. Results and Discussion

3.1. Ecosystem services of HSF

3.2. Ecological Sustainability of ESs in HSF from the year 2020–2030

3.3. Validation and verification of the predicted ecosystem services

3.4. Ecosystem Services Provision Index (ESPI) across each ACZ

3.5. Ecosystem Services Valuation (ESV)

3.6. Ecosystem services value change

3.7. Ecosystem service sensitivity analysis of HSF

3.8. Ecosystem service functions under each service category

4. Conclusions

Declarations

References

Additional Declarations

Status:

Version 1