All Earth

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/tgda21

Land use land cover change analysis and detection

of its drivers using geospatial techniques: a case of
south-central Ethiopia

Mehari Mariye, Li Jianhua & Melesse Maryo

To cite this article: Mehari Mariye, Li Jianhua & Melesse Maryo (2022) Land use land cover
change analysis and detection of its drivers using geospatial techniques: a case of south-
central Ethiopia, All Earth, 34:1, 309-332, DOI: 10.1080/27669645.2022.2139023

To link to this article: https://doi.org/10.1080/27669645.2022.2139023

© 2022 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group.

Published online: 07 Nov 2022.

Submit your article to this journal

Article views: 2618

View related articles

View Crossmark data

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tgda21
ALL EARTH
2023, VOL. 34, NO. 1, 309–332
https://doi.org/10.1080/27669645.2022.2139023

EARTH OBSERVATIONS

Land use land cover change analysis and detection of its drivers using
geospatial techniques: a case of south-central Ethiopia
Mehari Mariyea, Li Jianhuaa and Melesse Maryob
a
College of Environmental Science & Engineering, Tongji University, Shanghai, China; bEthiopian Biodiversity Institute; home-based in
Kotebe (Metropolitan) University, Addis Ababa, Ethiopia

ABSTRACT ARTICLE HISTORY

The rapid expansion of agriculture and human settlements has simplified natural ecosystems and Received 22 June 2022
harmed the earth’s biodiversity. The current study was conducted in south-central Ethiopia to Accepted 19 October 2022
identify LULC change dynamics, and analysis of their driving force using geospatial technology. KEYWORDS
A supervised maximum likelihood image classification method was employed in combination with Drivers; ArcGIS; LULC;
the visual interpretation of satellite images to categorise and map LULC classes of the study Ethiopia; ERDAS imagine
landscape. Semi-structured interviews, field observations, key informants, and Focus Group
Discussions (FGDs) were employed to identify major driving forces, periodic LULC changes and
impacts. The classification result showed a considerable decline in forestland from (43.1%) in 1973 to
(13.1%) in 2000. Similarly, grasslands declined from (45.5%) in 1973 to (6.3%) in 2018. On the other
hand, cropland has increased from (9.24%) in 1973 to (32.04%) in 1986 likewise between 2000 and
2018 its coverage was augmented from 45.4% to 51.1%, respectively. . Local communities perceived
population growth, settlement, urbanisation, expansion of farmlands, and fuel wood collection as
dominant drivers of LULC changes in the study area in thewatershed. The respondents also
observed that the decline in forest LULC triggered the loss of biodiversity, soil fertility, and water
availability. Hence, local and national regimes must take adequate measures to minimise the rapid
shift in land use and to balance the protection of the human livelihood with the environment.

1. Introduction thousands of people because they can provide food,

medicine, precipitation, (FAO, 2016) freshwater, and
The rapid expansion of agriculture and human settle­
storehouse for carbon sequestration that helps to miti­
ments has simplified natural ecosystems and harmed
gate the effects of global climate change (Belete et al.,
the earth’s biodiversity (Hailu et al., 2020). The integrity
2021). Despite the fact that forests serve vital environ­
of ecosystem products and services would be jeopar­
mental, cultural (Gurmessa, 2015) and socioeconomic
dised if land resources became more endangered and
roles in the survival of life on earth, the quality and
strained (Elias et al., 2019). Consequently, LULC
quantity of forests in Ethiopia are diminishing on
changes have recently been a key concern topic of
a regular basis due to LULC conversions (Keenan
environmental agenda in the scientific research com­
et al., 2015).
munity because of its substantial impacts on global
LULC changes are influenced by a wide range of
climate change (Elias et al., 2019), ecosystem services,
drivers. The expansion of agricultural and settlement
and biodiversity (Phalan et al., 2011) biogeochemical
land, population growth, and charcoal production in
processes (Houghton et al., 2012) socioeconomic and
social well-being (Dwivedi et al., 2007; Mengistu et al., South Central Ethiopia (Kindu et al., 2015), resident
2012), micro-climate conditions, groundwater density in northern China (Peng et al., 2015), expansion
recharge (Lambin & Geist, 2006), and environmental of cultivated land in highlands of northern Ethiopia
modification (Yohannes et al., 2020). LULC is the most (Tsehaye & Mohammed, 2015), and slope inclination
significant risk to the depletion of natural resources, in Southern Wollo (Belete et al., 2021). Furthermore,
especially wetlands, forests, and diverse fauna improper land management, resettlement plans, and
throughout the world that in turn contributes to global rapid population expansion in northwestern lowlands
environmental change (Belete et al., 2021). The con­ (Binyam et al., 2015). Likewise, unfavourable adminis­
version of forest cover to other land-use changes to tration strategies Daniel (2008), Illegal logging, fuel
meet human needs and wants is a major contributor to wood exploitation, and plantation (Yesuph &
environmental degradation (Tesfaye et al., 2014), habi­ Dagnew, 2019) were cited as drivers of LULC in
tat destruction, and ecological imbalance on the sur­ Ethiopia. Population increase, and poverty in Malawi
face of the earth (FAO, 2010). Forests are significant (Asres et al., 2016). Land scarcity, population expan­
elements of biodiversity and vital means of income for sion, and urbanisation in Kenya (Mariye et al., 2021).

CONTACT Li Jianhua [email protected] College of Environmental Science & Engineering, Tongji University, 200092 Shanghai, China
© 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
310 M. MARIYE ET AL.

The conversion of the landscape can be realised by Kembata Tembaro zone (KTz), studies on LULC change
visiting locations on the ground (Mary et al., 2013). and analysing the driving factors are still limited or not
However, these traditional approaches are believed to yet conducted. Therefore, it is very essential to study
be time-intense, tedious, and costly that does not deliver how the community perceives the driving forces and
a comprehensive outcome, whereas remotely sensed realise the complex interdependence of LULC changes
information can provide appropriate data swiftly, and (Meyfroidt, 2013; B Belayneh et al., 2021). The study
examining time-based changes in the LULC dynamics site, Kembata Tembaro, is recognised for its cereal
rapidly and can identify environmental alterations pre­ production, high population, and livestock but the
cisely (Asmamaw et al., 2014; Asokan & Anitha, 2019; area is prone to land degradation, deforestation, high
soil erosion, and has experienced rapid and extensive
Degife et al., 2018; Schaefer & Thinh, 2019; TekleL &
LULC change which might lead to deterioration of the
Hedlund, 2000; Yuan et al., 2005; Zewdie & Csaplovies,
environment and the loss of ecosystem services that
2017) and offer a better accuracy at low cost even over
humans rely on for existence (B Belayneh et al., 2021).
large geographical areas. Several scholars have
Therefore, comprehensive information on the
attempted to use digital satellite image data (Abebe,
dynamics of LULC transition, driving factors, and
2018; Haque & Basak, 2017; Mengistu et al., 2012;
causes of LULC change is necessary to build appropri­
Minale, 2013; Moges et al., 2015; Rahman, 2016; Rawat & ate environmental regulations and suitable land man­
Kumar, 2015; Shawul & Chakma, 2019; Singh, 1989) to agement approaches for the entire study area and
address problems of LULC change detection in various beyond. However, basic data on the trends, magni­
parts around the world. Remote sensing (RS) and geo­ tude, and extent of the LULC modification in the
graphic information systems (GIS) became influential and study landscape have not been addressed so far, and
significant methods for detecting LULC changes at multi­ consequently, the magnitude of the change, its driving
ple spatial scales (Dewan & Corner, 2014). GIS integrates factors, and consequences are poorly understood. In
the information obtained from RS to provide the study landscape, there is no practice of evaluating
a comprehensive knowledge of LULC modelling (Twisa LULC dynamics using a combination of remote sensing
& Buchroithner, 2019). Accordingly, both RS and GIS have and GIS approaches. The KTz in south-central Ethiopia
proven to be quite helpful for detecting LULC patterns is a significant example of a landscape in which it lacks
(Attri et al., 2015; Lu et al., 2010). The remotely sensed data attention to scientific information on LULC changes
provide a substantial connection between the localised existed. Moreover, some decades ago, KTz district
environmental study on conservation and monitoring of was known for its untouched forest LULC cover.
natural resources at several scales (Mariye et al., 2020). However, it is currently facing great human influences
Kotoky et al. (2012) discussed that LULC studies are that have placed the forest into a fragmented state.
becoming increasingly important in various fields, includ­ Dense forest areas were given to landless youth just
disregarding the environmental importance of the for­
ing agricultural development, settlement surveys, effec­
est ecosystem. Such activities are known to affect bio­
tive land use plans, and ecological studies.
diversity (for instance, woody plant species, medicinal
Employing Landsat images to evaluate LULC mod­
plants, and wild animals) more seriously. Furthermore,
ification at the basin and sub-basin level, as well as
this study helps to fill scientific information shortages
identifying the pace and degree of land cover change
in the land management sector of Ethiopia in general
is a significant approach to enhance proper manage­
and particularly in the southern central highlands of
ment of natural resources (Meshesha et al., 2016). The Ethiopia. Therefore, the main objective of the current
assessment of LULC alterations and the factors that study was designed to conduct LULC change detection
trigger these changes are the topic of ongoing scien­ using scientific tools such as GIS and remote sensing
tific inquiry that has piqued the curiosity of a wide through spatial and temporal scales. However, the
range of scientists (Angessa et al., 2019) specific objectives were aimed to (1) determine
Tracking LULC change by using GIS and remote dynamics and extent of LULC transition and (2) assess
sensing technology provides quantitative analyses of the driving factors and undertake remote sensing and
the transformation. However, it fails to describe the GIS supported LULC change analysis over the last
relationship between the driving factors and the 45 years. This study is envisioned to be used by land
cause of their change (Daniel et al., 2018). The majority use planners, environmentalists, decision-makers, and
of previous research on LULC alterations was focused other stakeholders to formulate environmentally
on particular areas, mainly in Northern Ethiopia and in sound policies and planning strategies based on
the central Rift Valley Lake, and only evaluated the a robust understanding of the dynamics of LULC
dynamics of LULC changes using remote sensing data change. It is intended to provide guidelines for main­
(B Belayneh et al., 2021); nevertheless, they did not taining ecosystem services and preserving and utilising
offer explanations on local people's views of the driv­ natural resources in KTz or alternative districts with
ing forces of LULC change (Burgi et al., 2017). In the similar settings.
 ALL EARTH 311

2. Methods and materials from the homesteads are dedicated for the cultivation
of annual crops such as barley, wheat, maize, teff,
2.1. The study area
sorghum, beans and potatoes. ‘Maxxe Sana’ denotes
The study area is situated in Kembata Tembaro district the longer rainy seasons that account for 80% and
(Figure 1) and is located about 350 km southwest of occur between June, July, and August. During this
Addis Ababa (capital of Ethiopia). Geographically, the period, the area receives a total annual average rainfall
district positioned between 71°03ʹ048”–75°00ʹ946” of about 1507 mm. The amount of rainfall is low during
latitude and 37° 341’ 711”–38° 078 ‘387” longitude the dry season that falls between October and
and covers about 44, 883 ha (448.83 km2). The study February. Crops are cultivated using the rainfall of the
landscape has a share of four woredas (The smallest main ‘Maxxe Sana’ and ‘Glalichi Sana’ seasons but the
administrative unit of the country), namely Doyogena, majority of the crop husbandry takes place during the
Angecha, Damiboya, and Kachabira. It is characterised ‘Maxxe Sana’. Consequently, soil loss by water erosion
by undulating mountainous topography, with an ele­ occurs during this period. Winter frost and irregular
vation stretching from 2300 to 3080 m.a.s. rainfall patterns in the summer and spring are the
l. (KTDARDO, 2019). two most significant factors affecting crop production
In the study area, there are two rainy seasons with in the study area. Because of rainfall-dependent farm­
bimodal rainfall. ‘Maxxe Sana’ and ‘Glalichi Sana’ The ing, farmers are always concerned about rainfall inten­
average annual rainfall (Houghton et al.), in the area sity and duration. According to Ethiopian Agro-
ranges from 1200 to 1800 mm. ‘Glalichi Sana’ locally ecological zone classification, 22.25% of the district is
denotes a shorter rainy season that extends from classified as Dega (high land, 2400–3200 metres above
March to May and contributes 19.8% of the total rain­ sea level), 70.75% Woyina Dega (midland, 1800–
fall (RF). In the study area, most farmers possess at least 2400 m. a. s .l), and 7% Kolla (lowland, 500–
two land types: the main land and home gardens. The 1800) m. a. s. l (Maryo, 2020). Based on meteorological
main land is allocated for home construction, cereal data obtained from the nearby Angecha and Durame
cultivation, grazing and tree plantations. Home gar­ stations, the minimum monthly temperature ranges
dens, on the contrary, are used for planting valuable between 12.2°C in 1997 as depicted in (Figure 2),
crops such as Enset (Ensete ventricosum), vegetables, whereas mean annual maximum temperature reaches
chat and fruit trees. Plots of land at distant locations 25.4°C in 2009 (EMS, 2013).

Figure 1. Map of the study area. Lower Right: location of the study area in Ethiopia, Upper-right: location of the study area in
Southern Region, Left side: study area with streams, contour, and Roads.
312 M. MARIYE ET AL.

Total of (77.4%) 763, 245 people live in the rural area, usgs.gov) of the USGS and (EROS) centre to quantify
while only (22.6%) 223,432 of the total population live in the magnitude and direction of LULC change. The
towns (CSA, 2013). The average population density was following factors were considered when selecting
728 people per km2 (Maryo, 2020). The average land­ satellite images: (i) the principal events that occurred
holding size for each family is less than 1 ha per house­ in the area and (Muriithi, 2016) image quality that
hold. The agricultural system of the study area is mainly reduces the impact of fire and clouds on land cover
characterised by the mixed farming system where the mapping. We accessed satellite images with
rural people depend on both crop and livestock produc­ a minimum cloud cover (<10%) between November
tion for their livelihood. Cattle, sheep, goat, pack ani­ and January, which is before the usual season of forest
mals, and poultry are the most common domestic fires. Moreover, we employed multispectral satellite
animals raised in the study landscape. Commonly data (Landsat 4, 5, 7, and 8 with sensors MSS, TM,
known as ‘false babana’, Enset (Ensete verticosum) is the ETM+, and OLI) from 1973, 1986, 2000, and 2018,
dominant land use, cultural heritage, and most impor­ respectively, to identify and evaluate satellite images.
tant food security crop for the local community of the The LULC classifications were interpreted using socio­
south-central region of Ethiopia. The overall land use of economic and additional data sets following previous
the study area is patterned in concentric circles around studies done by (Andualem et al., 2018; Dibaba et al.,
homesteads, and Ensete occupies the inner rings that 2020; Mohajane et al., 2018).
surround the homestead. The fact that the south-central We obtained ASTER, DEM, and topographic maps of
Ethiopia is known for its dense population concentra­ the study area (1:50,000) from the Geospatial
tion (due to historical and cultural reasons) and its sui­ Information Institute of Ethiopia, then the images were
table agro-ecological condition for human settlement, orthorectified into Universal Transfer Mercator (UTM)
the natural vegetation cover of the study area has been Zone 37 N, World Geodetic System (WGS) 1984.
depleted quite a long time ago. The only exception to Remote sensing and GIS tools, including ERDAS
this practice includes pockets of lands in holy sites, Imagine 2014 and ArcGIS 10.6, were used for image
which preserved ancient trees, and farm boundary processing and data analysis, respectively. At first,
strips. In the study area, with steep sloppiness and images were converted to UTM and geo-referenced
extremely undulating geography, animals cannot access using a datum selected by WGS-84 for Ethiopia. We
feeds near mountainous areas and are particularly kept digitised the demarcated study area in Arc GIS 10.6 to
close to the bottom and at marshy and swampy lands overlay the view on the spatial databases generated by
(Deribe et al., 2019). the photograph and the satellite image. Aerial photo­
graphs from 1986 to 2018 were interpreted via mirror
stereoscope for LULC type identification and classifica­
2.2. Data sources and analysis
tion, whereas post-classification change detection was
We accessed multi-sensor and multi-temporal Landsat used for the evaluation of LULC types on the aerial
images from the data portal (https://earthexplorer. photographs (Mariye et al., 2022). For the creation of

12.4 25.5
Min temprature
Max tempreture
12.2 25.0

12.0
Maximum temperature (oC)
minimum temperature (oC)

24.5

11.8
24.0

11.6
23.5

11.4
23.0

11.2
22.5
1990 1995 2000 2005 2010 2015

Figure 2. A climate diagram showing temperatures from Angecha and Durame stations (1990–2015). Note: The red lines and
vertical numbers on the left symbolise the area of cropland change in a hectare, while the blue lines and vertical numbers on the
right indicate the population growth trend in the study landscape over the last four and half decades.
 ALL EARTH 313

the latest land cover map, Google Earth imagine, field Value of the identified pixel
Area inpercent ¼ � 100
survey, and ground control points (GCPs) were used as Number of pixels in total
major sources of data. LULC trends and dynamics were (2)
assessed using Landsat imagery, which provides a range
of spatial, temporal, spectral, and multi-resolution cap­ � �
Pt2 Pt1
abilities for land cover analysis following Oettera et al. Percentage of LULC change ¼ �100 (3)
Pt1
(2000). The assortment of the suitable image acquisition
date is an essential for clear identification of the LULC Where Pt1 is the area of LULC at the initial period and Pt2
types from satellite imageries (Berihun et al., 2019; is the area of LULC at the final period. A positive result
WoldeYohannes et al., 2018). We coded and organised indicates that the extent of LULC has increased, whereas
the household (HH) questionnaires before conducting a negative result indicates that the amount has
analysis. Coding was established during the preparation decreased. We validated the classification results by con­
of the questionnaires. For analysis, the collected data structing a confusion matrix as a basis for determining
were input into an SPSS version 23 database following accuracy assessments (Belete et al., 2021; Miheretu &
the coding system. HHs’ socioeconomic characteristics Yimer, 2017). To improve the image quality and to esca­
were analysed quantitatively using descriptive statistics. late the accuracy assessment, 300 GCPs from seven land
While the qualitative data of the FGDs, KIIs, and obser­ cover classes (i.e. 35 grasslands, 38 bare lands, 35 settle­
vational notes were transcribed, arranged, and analysed. ments, 40 forestlands, 41 shrub lands, 41 wetlands, and
Then, we used SPSS Version 23 to analyse and summar­ 36 croplands) were used, respectively. Furthermore, we
ise the quantitative data gained from general infor­ used topographic maps obtained from the Ethiopian
mants during a formal survey. In order to enhance the Geospatial Information Institute to visualise the land­
quality of the image, different mosaicking, sub-setting, scape of the whole study area (Anchan et al.). This study
and radiometric enhancement techniques (Haze reduc­ has applied both unsupervised and supervised catalo­
tion and Histogram equalisation) were applied to the guing techniques following previous researchers (Belay
raw data following Belete et al. (2021) & Mengistu, 2019; Congalton, 2001; Congalton & Green,
After the image was processed, signatures were 2009). We collected ground trothing points with the
distributed per pixel by identifying the land into support of FGDs and KIIs during the transect walk. In
seven classes of 1973 and 2000. Furthermore, seven addition, Google Earth Engine images were used as
LULC categories have been recognised for the years a supplementary tool for GTPs. Moreover, GTPs for the
2000 and 2018, respectively. Image cataloguing was 1973 and 1986 LULC maps were collected by using the
based totally on the reflectance characteristics (false- false colour composite of 1973 and 1986 satellite images
colour composite) of the specific land cover classes in conjunction with elder knowledge based on previous
and also supplemented using field observation studies (Betru et al., 2019). In all scenarios, members of
(Belete et al., 2021), key informant interview, and the KIIs and FGDs were employed to collect GTPs by
FGDs. Each class was given a unique identification describing the dynamics of LULC occurring in and around
and assigned a selected colour to differentiate one their district. In supervised classification, specific land
from another. For each of the predetermined LULC cover types are delineated using training sites, whereas
categories, training samples had been selected via in unsupervised classification, land cover classes are
delimiting polygons around representative sites. formed based on the number of classes requested.
During this time, Google Earth extension was Moreover, we combined unsupervised classifications
employed to reduce confusion in interpreting the with visual signature editions based on the spectral
pixel. Furthermore, the results obtained from the values of recent images to determine the classification
supervised class with the help of ERDAS Imagine of older images 1973, MMS imagery (Andualem et al.,
2014 were imported to ArcGIS 10.6 for map layout 2018). A supervised signature extraction with the max­
preparation, reclassification, and pixel value estimation imum likelihood algorithm was used to categorise the
for all LULC classes (Belete et al., 2021; Othow et al., Landsat imageries (Amare, 2015; Congalton & Green,
2017). The conversion matrices were created using 2009).
ERDAS Imagine 2014, and the source and destination In Table 1, a spatial resolution of 15*15 m is used for
of each land cover value were evaluated in an Excel the panchromatic band 8. MMS: Multispectral Scanner;
spreadsheet (Kindu et al., 2013; Meshesha et al., 2016; ETM+: Enhanced Thematic Mapper Plus; TM: Thematic
Othow et al., 2017). In addition to calculating the area Mapper; OLI-TIRs: Operational Land Imager and
of LULC in hectares and percent, we also calculated the Thermal Infrared Sensor.
percentage change in LULC between the stated period
following Braimoh (2006) and Pontius et al. (2004).
2.2.1. Socio-economic survey
Count � 900 or 3600 To validate the data obtained from aerial photos and
Area in hectar ¼ (1)
10; 000 satellite imagery, a socioeconomic survey was
314 M. MARIYE ET AL.

undertaken with 121 randomly selected households. 2.2.2. Satellite data pre-processing and
The district, administration, and household partici­ classification
pants were chosen using a three-stage sampling tech­ Due to systematic and random errors present in raw
nique that included purposive and random sampling, satellite images, these images cannot be directly utilised
whilst the household respondents are being selected for any form of feature identification. Therefore, we
through systematic random sampling following Wubie applied the most common Landsat pre-processing opera­
et al. (2016). As a result, from the three elevation tions such as image enhancement, conversion of radi­
classes, namely Lower (500–1800 m above sea level), ance, solar correction, geometrical rectification, and
Middle (1801–2400 m. a. s. l.), and Upper (2401– normalisation were carried out to improve features and
3200 m. a. s. l.) using a household survey 40 house­ interpretability (Gebremicael et al., 2018; Lu & Weng,
holds from the 2 sample areas and 41 households from 2007) for LULC change following previous research
one sample area were chosen, respectively. Sampling study done by (Birhane et al., 2019; Tucker et al., 2004;
for the socio-economic survey was done in two phases. USGS, 2019). The images produced by Landsat sensors are
The first phase involved the assortment of the sam­ subject to distortion caused by sensor, solar, atmospheric,
pling sites, however, the second phase comprised and topographic factors. Therefore, employing these
selecting individual households from the chosen methods would minimise effects to the extent that is
kebeles (the smallest government administrative struc­ desired for a given application. ERDAS imagine was
ture in Ethiopia) using system sampling following pre­ used to implement topographic and atmospheric correc­
vious approaches (Amare, 2015). Furthermore, FGDs tions for airborne imagery, which reduces errors and gets
were carried out with 30 participants (12 women and the actual reflectance values for each LULC classification
18 men) to acquire additional information on the long- (Lu & Weng, 2007). In this study, we consider the following
term LULC practices. In general, four FGDs were con­ gap-filling or destriping methods to remove Landsat
ducted where each group was composed of six parti­ image strips: Geostatistical Neighbourhood Similar Pixel
cipants i.e. development agents, farmers, Kebeles Interpolator (GNSPI); the weighted linear regression (WLR)
cabinet members, youth association, and community algorithm; and the direct sampling (Oestreicher et al.,
elders who had been selected through the kebeles 2014) method. Furthermore, image difference was esti­
administrative bodies and the knowledgeable commu­ mated as the difference in the total number of equiva­
nity representatives. In order to engage in in-depth lently classed pixels between two images, which was
discussions, key informant interviews, and FGDs were computed by subtracting the initial state class totals
conducted to acquire information regarding the past from the final state class totals.
and present situations, including the drivers of LULC The basic sources of data to generate an up-to-date
change within the study landscape (Mariye et al., land cover map of the study area were Google Earth
2022). For the in-depth discussion, 15 people (age images and ground control point reading. The trends
>60 years) were purposefully chosen as key informants and dynamics of LULC changes were analysed using
(Danano et al.) to gather data on the trends of LULC Landsat image that offers a multi-temporal, multi-
alteration over the past 4 and half decades. We calcu­ spectral, and multi-resolution range of imagery for the
lated the study sample size following Kothari (2004) land cover analysis of the current study following Oettera
et al. (2000).
Z2 :p:q:N
n¼ (4)
e2 ðN 1Þ þ Z2 :p:q 2.2.3. Change detection
Monitoring changes of the earth’s surface features
Where n is the sample size and Z denotes the 95 con­ enable us to realise the interactions between humans
fidence limit (interval) under the normal curve, which and their environment for health management and the
would be 1.96. P = 0.1 (percentage of the population to use of natural resources as reported by Singh (1989).
be included in the sample that is 10 percent). q = none Numerous change detection techniques have been
event occurrence = 1–0.1, which denotes (0.9). N = is the created and utilised for different applications.
total number of households = 2100. e = allowable error However, they can be broadly divided into post-
term (margin of error or degree of accuracy) (0.05). classification and spectral change detection
approaches (Singh, 1989). The post-classification
2
Thus n ¼ 0:052 ð1:96 �0:1�0:9�2100
2100 1Þþ1:962 �0: 1�0:9
¼ 121 approach evaluates the change in LULC based on
a detail-categorised classification of land cover units
The overall procedure for the LULC change analysis (Ayele et al., 2018). Jianya et al. (2008) discussed that
was organised in the schematic diagram (Figure 3), and change detection approaches can be characterised in
seven classes of LULC categories were identified two groups: Bi-temporal altered detection is a direct
(Table 2), and the description in Table 2 is provided comparison, post-analysis comparison, and uniform
from various supporting data (Kindu et al., 2013). modelling that measures changes based on a simple
 ALL EARTH 315

Data acquisition

Primary data Secondary data

collection

Ground truth data, key

informant interview, Topographic map, Soil Satellite image and
FGDs, field survey and types, RF, Reports, Google earth
transect walk manuals, documents

Landsat images of 1973, Data analysis GCPs for LULC

1986, 2000, and 2018 ArcGIS, ERDAS Imagine, change observations
Origin pro 9, SPSS 23

Image Unsupervised
Image processing Supervised
Enhancement classification
Layer staking
sub setting
Initial LULC map

Accuracy Time series LULC

Assessment maps of (1973-2018)
Change detection LULC change
and Final LULC analysis
map

Analysis of Extent, Rate,

and change dynamics

Figure 3. Flowchart presenting techniques employed to attain the ultimate LULC change in the study area.

Table 1. Explanation of imagery statistics and sources used for LULC study in KTz.
Satellite image Imagery type Imagery date Used bunds Source Spatial Resolution Path/R Bands/colour
Land sat_ 4 MSS January–1973 4 bands, 1–4 USGS 57*57 169/55 Multi-spectral
Landsat_5 TM February–1986 5 bunds, 1–5 USGS 28.5*28.5 169/55 Multi-spectral
Landsat_7 ETM+ January–2000 8 bands, 1–8 USGS 15*15 169/55 Multi-spectral
Landsat_8 OLI-TIRS January–2018 8 bands, 1–8 USGS 30*30 169/55 Multi-spectral

Table 2. LULC categories and their description in the study landscape, Ethiopia.
Land cover
No type Their explanations
1 Settlement A land-use type that includes rural settlement area, educational, health, socio-economic facilities, residential houses,
administrative buildings, small-scale industrial areas, etc.
2 Wetland Land use that is waterlogged and swampy during the wet season, which dries in the sunny season.
3 Cropland Smallholder farmers owned land that is used to grow cropland. It is characterised by tilled and planted, bare crop fields, and
limited areas temporarily left as fallow.
4 Shrub land Areas covered by bushes, and shrubs > 20% cover, and mixed with grasses; less than 20% tree cover.
5 Forest land Land covered with dense trees > 80% canopy cover which includes evergreen forest land, mixed forest, and plantation forests
6 Grassland A land-use type where the land is dominated by grasses, forbs, and herbs with nil or little proportion of shrubs that are used for
communal grazing.
7 Bare land Areas with little or no vegetation cover consist of exposed soil and/or rock outcrops, and quarries.
316 M. MARIYE ET AL.

‘two epochs timescale comparison whereas temporal approach in cases where the historical maps were poor
trajectory analysis is a time series analysis and the (Fasika et al., 2019; Gashaw et al., 2017; Tilahun, 2015). This
changes are based on a ‘continuous’ time scale, focus­ study used a random sampling technique to select points.
ing on both changes between dates and progress of As a result, enough spatial distribution of sampling points
the modification over the period. To obtain the infor­ was attained for each LULC class. The accuracy measure
mation of LULC dynamics in terms of pattern and rate was worked out based on the confusion matrix that
conversion, post-classification change detection analy­ comprises the Producer’s and User’s accuracy (Asokan &
sis was carried out in ERDAS Imagine 14 using classifi­ Anitha, 2019; Thakkar et al., 2017). In order to compute the
cation images of 1973, 1986, 2000, and 2018. user accuracy, we divided the number of correctly classi­
According to Lu et al. (2004), accuracies of change fied pixels in each class by the number of training set
discovery results depend on the performance of the pixels per classified class, which indicates the percentage
image processing and categorisation approach. of correctly classified pixels per LULC class (Othow et al.,
According to Singh (1989), change detection is used 2017). Producer accuracy is computed by dividing the
when someone is interested to know changes that have number of correctly classified pixels by the total number
taken place in a particular region provided that satellite of pixels (reference totals), showing the percentage of
data over that area can be easily attained. The key objec­ correctly classified pixels using the reference data follow­
tive of the change detection technique is to obtain altera­ ing Belete et al. (2021). The low accuracy record of pro­
tions in two or more images of the identical sites captured ducers and users may have been related to the similarity
at, unlike periods. Different approaches have been of land cover classes at a spectral level. Belete et al. (2021)
adopted for change detection depending on its applica­ explain that low producer’s accuracy indicates that the
tion such as algebra-based, transform-based, classifica­ ground reference points for a category are classified
tion-based, neural network and fuzzy-based, and other incorrectly, while low user accuracy suggests that pixels
change detection approaches (Asokan & Anitha, 2019). could be classified that do not exist on the ground. The
overall classification accuracy was weighted by the num­
ber of samples in each class, i.e. the sum of all samples on
3. Result and discussion the diagonal divided by the total number of cells
(36 + 35 + 40 + 41 + 38 + 41 + 35)/300, which equals
3.1. Accuracy assessment for LULC maps from 88.7%. This result indicates that 88.7% of LULC classes
(1973–2018) have been properly classified. Overall, kappa statistics of
Based on the results of the post-classification accuracy 0.87 was achieved for the classification period of 2018
assessment, the overall user accuracy evaluation was images, which means that there is 87% in better agree­
found to be highest for wetland (100%) and lowest for ment than would be expected by chance alone (Table 3).
cropland (70.6%) in 2018, respectively, as depicted in The kappa coefficient ranges from +1 to – 1
(Table 3). During the final study periods, the producer indicates the level of accuracy between the refer­
accuracy, which is the proportion of properly classified ence data (true data) and the analysed images. In
data from total classified data was (92.7%) for bare land, a study by Hassan et al. (2016), the kappa coeffi­
(90.9%) forest, (81.4%) settlement, (100%) cropland, cient is associated with three possible groups of the
(91.1%) shrub land, (81%) grassland, and (85.4%) wetland, covenant: the value over 0.80 shows strong agree­
respectively, as shown in (Table 3). The overall user accu­ ment, the value between 0.40 and 0.80 denotes
racy for the classified maps of grassland and shrub land moderate covenant, and the value below 0.40 indi­
was (79.1%) to (97.8%) for the second study period. cates poor covenant (Belete et al., 2021). We also
Producer accuracy for individual LULC classes was highest computed a Kappa statistic for each classified map
for grassland (100%) and lowest for wetland (81%) in to measure the accuracy of the classification results.
2000. We collected a total of 785 Ground Control Points According to the maps from 1986, 2000, and 2018,
(GCPs) from a Global Positioning System (GPS) during our the overall accuracy found to be was 88.3%, 87.6%,
field visit and translate the walk with the help of Google and 88.7%, whereas kappa values were 0.84, 0.84,
Earth engine. Out of which, 485 points were used for and 0.87 for each classified map from 1986, 2000,
supervised classification, while the remaining 300 points
and 2018, respectively. Accordingly, for the subse­
were used for assessing the accuracy of satellite images.
quent analysis and change detection, this kappa
We retrieved the 1973 and 1986 reference points using
accuracy was generally considered reasonable.
Google Earth, Landsat images, historical reports and
maps, and field observations in 2018. A high resolution Kappa statistics for 2018 demonstrated a strong
of Google Earth was used to better identify the land-use covenant, while for 1986 and 2000 good agreement
classes on the historical maps of 1973 obtained from the was obtained as shown in (Table 3). Finally, it was
Ethiopian mapping agency since the resolution of the calculated following the previous study approach
maps was poor. Many other studies have used the same proposed by Tilahun (2015).
 ALL EARTH 317

Table 3. Accuracy of LULC change maps for 1986, 2000, and 2018.
LULC CL GL FL SHL BL WL SL UA% K
2018 CL 36 2 2 2 2 3 4 70.6 0.87
GL 0 35 0 0 0 2 4 85.4
FL 0 2 40 0 0 0 0 95.2
SHL 0 2 2 41 0 0 0 91.1
BL 0 0 0 2 38 0 0 95
WL 0 0 0 0 0 41 0 100
SL 0 2 0 0 1 2 35 87.5
PA% 100 81 90.9 91.1 92.7 85.4 81.4
OAA % 88.7
CL 49 0 2 5 0 0 0 87.5 0.84
GL 4 34 3 2 0 0 0 79.1
2000 FL 3 0 33 2 0 1 0 84.6
SHL 0 0 0 45 0 1 0 97.8
BL 0 0 0 0 0 0 0 0
WL 1 0 0 0 0 9 0 90
SL 0 0 0 0 0 0 0 0
PA% 86 100 87 83.3 0 81 0
OAA% 87.6
CL 43 0 4 2 0 0 0 87.8 0.84
GL 2 39 6 3 0 0 0 78
1986 FL 0 2 40 0 0 0 0 95.2
SHL 2 1 0 44 0 0 0 93.6
BL 0 0 0 0 0 0 0 0
WL 0 0 0 0 0 0 0 0
SL 0 0 0 0 0 0 0 0
PA% 91.5 92.9 80 89.8 0 0 0
OAA% 88.3
Note: CL; Crop land, GL = Grass land; FL = Forest land; SHL = Shrub land; BL = Bare land; WL = Wetland; SL = Settlement land; UA = User accuracy;
PA = Producer accuracy; OAA = Overall accuracy; K = Kappa statistics.

Pn Pn
N mii
i¼1 ð Gi C i Þ 79800 12; 800 67000
K¼ Pn i¼1 (5) K¼ ¼
N 2
ð G C
i iÞ
90000 12; 800 77200
i¼1
¼ 0:867,0: 87 ðvalue of kappa coefficient for 2018Þ:
Where: i denotes the class number, (N) represents the
total number of classified values in comparison to truth The greater the kappa coefficient the better is the seg­
mentation accuracy (Asokan & Anitha, 2019). The rate of
values, (mii) is the number of values that fall into the
LULC conversion for the four periods from 1973–1986,
truth class i and are also classified as values found
1986–2000, and 2000–2018, and 1973–2018 was com­
along the diagonal of the confusion matrix, (Ci) is the
puted using the previous research approach conducted
sum of all predicted values belonging to class i, and (Gi)
by Amanuel and Mulugeta (2014).
indicates the total number of truth values in class
Pn Pn Y2 Y1
i Where i¼1 mii and i¼1 ðGi Ci Þ are calculated R¼ (6)
using the equations given below. t
Where R = rate of change; Y2 = recent year land-use/cover
N ¼ 36 þ 43 þ 44 þ 45 þ 41 þ 48 þ 43 ¼ 300
in ha; Y1 = initial year LULC in ha, and t = interval year
Xn between initial and recent year.
i¼1
mii ¼ 36 þ 35 þ 40 þ 41 þ 38 þ 41 þ 35
¼ 266
3.2. Status of LULC classification in the study
Xn
ðGi Ci Þ ¼ ð36 � 51Þ þ ð43 � 41Þ þ ð44 � 42Þ landscape
i¼1
þ ð45 � 45Þ þ ð41 � 40Þ In the initial study period (1973), grasslands covered
þ ð48 � 41Þ þ ð43 � 40Þ 45.5% (20, 401.1 ha) followed by forestland 43.1% (19,
¼ 12; 800 318.6 ha), croplands 9.24% (4141.2 ha), shrub land
2.01% (901.1 ha), and bare lands (0.1%) wet lands
300
� ð36 þ 35 þ 40 þ 41 þ 38 þ 41 þ 35Þ � accounted the lowest percentage of the study land­
ð36�51Þ þ ð43�41Þ þ ð44�42Þ þ ð45�45Þ
þ ð41�40Þ þ ð48�41Þ þ ð43�40Þ scape, whereas the settlement lands were totally lost
K¼ � � �� (Table 4). However, in the second study period (1986),
2 ð36�51Þ þ ð43�41Þ þ ð44�42Þ þ ð45�45Þ
300Þ
þ ð41�40Þ þ ð48�41Þ þ ð43�40Þ croplands were dominant LULC classes, which
accounted for 32.04% (14, 360.03 ha), followed by
� �
ð1836Þ þ ð1763Þ þ ð1848Þ grasslands 27.67% (12, 401.1 ha), shrub lands 23.9%
300ð266Þ
þ ð2025Þ þ ð1640Þ þ ð1968Þ þ ð1720Þ (10, 705.8 ha), forestlands 16.29% (7302.03 ha), wet­
K¼ � �
ð1836Þ þ ð1763Þ þ ð1848Þ land 0.1% (32.1 ha), but bare land occupied the mini­
90000
þ ð2025Þ þ ð1640Þ þ ð1968Þ þ ð1720Þ mum share of the study area. In the third period (2000),
318 M. MARIYE ET AL.

about (45.4%) of the study area was occupied by crop­ farming. These findings are consistent with previous
lands, followed by shrub land 23.9% (10, 745.2 ha), research by Shiferaw and Singh (2011), Siraj et al.
grassland 14.2% (6345.8 ha), forestland 13.1% (2018), and Hailu et al. (2020) who have found 65%
(5882.3 ha), wetlands 3.2% (1431.6 ha), whereas bare augmentation in cropland respectively. Obviously, in
land and settlements share the lowest percentage. In subsistence farming, when modern technology is not
the final study period (2018), approximately 51.1% (22, used, agricultural input is insufficient or non-existent,
913.8 ha) of the study area was covered by croplands boosting yields is achieved by bringing additional
followed by Shrub land 23.5% (10, 545.5 ha), forestland parcels of land under farming, and this is specifically
10.8% (4850.9), grassland 6.3% (2816.9), wetland 3.9% what has been seen in the current study area. The
(1733.6 ha) even though bare lands and settlement current research results are consistent with those of
accounted the lowest proportion of the study area many previous Ethiopian studies. For instance,
(Table 4). The areal extent of LULC changes and their Amanuel and Mulugeta (2014) indicated agricultural
distribution, for each time phase, are illustrated land increased by 19.16%, 52.11%, and 65.6% in 1973,
(Table 4 and Figure 4). The LULC changes were classi­ 1986, and 2004, respectively. Similarly, Alemu et al.
fied into 4-time series (1973–1986), (1986–2000), (2015) point out that the percentage of cropland has
(2000–2018), (1973–2018), and LULC maps of the increased from 23.5% in 1985 to 39.11% in 2010,
study landscape and the subsequent statistical preci­ respectively. Alike, findings were reported by
sion are shown in (Figure 5, and Table 5), respectively. Andualem et al. (2018), who have found that cropland
increased by 13.78% between 2007 and 2018. Agidew
and Singh (2017) also come up with similar outcomes
3.3. Dynamics of LULC change between 1973 and
observing that cropland cover increased by 55.23%
2018
between 2000 and 2010. Moreover, the current results
3.3.1. State of cropland LULC are in opposition to that of a study by Deribew and
The majority of the rural inhabitants in Kembata Dalacho (2019) that outlines the decline of cropland
Tembaro district, as elsewhere in Ethiopia, rely mainly due to prolonged drought, outbreaks of disease, and
on agriculture, especially crop cultivation. The LULC the displacement of people, between 1973 and 1986,
trend analysis conducted over the past four consecu­ leading to a conversion of cropland to grassland and
tive study periods 1973–1986, 1986–2000, 2000– shrub land. However, in the current study area infor­
2018, and 1973–2018 shows that the study area mants have reported that the drought happened for
experienced substantial LULC changes (Table 5). just a short time and was not severe. Furthermore,
Between 2000 and 2018, the amount of cropland Martínez et al. (2009) documented a significant reduc­
augmented with an annual increasing rate of tion of farmland in Mexico, with an annual decreasing
142.3 ha/year. The study landscape recorded an rate of – 7% and – 20% between 1973–1990 and
expansion in cropland between 1973 and 1986 of 1990–2003, respectively.
4141.2 ha (9.2%) which then increased to 14, 360.ha
(32%), whereas in the final period (2018) the overall 3.3.2. State of Wetland LULC
situation was changed and it occupied the largest Land covered by wetlands in the study landscape was
portion 22, 913.8 ha (51.1%) followed by shrub land 15.6 ha accounting for (0.04%) of the total area in 1973.
(23.5%), forest (10.8%), and grassland (6.3%). The The wetland cover extent and its proportional share in
remaining shares were covered by grassland, bare the years 1986, 2000, and 2018 were (0.07%), (3.2%),
land, and wetland. Between 1973 and 1986 cropland and (3.9%), respectively. The land cover of wetland had
gained 246.8% area from other land cover classes and shown a continuous increase in the study area
experienced a positive rate of change (Table 7). This (Table 4). In summary, the trend analysis revealed
type of transition is typically caused by rapid popula­ that wetlands increased by (1718 ha) between 1973
tion expansion happened at the expense of grass­ and 2018, with a percentage change of (1112.8%) and
lands, and forestland. According to evidence an annual increasing rate of (38.2%). These study find­
collected through key informant interviews and dis­ ings contradict the outcomes of (Y. Belayneh et al.,
cussions with FGDs, the expansion of croplands, as 2018), who found a significant decrease and full drying
well as the reduction of grassland and forestland up of wetlands in Ethiopia. Similarly, Hailu et al. (2020)
cover was happened as a result of population pres­ found out that from 1973 to 2019, there was a steady
sure. Moreover, as discussed by participants of the reduction in the entire area of wetland, with an aver­
FGDs, during this period, crop lands were abundant age lessening rate of 172.6 ha/year.
and the population pressure was low in most places
of the study area. Between 1973 and 1986, the highest 3.3.3. State of Bare LULC
annual increase rate of cropland (and hereafter the At the beginning of the current study, bare land cov­
loss of grassland, forestland, and bare land) has been ered only a few areas 35.41 ha in (1973), 2.03 ha in
noted, which may be related to a rise in small-scale the second period, and 1.21 ha in the third period.
 ALL EARTH 319

However, this had expanded to 1399.55 ha (3.2%) in revealed that the forest cover of the study area has
the final study period (2018). Between 1973 and 1986 been gradually declining at an annual rate of –
the bare land LULC showed increasing and decreasing 321.5 ha/year between 1973 and 2018 (Table 5). This
trends. However, in general, the LULC trend analysis finding is in agreement with the study by Assen and
showed that bare land augmented by (1364.1 ha) in Nigussie (2009) and Siraj et al. (2018), both of which
the final study period (1973–2018), with a percentage recognised that Ethiopia’s forest cover had been con­
change of (3852.4%) and an annual increasing rate of verted into other land uses. The findings of this study
30.31 ha/year has been recorded over the last 45 years. are similar to those of several previous Ethiopian studies.
This could be attributed to the sloppy area cultivation For example, Degife et al. (2018) have described forest­
practices and incompatible farming methods. land decline from 16.41% in 1987 to 15.50% in 2017, and
Informants confirmed that several people confessed Daye and Healey (2015) indicated analogous outcomes
to engaging in deforestation to increase their produc­ in south-west Ethiopian, and Degifea et al. (2019)
tivity and expand agricultural land. Moreover, growth reported a similar trend in the Lake Hawassa watershed,
in the human population in the study area in conjunc­ and Dibaba et al. (2020) noticed that lessening of forest
tion with migration and formal settlement pro­ coverage from (21.55%) in 1987, (17.61%) in 2020, and
grammes led to an increase in the bare land cover. In (9.18%) in 2017. Likewise, WoldeYohannes et al. (2018)
the current study area, climate change and soil erosion also stated forestland has declined at an annual rate of
(e.g. water and wind erosion) linked to inadequate land 1.71 ha/year during 1985–2010. Similarly (Hassen &
management resulted in infertile and bare land. This Assen, 2017) conveyed the shrinking of forest coverage
finding is in contrast with research done by Hailu et al. by 23.1% between 2004 and 2014. Agidew and Singh
(2020) who reported a decline in bare land from (2017) stated that there was a decline in forestland in
2872.1 ha (7%) in 1973 to 551.34 ha (1.3%) in 2019. Northeastern Ethiopia. Tewabe et al. (2020) also indi­
Despite this, the current study finding is in covenant cated analogous outcomes. Amanuel and Mulugeta
with those from a wide range of previous studies in (2014) reported that forestland declined at an average
Ethiopia and beyond. For example, (Tolessa et al., 2017) rate of 22.64 ha/year from 1973 to 2004. It also reported
in the central highlands of Ethiopia, from 0 in 1973 to the decline of forestland in the lake Wanchi watershed.
(739.08%) in 2015, and Siraj et al. (2018) noticed an In contrast to the current study, Andualem et al. (2018)
increase in bare land with an annual rate of 5.23 ha found that between 2007 and 2018, the forest LULC
per year between 1973 and 2015, and Bekele et al. increased by 1.46%. Similarly, Gebrelibanos and Assen
(2018) observed an increased rate of 91 ha per year (2013) confirmed that there was a consistent increase in
from 1998 to 2011, Olorunfemi et al. (2018) reported forest cover between 1964 and 2006. Informants of the
similar findings in southern Nigeria. Additionally, Abate current study mentioned the following reasons for the
(2011) reported that bare land expanded at an annual loss of the forest. Fundamental reasons include the out­
increasing rate of 256 ha per year between 1972 and break of the 1985/86 famine, which forced families to
1985. extract locally available forest resource. Similarly, raise in
population caused division of forestland and other
3.3.4. State of Forestland LULC shrub lands aimed to maximum production.
The study results showed there was a continuous per­ Furthermore, in 1991/92 there was a downfall of the
centage loss of forestland in all study periods. Its share military government and replacement of non-military
was 19, 318.62 ha (43.1%), 7302.03 ha (16.29%), government caused weak and low enforcement where
5882.33 ha (13.13%), and 4850.91 (10.82%), in 1973, forestlands were illegally converted into other land use
1986, 2000, and 2018, respectively (Table 4). The reason types (Demissie et al., 2017). Youths with no farms and
for this may have been the rapid annual population farmers with very small farm land contribute negatively
growth in the area by 2.8% (CSA, 2007) which likely to the LULC changes by cutting trees for firewood and
required more agricultural land. Moreover, the result charcoal, which poses a burden upon the existing forest

Table 4. Area covered their status and percentage change of the classified LULC categories between 1973, 1986, 2000, and 2018 in
the study landscape.
1973 1986 2000 2018
LULC Classes area (ha) % area (ha) % area (ha) % area (ha) %
Crop land 4141.2 9.24 14,360.03 32.04 20,351.64 45.41 22,913.82 51.13
Grass land 20,401.072 45.5 12,401.1 27.67 6345.77 14.16 2816.98 6.29
Forest land 19,318.62 43.1 7302.03 16.29 5882.33 13.13 4850.91 10.82
Shrub land 901.11 2.01 10,705.8 23.9 10,745.21 23.98 10,545.5 23.53
Bare land 35.41 0.1 2.03 0.01 1.21 0.03 1399.6 3.12
Wet land 15.60 0.04 32.09 0.072 1431.61 3.19 1733.6 3.87
Settlements land 0 0 10.03 0.02 55.26 0.12 552.69 1.23
Total 44,813.03 100 44,813.03 100 44,813.03 100 44, 813.03 100
320 M. MARIYE ET AL.

Settlement

LULC Classes identified in the study area

2018
Wet Land
2000
1986
Bare Land
1973

Shrub Land

Forest Land

Grass Land

Crop Land

0 5000 10000 15000 20000 25000

LULC Change dynamics in the study landscape (ha)

Figure 4. LULC distribution of the study landscape.

lands. There are many resource poor individuals who from 5.8% in 1980 to 4.1% in 2006 with an annual decreas­
practice this form of economic activity in the area, espe­ ing rate of 1.17%. Abate (2011) reported that grasslands
cially jobless and landless youths. This situation needs declined by 73 ha/year between 1972 and 1985 in the
immediate attention; otherwise, it will result in severe south Wello highlands. Agidew and Singh (2017)
socioeconomic and environmental disruptions. described a decline in grassland between 1973 and
2015 in Northeastern Ethiopia. Bekele et al. (2018)
3.3.5. State of Grassland LULC reported that grassland progressively declined during
In the mid–1970s, grasslands were the dominant type of the study period from (1985–2011) in the Awash river
LULC within the area. Within the mid-20th, grassland was basin. A study conducted by WoldeYohannes et al.
rapidly substituted by cropland because of high rates of (2018) also conveyed similar results in southern Ethiopia
population raise and also the need for a various range of during 1985–2010. Mikias (2015) also arrived at similar
products, pushing farmers to alter some of their lands to outcomes indicated that grazing land was reduced by
other land-use categories. Grassland covered a total area an annual rate of 13.58 ha/year. Similarly, Tewabe et al.
of (45.5%) in 1973, (27.7%) in 1986, (14.2%) in 2000, and (2020) reported that there was a lessening of grassland at
(6.3%) in the study area in 2018. Based on the current an annual rate of 10 ha/year in 1986–2018. Andualem
LULC change analysis, grasslands exhibited a con et al. (2018) had stated grassland cover decreased by
siderable decline by (–7999.97 ha) with a percentage 15.97% between 2007 and 2018 in the Ribb watershed.
change of (–39.2%) and an annual decreasing rate of –
615.4 ha/year changes in the initial study period (1973– 3.3.6. State of shrub LULC
1986) and by (–3528.8 ha) an annual lessening rate of Shrub land was the second largest in percentage cover
change – 196 ha/year in the third period, respectively between the third and fourth periods when compared to
(Table 5). In summary, the LULC trend analysis revealed the other LULC categories that accounted for 10,
that grassland lessened by (–17,584.1 ha) between 1973 745.21 ha (23.9%) and 23.53% (Figure 4). The amount of
and 2018, with a percentage change of (–86.2%) and an shrub lands, on the other hand, decreased steadily in the
annual declining rate of (–390.8%). FGDs and key infor­ third period while declining with a percentage change of
mant interviews corroborate with satellite data, which (–1.86%) and an annual shrinking rate of (–11.1) ha/year
reveals a continual decrease during the study period, (Tables 5 and 7). Shrub land experienced the biggest
which is connected to periodic fires that occur during conversion with nearly 667 ha (73.65%) of its entire area
the dry period and the expansion of farming activities. transformed to cropland 499 ha (49.6%), wetland 115 ha
Moreover, based on responses from local residents (12.7%), and the rest to other LULC categories as depicted
obtained through interviews and group discussions, an in (Table 6). According to key informants and FGDs,
increase in the desire for agricultural production on exist­ this year also experienced a time when Ethiopia was
ing land and the establishment of new households in the severely impacted by famine and drought. Shrub land
study landscape may have contributed to the reduction significantly diminished in the third period, when this
of grass land. Thus, the present study found that conver­ land cover was officially disseminated for youths for culti­
sion of grass land into cropland and settlement is vation activities. This might be attributable to the expan­
a common practice in the study landscape. This finding sion of cropland and settlement area at the expense of
is consistent with the result of Asmamaw (2011) who have shrub land. In general, the LULC trend analysis showed
found that grasslands in northeastern Ethiopia declined that shrub land augmented by (9644.37 ha) in the final
 ALL EARTH 321

Figure 5. Classified LULC map of the study area in 1973, 1986, 2000, and 2018.

Table 5. Rate of LULC dynamics from the time when (1973–2018) in the study landscape.
1973–1986 1986–2000 2000–2018 1973–2018
RC
LULC Classes AC (ha) RC (ha/year) AC (ha) RC (ha/year) AC (ha) (ha/year) AC (ha) RC (ha/year)
Crop land 10,218.8 786.1 5991.6 428 2562.2 142.3 18,772.6 417.2
Grass land –7999.97 –615.4 –6055.3 –432.5 –3528.8 –196 –17,584.1 –390.8
Forest land –12,016.6 –924.4 –1419.7 –101.4 –1031.4 –57.3 –14,467.7 –321.5
Shrub land 9804.7 754.2 39.4 2.82 –199.7 –11.1 9644.37 214.3
Bare land –33.4 –2.6 –0.82 –0.06 1398.3 77.7 1364.1 30.31
Wet land 16.5 1.3 1399.5 99.9 301.99 16.8 1718 38.2
Settlement 10.03 0.77 45.2 3.2 497.4 27.6 552.7 12.3
Note: AC = Area of change; RC = Rate of change
322 M. MARIYE ET AL.

Table 6. An error matrix of LULC change detection from 1973 to 2018 in the study landscape.
LULC change from the initial year 1973 (ha)
LULC Classes Crop land Grass land Forest land Shrub land Bare land Wet land Settlements Row Total Loss

LULC change to the Final year 2018

Crop land ha 2733 12,649 7002 449 15 6 0 22,854 20,121
% 64.4 61.5 36.1 49.6 42.2 37.5 0
Grass land ha 280 1246 1266 7 0 0 0 2799 1533
% 6.6 6.1 6.5 0.8 0.92 2.1 0
Forest land ha 82 900 3796 49 1 0 0 4828 1032
% 1.93 4.4 19.6 5.4 1.8 2.1 0
Shrub land ha 597 3238 6468 227 8 3 0 10,541 10,314
% 14.1 15.8 33.4 25.1 22.02 16.7 0
Bare land ha 170 1002 168 41 2 0 0 1383 1381
% 4 4.9 0.9 4.5 6.4 0 0
Wet land ha 284 1061 258 115 4 2 0 1724 1722
% 6.7 5.2 1.3 12.7 11.9 10.4 0
Settlements ha 51 251 249 6 0 0 0 557 557
% 1.2 1.2 1.28 0.65 0.92 0 0
Column Totl 4197 20,347 19,207 894 30 11 0 – –
Gain 1464 19,101 15,411 667 28 9 0 – –
Net change −18,657 17,568 14,379 –9647 –1353 –1713 –557 – –
NP –6.83 14.1 3.79 –9874 −676.5 −856.5 – – –
Note: Those figures in bold are the sum of diagonals and represent overall persistence (i.e. the landscape that did not change). The NP symbolises the net
change-to-persistence ratio (i.e. the net change versus the diagonals for each class). Where net change, loss, gain, and net persistence are calculated
using the equations given below..

study period (1973–2018), with a percentage change of forestation in the study area. These findings are consistent
(1070.3%) and an annual increasing rate of (214.3 ha/ with Miheretu and Yimer (2017) they reported that settle­
year). This finding is in contrast with those of several ment areas had expanded between 1964 and 2014 in the
earlier Ethiopian studies. For example, (Angessa et al., Gelana sub-watershed. Initially, when the study started in
2019) stated that shrub land declined annually at a rate the year 1973, the pattern of settlement land was cover­
of 1.3% in the lake Wanchi watershed. Agidew and Singh ing 0% of the total studied area. However, in 2018 the
(2017) identified that shrub lands decreased from 28.4% observed LULC pattern was augmented with coverage of
to 24.6% between 1973 and 2015. Amare (2015) came up 552.69 ha (Table 4). This could be associated with the
with comparable outcomes that identified a decline in growth of residential and other infrastructures, as well as
bushland cover in the Infraz watershed. In covenant with settlers’ occupancy of public lands. During the specified
the current study, Mikias (2015) reported that shrub land years, this land cover has changed at an increasing rate. In
increased by an annual rate of 22.8 ha/year between 1973 the final study period (1973–2018), the overall percentage
and 2010. expansion of settlement was 55,269%, with an annual
increasing rate of 12.3 ha/year. In terms of the settlement,
3.3.7. State of settlement LULC a higher percentage increase does not mean the land was
In the image of 1986, the settlement LULC is identified as covered with housing in the study period; rather, it refers
a separate class. Within this period, it covered 10.03 ha to the proportion of land covered by housing during that
(0.02%) and quickly expanded to 552.7 (1.23%) of the total period.
study area in 2018. The size of the settlement increased
from 0% in 1973 to 0.12% in 2000 and further augmented 3.3.8. LULC change matrix
to 1.23% in 2018 (Table 4). The trend showed a consistent The LULC conversion matrices were analysed to deter­
expansion of the settlement LULC over the last four and mine where the major LULC changes originated and
a half decades being analysed from (1973–2018). There went. A summary of the results of the analysis from
has been a significant expansion of rural settlement at the 1973 to 2018 is presented in Table 6. However, the extent
expense of forestland, indicating widespread de of the changes in each LULC class has all been different,

Table 7. Change in LULC (ha) and percentage share in the study landscape between 1973 and 2018.
1973–1986 1986–2000 2000–2018 1973–2018
LULC Classes Area (ha) Area (%) Area (ha) Area (%) Area (ha) Area (%) Area (ha) Area (%)
Crop land +10,218.8 +246.8 +5991.6 +41.7 +2562.2 +12.6 +18,772.6 +453.3
Grass land –7999.97 –39.2 –6055.3 –48.83 –3528.8 –55.61 –17,584.1 –86.2
Forest land –12,016.6 –62.2 –1419.7 –19.44 –1031.4 –17.3 –14,467.7 –74.9
Shrub land +9804.7 +1088.1 39.4 +0.37 –199.7 –1.86 +9644.37 +1070.3
Bare land –33.4 –93.3 –0.82 –40.4 +1398.3 +1155 +1364.1 +3852.4
Wet land +16.5 +105.7 1399.5 +4361.2 +301.99 +21.1 +1718 +11,012.8
Settlement +10.03 +1003 45.2 +450.9 +497.4 +900.2 +552.7 +55,269
 ALL EARTH 323

even though all the classes have been undergoing (3238), cropland (597) ha over the last four and a half
changes in the study area. The outcome revealed that decades. Correspondingly, its original coverage was lar­
between 1973 and 2018 there was a significant LULC gely changed to the cropland (449 ha), wetland (115),
dynamics in which (30 ha) of bare land, (19,207) of forest­ forestland (49), bare land (41) ha, respectively. From this
land, (4197) cropland, (894) shrub lands, (20,347) grass­ example, unchanged coverage was 227 ha. The bare land,
lands, and (11 ha) of wetlands were changed one to which covered 30 ha in the first year (1973), was changed
another, respectively (Table 6). The forestland, which cov­ into 1383 ha in the final study period (2018). It gained
ered 15, 411 ha in the initial year (1973), was changed into from other land use types and lost to other land-use
4828 ha in the final study period (2018). It was gained types. Therefore, the unchanged coverage was 2 ha. It
from other land use categories and was lost to other land- gained 170, 1002, 41,168 ha from cropland; grassland,
use types. Therefore, unchanged coverage was 3796 ha. It shrub land, and forestland, respectively. Likewise, its ori­
gained 82, 900, and 49 ha from cropland; grassland, and ginal coverage was changed to the cropland, shrub land,
shrub land, respectively. Similarly, its original coverage wetland, and forestland with the extent of 15, 8, 4, and
was changed to the cropland, shrub land, grassland, and 1 ha, respectively, as illustrated in (Table 6). In general,
settlement with the extent of 7002, 6468, 1266, and cropland, bare land, wetlands, and shrub lands experi­
249 ha, respectively (Table 6). Cropland was a major por­ enced the least persistent cover types, while grassland
tion of the study area that gained additional area from was the most persistent (Table 6). The net change-to-
other LULC categories, which comprise forestland (7002), persistence ratio was large for forestland (positive), crop­
shrub land (449), grassland (12, 649), wetland (6), and bare land (negative), grass land (positive) and bare land (nega­
land (15), respectively, over the last four and half decades. tive), wetland (negative), and shrub land (negative),
From this example, the unchanged coverage was respectively, as shown in (Table 6). Throughout the total
2733 ha. Conversely, cropland had been changed to landscape, 8006 ha (i.e. the sum of diagonal elements)
another form of LULC category, which comprises settle­ remain unchanged (Table 6). Contrary to other land uses,
ment, grassland, shrub land, forestland, bare land, and cropland to shrub land experienced the largest change
wetland with the extent of 51, 280, 597, 82, 170, and matrix.
284 ha, respectively (Table 6).
3.3.9. Rate of LULC dynamics
Loss ¼ Row total
Table 7 already illustrates the rate of change in crop­
diagonals of each class ðunchangedÞ (7)
land, grassland, forestland, wetland, barren land, bush
land, and settlement area cover for the study land­
Net change ¼ gain loss (8) scape. This result indicates that although resources
are fixed, land cover types are subject to varying
Gain ¼ Column total rates of change. However, the rate of change varies
diagonals of each class ðunchangedÞ (9) considerably among the various land cover types. The
LULC analysis disclosed that cropland and settlement
Net persistence ¼ Net change=diagonals of area expanded at rates of 417.2 ha/year and 12.3 ha/
(10) year, respectively, between 1973 and 2018, resulting in
each class ðunchanged
the loss of grassland, and forestland. In the same per­
The amount of land area for each LULC type in the initial iod, grassland and forestland decreased by – 390.8 ha/
study year (1973) is summed in the row, and the amount year and – 321.5 ha/year, respectively. In contrast to
of land area converted to each LULC type in the year 2018 the first and second study periods, between 2000 and
is summed in the column. The area of each LULC class that 2018 bare land increased by 77.7 ha/year. However, in
stayed unaltered is shown by the bold diagonal values, the same study period grass, forest, and shrub land
while the off-diagonal numbers reflect the changed area. rapidly decreased with the annual rate of – 196 and –
The amount of land that was changed from one land- 57.3 and – 11.1 ha/year (Table 5). In the last 45 years,
cover type to another is represented by the values in each wetland, settlement areas, shrub, and bare land have
cell as depicted in (Table 6). On the other hand, grassland, grown at a rate of 38.2, 12.3, 214.3, and 30.31 ha/year,
which covered 20, 347 ha in the initial study period (1973), respectively. In contrast, forestland and grass land
was changed into 2799 ha in 2018. Moreover, it acquired have declined at an average rate of – 321.5 and –
additional land from other LULC categories, mainly forest­ 390.8 ha/year, respectively.
land (1266), cropland (280), and shrub land (7) ha, respec­
tively (Table 6). Likewise, its original coverage was
3.4. The major drivers of LULC change in the
changed to (12,649), cropland (3238), bare land (1002),
study landscape
wetland (1061), forest (900), settlement (251) ha, respec­
tively, during the last four and half decades. Shrubland We collected data from KIIs and FGDs on factors that
has obtained additional areas from other LULC categories drive LULC changes as well as the perceptions of local
that encompass mainly the forestland (4668), grassland people about the alterations. These interviews and
324 M. MARIYE ET AL.

discussions were guided by a checklist of questions (Table 8). The ranks are derived based on how the
during the KIIs and FGDs aimed at understanding the variables were selected frequently by the respondents
trends, patterns, and driving forces of LULC modifica­ during the period under review (Munthali et al., 2019).
tions. A key outcome of the KIIs and FGDs was to identify Population growth, settlement, agricultural expansion,
the major proximate drivers and underlying causes of charcoal production, and lack of financial resources
LU/LCC. An overall of more than 12 elements were were the top five ranked underlying drivers of LULC
mentioned by the informants as the major drivers of changes in the study area, with population growth,
LULC changes in the study landscape (Figure 6). settlement, and agricultural expansion ranked
Conversely, there were differences in each of the factors first, second, and third, respectively (Table 8). Similar
to which the local community viewed as drivers of LULC results were also revealed during key informant inter­
changes. The majority of the respondents (>96.5%,
view and FGDs in which population growth, settlement,
N = 180) have identified population growth, agricultural
and agricultural expansion were identified as the main
expansion, settlement, and charcoal production, respec­
causes of LULC change in the study landscape.
tively, as the major important drivers for the observed
In this study, about (78.6%) of the sampled house­
LULC change in the study landscape. The LULC change
holds were male-headed and (21.4%) females, respec­
analysis indicates that the cropland in the study area has
tively (Table 9). The average ages of the respondent
increased significantly over the last 45 years (1973–
were found to be 43.5 years. The average number of
2018). Likewise, a significant number of respondents
active labour force and land holding of the respon­
(96.5%) believed that LULC changes were caused by
dents were also identified 4.3 persons and 1.8 hectare,
human interference, mainly agricultural expansion. In
respectively (Munthali et al., 2019). Among the entire
response to the significant increase in food demand,
respondents, 60% of them owned less than 2 hectare
crop lands have been expanding by encroaching uncul­
of land, whereas 14% owned greater than 2 hectares’.
tivated areas upon forest lands.
With respect to their education level, 60% of the
Additionally, 82% and 79.6% of the respondents
sampled respondents were illiterates, 37.1% can read
perceived the lack of alternative income sources or
and write and 2.9% have attended primary education
financial resources and poverty as among the major
and above, respectively. The socioeconomic and
drivers of LULC changes in the study landscape. Lack
demographic attributes of the sampled households
of law enforcement (70%), construction (78.8%), bush
are presented in Table 9. The majority of the sampled
fires (56%), and firewood by (25%) were revealed as
respondents were married, about 81%, whereas 19% of
important dynamics accountable for LULC modifica­
them are unmarried. The farm size of the respondents
tions in the study landscape for over the last four and
varied from <1.25 to >2 ha, with an average of 1.8 ha
a half decades. The informants reported that livestock
(Munthali et al., 2019). Approximately, 85% of the
was permitted to graze on the residual crop stalks on
sampled household respondents were involved in agri­
the croplands after harvest and on the shared grazing
cultural occupations, and a small portion of the defen­
lands, which diminished grasslands over the study
dants (25%) were involved in off-farm activities, such as
periods (Figure 7)
government employee and business.
On the contrary, respondents had a low perception
of land-related policy (11%) and demand for timber
(14%) as significant drivers of LULC changes. The infor­
mants reported soil erosion as an important cause for
the decline of agricultural production, and soil fertility, Settlement
1973-2018
LULC classes identfied in the study area

particularly in the upper stream part of the study land­ 2000-2018

Wet land
scape. Key informants and FGD discussants revealed the 1986-2000
1973-1986
disappearance of versatile indigenous tree types such as Bare land
Cordia Africana shrinking of communal grazing lands,
and the deterioration of highly valuable medicinal plant Shrub land

species are other influences of the LULC change in the

Forest land
study watershed. During the field study, we observed
that the pasture lands were insufficient for supporting Grass land
the livestock population in the area, which could have
Crop land
impacted household income leading to an overreliance
on the available natural resources, which inevitably -20000 -10000 0 10000 20000 30000 40000
would have damaged the ecosystem. Rate of LULC change ha/year
From a range of different divers, respondents per­
Figure 6. Rate of LULC change from the time when (1973–
ceived 12 human-related activities as major drivers of 2018) in the study area. Note: Ptn; Production; Plcy; Policy; FRs;
LULC change in the study landscape as depicted in Financial resources; LEnt; Law enforcement; Exp; Expansion
 ALL EARTH 325

Table 8. Drivers of LULC change ranking. 2006), whereas in line with the present study declining
Drivers Percent (%) Rank in forest coverage by (23.1%) was observed between 2004
Population growth 12.2 1 and 2014 in Lake Tana watershed of Ethiopia (Hassen &
Settlement 12.09 2
Agriculture Expansion 12.02 3 Assen, 2017). Wetlands were also diminished throughout
Charcoal Production 11.77 4 the whole study period. This is also associated with the
Lack of Financial resources 10.2 5
Poverty 9.91 6 reallocation of wetland areas for newly married youth as
Construction 9.8 7 a result of the ever-increasing human population in the
Lack of Law enforcement 8.72 8
Bush fires 6.97 9
area, establishment of some infrastructures like commu­
Firewood 3.11 10 nity health centres, schools, farmer training centres, etc. as
Demand for timber production 1.74 11 it was confirmed by FGDs. Moreover, overgrazing of wet­
Land related Policy 1.37 12
lands assisted in the conversion of wetlands to settle­
ments and croplands as described during the FGDs.
3.5. Analysis of LULC changes from 1973–2018 in Conversely, croplands, urban built-up areas, and bare
the study landscape lands showed an increasing trend throughout the entire
Demissie, al., 2017; Eshetu, 2014; Kindu et al., 2015), and study period. Zeleke and Hurni (2001) described similar
the weak institution in the study landscape (Temesgen trends owing to the expansion of cropland causing the
et al., 2013). As per pieces of evidence from field observa­ shortage of appropriate land for agriculture for the ever-
tion and FGDs, it was also a period when a military gov­ growing population in the area. Furthermore, high
ernment (Dirge) has proclaimed a ‘National production human population growth in the area (Figure 8) and the
and cultural development campaign’ that was aimed at aforementioned national campaign also assisted for such
maximising agricultural produce after a great famine and escalation to croplands that corroborates the result of
drought period that took place in 1985. The KIIs and FGDs FGDs. Likewise, degraded land was augmented by over
outcomes with local communities confirmed that it was 44% throughout the whole study period. Conversely,
a time for the application of agricultural fertilisers; massive grasslands and shrub lands indicated both dwindling
settlements in the forested area and reallocation of vege­ and growing trends all over the study periods (Table 4),
tated land to farm less peasants. The collection of local showing the temporal variations of land cover classes and
building material and fuel wood, and civil war in the the columns of spatial variation of land cover classes
middle of 1990 and 1991 were additional causes for the
forestland declines. In contrast to the current study car­
3.6. Implications for policy formulation
ried out by Gebrelibanos and Assen (2013) in the high­
lands of Northern Ethiopia confirmed that there was Agricultural policy in Ethiopia prohibits the cultiva­
a consistent increase in forest cover between (1964– tion of steep slopes (more than 30% gradient)

Table 9. Household characteristics sampled in the study area.

Household Characteristic Value by Percent (%)
Gender
Male 78.6
Female 21.4
Age (Year)
< 45 64.3
> 45 35.7
Education
Illiterate 60
read and write 37.1
Primary and above 2.9
Income obtained from 1 ha of land (ETB)
< 140,000 1.4
140,000–200,000 8.6
>200,000 90
Active labour force in the household (person)
<3 24.3
3–5 60.0
>5 15.7
Total area of land owned (Ha)
< 1.25 24.3
1.25–2 60
>2 14.3
Marital status
Married 81
Unmarried 19
Occupation
Farmer 85
Off-farm activities 25
Ethnic composition
Kembata 78
326 M. MARIYE ET AL.

Demand for timber

No
Yes
Charcoal Ptn
Bush fires
Firewood
Land related Plcy
Lack of FRs
Poverty
Population growth
Construction
Lack of LEnt
Settlement
Agriculture Exp

0 20 40 60 80 100
Drivers of LULC change pereceied by local people (%)

Figure 7. Drivers of LULC change in the study landscape between 1973 and 2018. Note: The red lines and vertical numbers on the
left symbolise minimum temperature variations, while the blue lines and vertical numbers on the right indicate the maximum
annual temperature trend.

except for protective forestry (MoA, 1995). Several Residents and Housing Information of Ethiopia, the total
factors have contributed to the degradation of the number of inhabitants of the study district was 78, 565
watershed, including steep slope cultivation, distur­ (CSA, 2007). However, inhabitants were projected to be
bance of spare mountain plant covers by animals 93, 170 in 2014 and 182,338 in 2017 (CSA, 2013). They also
and humans, poverty, and food insecurity. In order pointed out that human arrivals from the adjacent
to reward farmers who practice effective land man­ regions, predominantly at the beginning of the Socialist
agement, the agricultural policy of the region or Administration (Dirge regime) in Ethiopia, have amplified
district should provide financial incentives. the stress to the forest resources and increased the area of
A reduction in land tax should be offered to farmers the cropland cover leading to the LULC modification.
who accept courses of training and prove that they According to key informants, the major causes of increase
have learned from them when extension agents in population pressure in the study area are the natural
insist to monitor their progress. The conservation reproduction of indigenous populations, in-migration,
of the land use should be in the form of positive urbanisation and resettlement. Data synthesised from
practices, rather than simply prohibiting certain bad key informant interviews and FGDs alluded to population
ones (FAO, 1986). The government needs to coordi­ growth as the most important driver of the demographic
nate various departments in order to manage element that is triggering LULC change. As noted in the
resources sustainably. Moreover, Ethiopia needs to Ethiopian Central Statistics report, the area under study
adopt an effective population policy to reduce the has seen a rapid increase in population in the past four
high fertility rate through family planning. and a half decades, which leads to an increase in food
Agricultural lands become scarce and fragmented consumption from agricultural expansion (Figure 8).
due to overpopulation, which further contributes Rapid population growth resulted in growing farmland
to resource pressures and environmental degrada­ and the destruction of forests. An increase in population
tion. The policy conserving high slopes for forests will negatively affect the natural environment due to the
should be intensified by re-afforestation, promoting high demand for resources.
more suitable soil and water conservation practices. LULC changes are dynamic and nonlinear, which
Land degradation must avoided just as women’s means that conversion from one land use to another
health must be protected from degradation. does not follow the same pattern owing to natural or
manmade reasons such as policy changes, population
expansion, and a loss in land productivity. The military
3.7. Perceptions on LULC Change from the local
administration announced the nationalisation of all rural
community
land by removing private and common property rights,
All the respondents (N = 180) witnessed population giving all land use rights (Meshesha et al., 2014). This
growth and the expansion of cultivated land as the fore­ policy made the state the only owner of the property
most drivers of LULC alterations. According to the 2007 and land-related resources, rendering the state incapable
 ALL EARTH 327

2400 130000

2200 Crop land 120000

population
2000

Population growth
Croplands area in (ha)
110000
1800

1600 100000

1400
90000
1200

80000
1000

5 0 5 0 5 0 5 0 5
198 199 199 200 200 201 201 202 202
Year

Figure 8. Population and croplands patterns from 1986 to 2018 in study landscape.

of monitoring and enforcing regulations. As a result of use from fuelwood to crop residue, cow dung, and
poor land management techniques, forest area has been eucalyptus plantations due to deforestation.
converted to settlements and cropland and has been Historically, 95% of the households utilised the natural
severely degraded (Mekasha et al., 2014; Tolessa et al., forest and shrubland for energy sources before
2017). The current government of Ethiopia likewise con­ 20 years ago, but today a few of them have access to
tinues the same situation, where land is owned by natural forest and shrubland to collect wood for cook­
Ethiopia’s nations, nationalities, and people, in terms of ing and heating. In the focus group discussion, female
Article 40 (3) of the constitution was ratified in (FDRE, respondents explained that previously, when the
1994). As a result, many land-use categories may be women collected fuelwood from the natural forest,
simply converted to cultivated land. In Ethiopia, for exam­ they took chances for socialisation and informal com­
ple, peasants possess more legal rights over forests if they munication with their friends and discussed social
convert them to cropland, as the law declares forests as issues. Currently, they consider deforestation as
state property. Farmers are compelled to legally or illeg­ a disadvantage since they have lost such communica­
ally change forests into cultivation land in order to secure tion. In the interview sessions, women are faced with
the usage of the land for an undetermined period of time. long distances to travel to fetch water for household
According to LULC trend analysis, forest cover declined consumption since the nearby water sources have
over the study period (1973–2018). This is consistent with dried up, and climate change is cited as a cause for
the findings of other LULC studies done in Ethiopia the drying of small brooks and springs. The male farm­
(Meshesha et al., 2014; Minale and Rao, 2012; Tsegaye ers in the present-day move their livestock long dis­
et al., 2010) and other tropical areas (Lira et al., 2012; tances to reach water sources for drinking.
Oestreicher et al., 2014).
Because of government and policy changes, forests
4. Conclusion and recomendation
were converted into cropland, and settlement resulted
in the reduction of ecosystem services. The extensive Remote sensing data analysis indicated a notable
use of forestland in tropical countries has led to an decline in forest cover and a considerable increase in
adverse impact on forest ecosystems as evident by cropland in the study landscape during the last four
several studies in relation to agricultural policy (Lira and a half decades. The trend of LULC alterations
et al., 2012; Oestreicher et al., 2014). According to the viewed by respondents was congruent with the find­
current study area, forestland has been lost and crop­ ings reported from remote sensing image analysis.
land and settlements have expanded. The current Seven LULC categories were identified in the current
demographic changes and the continued variation in study where significant land-use changes occurred in
LULC coupled with uncertain climate conditions, sig­ the area throughout the reference years of 1973, 1986,
nificantly affect livelihoods and put the farmer’s pro­ 2000, and 2018. Forestland declined from 19, 318.62 ha
duction system under pressure. The discussions also (43.1%) in 1973 to 4850.91 ha (10.8%) in 2018. The
revealed that some native trees, which were used for study further revealed that the total amount of forest­
house construction, are in danger, like Prunes Africana, land cleared between 1973 and 2018 was estimated to
Podocarpus falcatus, and Cordia Africana. Furthermore, be 14,467.7 ha (74.9%); similarly, grasslands declined
the area also experienced a shift in household energy from (39.2%) in 1973–1986 to (86.2%) in 1973–2018 ha/
328 M. MARIYE ET AL.

year over the last four and a half decades, respectively. Disclosure statement
Conversely, the settlement land has risen from 1003%
No potential conflict of interest was reported by the
to 55,269% during a similar study period. The findings author(s).
revealed that forest and grasslands were the most
predominant LULC type in the earlier decade, but it is
now preceded by croplands. There has been an Author’s contribution
increasing trend towards croplands, settlements, and Conceptualisation: Melesse Maryo, Data curation: Mehari
bare lands throughout the entire study period at the Mariye, Formal analysis: Mehari Mariye, Investigation:
expense of other LULC classes. This implies that the Mehari Mariye, Methodology: Mehari Mariye, Software:
natural resource conservation measures that have Mehari Mariye, Supervision: Melesse Maryo and Li Jianhua,
been practiced by the governmental and non- Validation: Melesse Maryo and Li Jianhua, Writing –original
draft: Mehari Mariye, Writing – review and editing: Mehari
governmental organisations in the study area did not
Mariye, Melesse Maryo and Li Jianhua. All authors reviewed
bring the desired conservation effects. the results and approved the final version of the manuscript.
The majority of the respondents perceived popula­ The author declares responsibility for conceiving and design­
tion growth, settlement, agricultural expansion, and ing the study, collecting data, analysing and interpreting the
charcoal productions were recognised as dominant results, and drafting the manuscript.
drivers of LULC change in the study area. If the LULC
trend continues, it will have major environmental and References
economic ramifications, as well as a negative influence
on local people’s livelihoods. In order to maintain sus­ Abate, S. (2011). Evaluating the land use and land cover
tainable rural livelihoods, appropriate land resource dynamics in borena woreda of south wollo highlands,
Ethiopia. Journal of Sustainable Development in Africa, 13
management policies, and designing appropriate (1), 87–107.
population strategies that are based on community Abebe, S. K. (2018). Assessment of land use/ land cover
level should be considered. change using GIS and remote sensing techniques: A case
To reduce and mitigate the rapid rates of LULC con­ study of dendi district Oromiya Regional State, Ethiopia.
versions at local, distant downstream systems and else­ Journal of Environment and Earth Science, 8, 123–128.
Agidew, A. A., & Singh, K., N. (2017). The implications of land
where with similar social-ecological settings, the
use and land cover changes for rural household food
application of integrated LULC management strategies insecurity in the Northeastern highlands of Ethiopia: The
is critically important. In this regard, it is suggested that case of the Teleyayen sub-watershed. Agriculture & Food
(1) Promote the rehabilitation and conservation of the Security, 6(56). https://doi.org/10.1186/s40066-017-0134-4
natural vegetation covers and environmentally sensi­ Alemu, B., Garedew, E., Eshetu, Z., & Kassa, H. 2015.
tive areas, so as to benefit both the upstream and International Research Journal of Agriculture and Soil
Science. International Research Journal of Agriculture and
downstream landscape dwellers. (2) Control the expan­ Soil Science, 5, 1, 28–44. https://doi.org/10.14303/irjas.
sion of crop cultivation into marginal lands including Amanuel, A., & Mulugeta, L. (2014). Detecting and quantify­
steep slopes through the application of smart agricul­ ing land use/ land cover dynamics in Nadda Asendabo
tural practices that increase the efficiency of using lim­ Watershed, South Western Ethiopia. International Journal
ited space with increased crop production. (3) Integrate of Environmental Sciences, 3(1), 45–50.
Amare, S. (2015). Land use/cover change at infraz watershed,
multipurpose perennial highland fruit tree plantation
northwestern Ethiopia. Journal of Landscape Ecology, 8(1),
with Enset dominant agroforestry system. Further 78–83. https://doi.org/10.1515/jlecol-2015-0005
more, we recommend that policy instruments such as Andualem, T. G., Belay, G., & Guadie, A. (2018). Land use
tax incentives for those who take care of their lands, change detection using remote sensing technology.
support for off-farm economic activities in rural areas, Journal of Earth Science & Climatic Change, 9(10), 1–6.
increase agricultural production through provision of https://doi.org/10.4172/2157-7617.1000496
Angessa, A. T., Lemma, B., & Yeshitela, K. (2019). Land-use and
affordable inputs, and implement the land registration land-cover dynamics and their drivers in the central high­
and certification process as quickly as possible. lands of Ethiopia with special reference to the Lake
Wanchi watershed. GeoJournal, 86(3), 1225–1243. https://
doi.org/10.1007/s10708-019-10130-1
Acknowledgments Asmamaw, B. L., Mohammed, A. A., & Lulseged, D. T. (2014).
Land use/cover dynamics and their effects in the Gerado
We stretch out our appreciation to the KTz district catchment, northeastern Ethiopia. International Journal of
Agricultural and Rural Development Office, farmers, and Environmental Studies, 68(6), 883–900. https://doi.org/10.
development agents, particularly from Doyogena, Angecha, 1080/00207233.2011.637701
Damiboya, and Kacha Bira districts, who gave us pertinent Asokan, A., & Anitha, J. (2019). Change detection techniques
data on the LULC management scheme of the study area and for remote sensing applications: A survey. Earth Science
its adjacent landscape. We also recognise the Ethiopian Informatics, 12(2), 143–160. https://doi.org/10.1007/
Meteorological Services and Ethiopian Space Science and s12145-019-00380-5
Geospatial Information Institute for providing us with data Asres, R. S., Tilahun, S. A., Ayele, G. T., & Melesse, A. M. (2016).
on precipitation, satellite information, and historical land use Analyses of land use/land cover change dynamics in the
map, respectively. upland watersheds of upper blue Nile Basin. Springer
 ALL EARTH 329

International Publishing, 73–91. https://doi.org/10.1007/ Burgi, M., Bieling, C. K., Hackwitz, V., Kizos, T., Lieskovsky, J.,
978-3-319-18787-7_5 Martı´n, M. G., McCarthy, S., Mu Ller, M., Palang, H.,
Assen, M., & Nigussie, T. (2009). Land use/cover changes Plieninger, T., & Printsmann, A. (2017). Processes and driv­
between 1966 and 1996 in Chirokella micro-watershed, ing forces in changing cultural landscapes across Europe.
Southeastern Ethiopia. East African Journal of Sciences, 3 Landscape Ecol, 32(11), 2097–2112. https://doi.org/10.
(1), 1–8. https://doi.org/10.4314/eajsci.v3i1.42778 1007/s10980-017-0513-z
Attri, P., Chaudhry, S., & Sharma, S. (2015). Remote Sensing & Congalton, R. G. (2001). Accuracy assessment and validation
GIS based Approaches for LULC Change Detection. of remotely sensed and other spatial information.
International Journal of Current Engineering and International Journal of Wildland Fire, 10(4), 321–328.
Technology, 5, 2347–5161. https://doi.org/10.1071/wf01031
Ayele, G. T., Tebeje, A. K., Demissie, S. S., Belete, M. A., Congalton, R. G., & Green, K. (2009). Assessing the accuracy of
Jemberrie, M. A., Teshome, W. A., Mengistu, D. T., & remotely sensed data: Principles and practices (2nd ed.).
Teshale, E. Z. (2018). Time series land cover mapping and Boca Raton: Taylor & Francis, 11(6), 448–449. https://doi.
change detection analysis using geographic information org/10.1016/j.jag.2009.07.002
system and remote sensing, Northern Ethiopia. Air, Soil CSA., (2007). Summary and statistical report of the 2007 popu­
and Water Research, 11, 1–18. https://doi.org/10.1177/ lation and housing census results.
1178622117751603 CSA. (2013). Population projection of Ethiopia for all regions at
Bekele, D., Alamirew, T., Kebede, A., Zeleke, G., & woreda level from 2014-17. Central Statistical Agency.
Melesse, A. M. (2018). Land use and land cover dynamics Daniel, A. M. (2008). Remote sensing and GIS-based Land use
in the Keleta watershed, Awash River basin, Ethiopia. and land cover change detection in the upper Dijo river
Environmental Hazards, 18(3), 246–265. https://doi.org/10. catchment, Silte zone, southern Ethiopia. 1–34.
1080/17477891.2018.1561407 Daniel, B., Tena, A., Asfaw, K., Gete, Z., & Assefa, M. M. 2018.
Belay, T., & Mengistu, D. A. (2019). Land use and land cover Land use and land cover dynamics in the Keleta
dynamics and drivers in the Muga watershed, Upper Blue watershed, Awash River basin, Ethiopia. Environmental
Nile basin, Ethiopia. Remote Sensing Applications: Society hazards, 18, 3, 246–265. https://doi.org/10.1080/
and Environment, 15. https://doi.org/10.1016/j.rsase.2019. 17477891.2018.1561407.
100249 Daye, D. D., & Healey, J. R. (2015). Impacts of land-use change
Belayneh, B., Eyasu, E., & Medina-Solis, C. E. (2021). Land use/ on sacred forests at the landscape scale. Global Ecology
land cover change and its driving forces in Shenkolla and Conservation, 3, 349–358. https://doi.org/10.1016/j.
Watershed, South Central Ethiopia. The Scientific World gecco.2014.12.009
Journal, 2021, 1–13. https://doi.org/10.1155/2021/9470918 Degifea, A., et al, Worku, H., & Gizaw, S., et al (2019). Land use
Belayneh, Y., Ru, G., Guadie, A., Teffera, Z. L., & Tsega, M. land cover dynamics, its drivers and environmental impli­
(2018). Forest cover change and its driving forces in cations in Lake Hawassa Watershed of Ethiopia. Remote
FagitaLekoma District, Ethiopia. Journal Forest Resources, Sensing Applications: Society and Environment, 14, 178–190.
31(5), 1567–1582. https://doi.org/10.1007/s11676-018- Degife, A. W., Zabel, F., & Mauser, W. (2018). Assessing land
0838-8 use and land cover changes and agricultural farmland
Belete, F., Maryo, M., & Teka, A. (2021). Land use/land cover expansions in Gambella Region, Ethiopia, using Landsat
dynamics and perception of the local communities in Bita 5 and Sentinel 2a multispectral data. Heliyon, 4(11), 1–28.
district, south western Ethiopia. International Journal of https://doi.org/10.1016/j.heliyon.2018
River Basin Management, 1–12. https://doi.org/10.1080/ Demissie, F., Yeshitila, K., & Kindu, M. (2017). Land use/land
15715124.2021.1938092 cover changes and their causes in Libokemkem District of
Berihun, M. L., Tsunekawa, A., Haregeweynm, N., South Gonder, Ethiopia remote sensing applications.
Meshesha, D. T., Adgo, E., Tsubo, M., Masunagaf, T., Society and Environment, 8, 224–230. https://doi.org/10.
Fenta, A. A., Sultan, D., & Yibeltal, M. (2019). Exploring 1016/j.rsase.2017.10.001
land use/land cover changes, drivers and their implica­ Deribew, K.T., & Dalacho, D.W. (2019). Land use and forest
tions in contrasting agro-ecological environments of cover dynamics in the North-eastern Addis Ababa, central
Ethiopia. Land Use Policy, 87, 1–15. https://doi.org/10. highlands of Ethiopia. Environmental Systems Research, 8.
1016/j.landusepol.2019.104052 Dewan, A., & Corner, R. (2014). Dhaka megacity: Geospatial
Betru, T., et al, Tolera, M., & Sahle, K., et al (2019). Trends and Perspectives on Urbanisation, environment and health.
drivers of land use/land cover change in Western Ethiopia. Springer Science & Business Media.
Applied Geography, 104, 83–93. Dibaba, W. T., Demissie, T. A., & Miegel, K. (2020). Drivers and
Binyam, A., Efrem, G., Zewdu, E., & Kassa, M. (2015). Land use implications of land use/land cover dynamics in finchaa
and land cover changes and associated driving forces in Catchment, Northwestern Ethiopia. Land, 9(113), 1–20.
North Western Lowlands of Ethiopia. International https://doi.org/10.3390/land9040113
Research Journal of Agriculture and Soil Science, 5(1), DWARDO. (2019). Doyogena Woreda Agriculture and Rural
28–44. https://doi.org/10.14303/irjas.2014.063 Development Office. Information on socio-economic activ­
Birhane, E., Ashfare, H., Fenta, A. A., Hishe, H., ities in Kembata Tembaro zone.
Gebremedhin, M. A., Wahed, H. G., & Solomon, N. (2019). Dwivedi, R. S., Sreenivas, K., & Ramana, K. V. (2007). Cover:
Land use land cover changes along topographic gradients Land-use land-cover change analysis in part of data.
in Hugumburda national forest priority area, Northern International Journal of Remote Sensing, 26(7), 1285–1287.
Ethiopia. Remote Sensing Applications: Society and https://doi.org/10.1080/01431160512331337763
Environment, 13, 61–68. https://doi.org/10.1016/j.rsase. Elias, E., Seifu, W., Tesfaye, B., Girmay, W., & Tejada Moral, M.
2018.10.017 (2019). Impact of land use/cover changes on lake ecosys­
Braimoh, A.K. (2006). Random and systematic land-cover tem of Ethiopia central rift valley. Cogent Food &
transitions in northern Ghana. Agriculture, Ecosystems & Agriculture, 5(1), 1–20. https://doi.org/10.1080/23311932.
Environment, 113, 254–263. 2019.1595876
330 M. MARIYE ET AL.

EMS. (2013) . ethiopian Meteorological Service Database. Addis Keenan, R. J., Reams, G. A., Achard, F., de Freitas, J. V.,
Ababa. Grainger, A., & Lindquist, E. (2015). Dynamics of global
Eshetu, A. A. (2014). Forest resource management systems in forest area: Results from the FAO global forest resources
Ethiopia: Historical perspective. International Journal of assessment 2015. Forest Ecology and Management, 352,
Biodiversity and Conservation, 6(2), 121–131. https://doi. 9–20. https://doi.org/10.1016/j.foreco.2015.06.014
org/10.5897/IJBC2013.0645 Kindu, M., Schneider, T., Teketay, D., & Knoke, T. (2013). Land
FAO., (1986). The highlands reclamation study – Ethiopia final use/land cover change analysis using object-based classi­
report. Food and Agriculture Organization of the United fication approach in munessa-shashemene landscape of
Nations, Rome.Volumes 1 and 2. the Ethiopian highlands. Remote Sensing, 5(5), 2411–2435.
FAO., (2010). Global forest resources assessment country report https://doi.org/10.3390/rs5052411
Ethiopia PP43. Kindu, M., Schneider, T., Teketay, D., & Knoke, T. (2015). Drivers of
FAO. (2016). State of the world’s forests 2016. Forests and agri­ land use/land cover changes in Munessa-Shashemene land­
culture: Land-use challenges and opportunities. Rome, 107. scape of the south-central highlands of Ethiopia.
Fasika, A., Motuma, T., & Gizaw, T. (2019). Land Use Land Environmental Monitoring and Assessment, 187(7), 1 17.
Cover Change Trend and Its Drivers in Somodo https://doi.org/10.1007/s10661-015-4671-7
Watershed South Western, Ethiopia. African Journal of Kothari, C. R. (2004). Research methodology methods & tech­
Agricultural Research, 14, 102–117. niques (2nd) ed.). New Age International (P) Limited.
F.D.R.E. (1994). Federal Democratic Republic of Ethiopia (FDRE), Kotoky, P., Dutta, M. K., & Borah, G. C. (2012). Changes in
1994, Constitution of the Federal Democratic Republic of Landuse and Landcover along the Dhansiri river channel
Ethiopia (p. 50). Addis Ababa. assam – A remote sensing and GIS approach. Journal
Gashaw, T., Tulu, T., & Argaw, M. (2017). Erosion risk assessment Geological Society of India, 79(1), 61–68. https://doi.org/
for prioritization of conservation measures in Geleda 10.1007/s12594-012-0002-6
watershed, Blue Nile basin, Ethiopia. Environmental Systems K.T.D.A.R.D.O. (2019). Kembata Tembaro District Agriculture
Research, 6(1), 1–14. https://doi.org/10.1186/s40068-016- and Rural Development Office. Information on Socio-
0078-x Economic and Geo-Spatial Annual Statistical Abstract.
Gebrelibanos, T., & Assen, M. (2013). Land use/land cover southren central Ethiopia.
dynamics and their driving forces in the Hirmi watershed Lambin, E. F., & Geist, H. J. (2006). Land-use and landcover
and its adjacent agro-ecosystem, highlands of Northern change: Local processes and global impacts. Springer
Ethiopia. Journal of Land Use Science, 10(1), 81–94. https:// Science & Business Media.
doi.org/10.1080/1747423X.2013.845614 Lira, et al (2012). Land-use and land-cover change in Atlantic
Gebremicael, T. G., Mohamed, Y. A., van der Zaag, P., & Forest landscapes. Forest Ecology and Management, 278,
Hagos, E. Y. (2018). Quantifying longitudinal land use 80–89.
change from land degradation to rehabilitation in the Lu, D., Mausel, P., Brondi´Zio, E., & Moran, E. (2004). Change
headwaters of Tekeze-Atbara Basin, Ethiopia. Science of detection techniques. International Journal of Remote
the Total Environment, 622-623, 1581–1589. https://doi. Sensing, 25(12), 2365–2407. https://doi.org/10.1080/
org/10.1016/j.scitotenv.2017.10.034 0143116031000139863
Gurmessa, F. (2015). Forest loss and climate change in Lu, D., et al, Mausel, P., & Brondízio, E., et al (2010). Change
Ethiopia. Research Journal of Agriculture and detection techniques. International Journal of Remote
Environmental Management, 4(5), 216–224. Sensing, 25, 2365–2401. https://doi.org/10.1080/
Hailu, A., Mammo, S., & Kidane, M. (2020). Dynamics of land 0143116031000139863
use, land cover change trend and its drivers in Jimma Lu, D., & Weng, Q. (2007). A survey of image classification
Geneti District, Western Ethiopia. Land Use Policy, 99, methods and techniques for improving classification
1–11. https://doi.org/10.1016/j.landusepol.2020.105011 performance. International Journal of Remote Sensing, 28
Haque, I. M., & Basak, R. (2017). Land cover change detection (5), 823–870. https://doi.org/10.1080/01431160600746456
using GIS and remote sensing techniques: A spatio-temporal Mariye, M., Maryo, M., Changming, Y., Lakew Teffera, Z.,
study on Tanguar Haor, Sunamganj, Bangladesh. The Weldegebrial, B. (2020). Effects of land use and land
Egyptian Journal of Remote Sensing and Space Sciences, 20 cover change on soil erosion potential in Berhe district:
(2), 251–263. x. https://doi.org/10.1016/j.ejrs.2016.12.003 A case study of Legedadi watershed, Ethiopia.
Hassan, Z., Shabbir, R., Ahmad, S. S., Malik, A. H., Aziz, N., International Journal of River Basin Management, 20(1).
Butt, A., & Erum, S. (2016). Dynamics of land use and land https://doi.org/10.1080/15715124.2020.1767636.
cover change (LULCC) using geospatial techniques: A case Mariye, M., Maryo, M., & Jianhua, L. (2021). The study of land
study of Islamabad Pakistan. SpringerPlus, 5(1), 1–12. use and land cover (LULC) dynamics and the perception of
https://doi.org/10.1186/s40064-016-2414-z local people in Aykoleba, Northern Ethiopia. African
Hassen, E. E., & Assen, M. (2017). Land use/cover dynamics Journal of Environmental Science and Technology, 15(7),
and its drivers in Gelda catchment, Lake Tana watershed, 282–297. https://doi.org/10.5897/AJEST2021.3022
Ethiopia. Environmental Systems Research, 6(4), 1–13. Mariye, M., Maryo, M., & Li, J. (2022). The Study of Land Use
https://doi.org/10.1186/s40068-017-0081-x and Land Cover (LULC) dynamics and the perception of
Houghton, R. A., House, J. I., Pongratz, J., van Derwerf, G. R., local people in Aykoleba, Northern Ethiopia. Journal of the
DeFries, R. S., Hansen, M. C., Le Quere, C., & Ramankutty, N. Indian Society of Remote Sensing, 50(5), 775–789. https://
(2012). Carbon emissions from land use and land-cover doi.org/10.1007/s12524-021-01462-y
change. Biogeosciences, 9(12), 5125–5142. https://doi.org/ Martínez, M.L., et al, Pérez-Maqueo, O., & Vázquez, G., et al
10.5194/bg-9-5125-2012 (2009). Effects of land use change on biodiversity and
Jianya, G., Haigang, S., Guorui, M., & Qiming, Z. 2008. A review ecosystem services in tropical montane cloud forests of
of multi-temporal remote sensing data change detection Mexico. Forest Ecology and Management, 258, 1856–1863.
algorithms. The International Archives of the Mary, T., Ekwal, I., & Tahir, H. (2013). Evaluation of land use/land
Photogrammetric, Remote Sensing and Spatial Information cover changes in Mekelle City, Ethiopia using Remote Sensing
Sciences Beijing, China, XXXVII Part B7. and GIS. Computational Ecology and Software, 3(1), 9–16.
 ALL EARTH 331

Maryo, M. (2020). Agrobiodiversity in Enset-based agricultural the Brazilian Amazon: Quantitative characterization and
landscape of Kambatta Tembaro. International Book qualitative insights into the influence of knowledge,
Market Service Ltd,Omniscriptum Publishing Group. values, and beliefs. Human Ecology, 42(2), 521–540.
Mengistu, D. A., Waktola, D. K., & Woldetsadik, M. (2012). https://doi.org/10.1007/s10745-014-9667-3
Detection and analysis of land-use and land-cover Oettera, D. R., Cohenb, W. B., Berterretchea, M.,
changes in the Midwest escarpment of the Ethiopian Rift Maiersperger, T. K., & Kennedy, R. E. (2000). Land cover map­
Valley. Journal of Land Use Science, 7(3), 239–260. https:// ping in an agricultural setting using multiseasonal Thematic
doi.org/10.1080/1747423X.2011.562556 Mapper data. Remote Sensing of Environment, 76(2), 139–155.
Meshesha, T. W., Tripathi, S. K., & Khare, D. (2016). Analyses of https://doi.org/10.1016/S0034-4257(00)00202-9
land use and land cover change dynamics using GIS and Olorunfemi, I. E., Fasinmirin, J. T., Olufayo, A. A., & Komolafe, A. A.
remote sensing during 1984 and 2015 in the Beressa (2018). GIS and remote sensing-based analysis of the impacts
Watershed Northern central highland of Ethiopia. of land use/land cover change (LULCC) on the environmental
Modeling Earth Systems and Environment, 2(168), 1–12. sustainability of Ekiti State, southwestern Nigeria.
https://doi.org/10.1007/s40808-016-0233-4 Environment, Development and Sustainability, 22(2), 661–692.
Meshesha, D.T., et al, Tsunekawa, A., & Tsubo, M., et al (2014). https://doi.org/10.1007/s10668-018-0214-z
Land-use change and its socio-environmental impact in Othow, O. O., Legesse, G. S., & Obsi, G. D. (2017). Analyzing the
Eastern Ethiopia’s highland. Regional Environmental Rate of land use and land cover change and determining the
Change, 14, 757–768. causes of forest cover change in Gog District, gambella regio­
Meyfroidt, P. (2013). Environmental cognitions, land change, nal state, Ethiopia. Journal of Remote Sensing & GIS, 6(4).
and social–ecological feedbacks: An overview. Journal of https://doi.org/10.4172/2469-4134.1000219
Land Use Science, 8, 341–367. https://doi.org/10.1080/ Peng, G., Xiang, N., Bing, W., & Zheng, Y. (2015). Land use
1747423X.2012.667452 changes and its driving forces in hilly ecological restora­
Miheretu, B. A., & Yimer, A. A. (2017). Land use land cover tion area based on gis and rs of northern China. Scientific
changes and their environmental implications in the Reports, 5, 1–11. https://doi.org/10.1038/srep11038
Gelana sub-watershed of Northern highlands of Ethiopia. Phalan, B., Onial, M., Balmford, A., & Green, R. E. (2011).
Environmental Systems Research, 6(7). https://doi.org/10. Reconciling food production and biodiversity conservation:
1186/s40068-017-0084-7 Land Sharing and Land sparing compared. Science, 333(6047),
Mikias, B. M. (2015). Land Use/Land Cover dynamics in the 1289–1291. https://doi.org/10.1126/science.1208742
central rift valley region of Ethiopia: case of arsi Negele Pontius, R.G., Shusas, E., & McEachern, M. (2004). Detecting
District. African Journal of Agricultural Research, 10(5), important categorical land changes while accounting for
434–449. https://doi.org/10.5897/AJAR2014.8728 persistence. Agriculture, Ecosystems & Environment, 101,
Minale, A. S. (2013). Retrospective analysis of land cover and 251–268.
use dynamics in gilgel abbay watershed by using GIS and Rahman, M. T. (2016). Detection of Land Use/land cover
remote sensing techniques, Northwestern Ethiopia. changes and Urban Sprawl in Al-Khobar, Saudi Arabia:
International Journal of Geosciences, 4(7), 1003–1008. An analysis of multi-temporal remote sensing data. ISPRS
https://doi.org/10.4236/ijg.2013.47093 International Journal of Geo-Information , 5(15), 2–17.
Minale, A.S., & Rao, K.K. (2012). Impacts of land cover/use https://doi.org/10.3390/ijgi5020015
dynamics of Gilgel Abbay catchment of Lake Tana on Rawat, J. S., & Kumar, M. (2015). Monitoring land use/cover
climate variability, Northwestern Ethiopia. Applied change using remote sensing and GIS techniques: A case
Geomatics, 4, 155–162. study of Hawalbagh block, district Almora, Uttarakhand,
MoA. (1995). Land use systems and soil conditions of Ethiopia. India. The Egyptian Journal of Remote Sensing and Space
Moges, K. B., Amare, S. M., & Alemu, B. D. (2015). Sciences, 18(1), 77–84. https://doi.org/10.1016/j.ejrs.2015.
Multitemporal Land use land cover change and dynamics 02.002
of blue nile basin by using GIS and remote sensing tech­ Schaefer, M., & Thinh, N. X. (2019). Evaluation of land cover
niques, North-Western Ethiopia. International Journal of change and agricultural protection sites: A GIS and remote
Environmental Sciences, 4(2), 81–88. sensing approach for Ho Chi Minh City, Vietnam. Heliyon, 5
Mohajane, M., Essahlaoui, A., Oudija, F., El Hafyani, M., El (5), e01773. https://doi.org/10.1016/j.heliyon.2019.e01773
Hmaidi, A., El Ouali, A., Randazzo, G., & Teodoro, A. A. Shawul, A. A., & Chakma, S. (2019). Spatiotemporal detection of
(2018). Land use/land cover (LULC) using Landsat data land use/land cover change in the large basin using inte­
series (MSS, TM, ETM+ and OLI) in azrou forest, in the grated approaches of remote sensing and GIS in the Upper
central middle atlas of Morocco. Environments, 5(131), Awash basin, Ethiopia. Environmental Earth Sciences, 78(141),
1–16. https://doi.org/10.3390/environments5120131 1–13. https://doi.org/10.1007/s12665-019-8154-y
Munthali, M. G., Davis, N., Adeola, A. M., Botai, J. O., Shiferaw, A., & Singh, K. L. (2011). Evaluating the land use and
Kamwi, J. M., Chisale, H. L. W., & Orimoogunje, O. O. I. land cover dynamics in Borena Woreda South Wollo
(2019). Local perception of drivers of land-use and Highlands, Ethiopia. J. Bus. Econ, 2(1), 69–104. 2. EJBE
land-cover change dynamics across Dedza District, Singh, A. (1989). Review Article Digital change detection
Central Malawi region. Sustainability, 11(3), 832. https:// techniques using remotely-sensed data. International
doi.org/10.3390/su11030832 Journal of Remote Sensing, 10(6), 989–1003. https://doi.
Muriithi, F. K. (2016). Land use and land cover (LULC) changes in org/10.1080/01431168908903939
semi-arid sub-watersheds of Laikipia and Athi River basins, Siraj, M., Zhang, K., & Moges, K. (2018). Retrospective analysis
Kenya, as influenced by expanding intensive commercial of land use land cover dynamics using GIS and remote
horticulture. Remote Sensing Applications; Society and sensing in central highlands of Ethiopia. Journal of
Environment, 3, 73–88. https://doi.org/10.1016/j.rsase.2016. Landscape Ecology, 11(2), 31–52. https://doi.org/10.2478/
01.002 jlecol-2018-0005
Oestreicher, J. S., Farella, N., Paquet, S., Davidson, R., TekleL, K., & Hedlund, L. (2000). Land cover changes between
Lucotte, M., Mertens, F., & Saint-Charles, J. (2014). 1958 and 1986 in Kalu District, Southern Wello, Ethiopia.
Livelihood activities and land-use at a riparian frontier of Mountain Research and Development, 20(1), 42–51.
332 M. MARIYE ET AL.

Temesgen, H., Nyssen, J., Zenebe, A., Haregeweyn, N., Twisa, S., & Buchroithner, M. F. (2019). Land-use and
Kindu, M., Lemenih, M., & Haile, M. (2013). Ecological suc­ land-cover (LULC) change detection in Wami River Basin,
cession and land use changes in a lake retreat area (Main Tanzania. Land, 8(136), 1–15. https://doi.org/10.3390/
Ethiopian Rift Valley). Journal of Arid Environments, 91, land8090136
53–60. https://doi.org/10.1016/j.jaridenv.2012.12.001 USGS. 2019. Landsat Collection 1 Level 1 product definition.
Tesfaye, S., Guyassa, E., Raj, J. A., Birhane, E., & Wondim, G. T. Survey, D.o.t.I.U.S.G Ed.
(2014). Land Use and Land Cover Change, and woody WoldeYohannes, A., Cotter, M., Kelboro, G., & Dessalegn, W.
vegetation diversity in human driven landscape of Gilgel (2018). Land use and land cover changes and their effects
Tekeze Catchment, Northern Ethiopia. International on the landscape of Abaya-Chamo Basin, Southern Ethiopia.
Journal of Forestry Research, 1, 1–10. dx. https://doi.org/ Land, 7(2), 2. https://doi.org/10.3390/land7010002
10.1155/2014/614249 Wubie, M. A., Assen, M., & Nicolau, M. (2016). Patterns, causes
Tewabe, D., Fentahun, T., Li, F., & Li, F. (2020). Assessing land and consequences of land use/cover dynamics in the
use and land cover change detection using remote sen­ Gumara watershed of lake Tana basin, Northwestern
sing in the Lake Tana Basin, Northwest Ethiopia. Cogent Ethiopia. Environmental Systems Research, 5(8), 1–12.
Environmental Science, 6(1), 1778998. https://doi.org/10. https://doi.org/10.1186/s40068-016-0058-1
1080/23311843.2020.1778998 Yesuph, A. Y., & Dagnew, A. B. (2019). Land use/cover
Thakkar, A. K., Desai, V. R., Patel, A., & Potdar, M. B. (2017). spatiotemporal dynamics, driving forces and implica­
Post-classification corrections in improving the classifica­ tions at the Beshillo catchment of the Blue Nile Basin,
tion of Land Use/Land Cover of arid region using RS and North Eastern Highlands of Ethiopia. Environmental
GIS: The case of Arjuni watershed, Gujarat. India the Systems Research, 8(21), 1–30. https://doi.org/10.1186/
Egyptian Journal of Remote Sensing and Space Sciences, s40068-019-0148-y
20(1), 79–89. https://doi.org/10.1016/j.ejrs.2016.11.006 Yohannes, H., Soromessa, T., Argaw, M., & Dewan, A.
Tilahun, A. (2015). Accuracy Assessment of Land Use Land (2020). Changes in landscape composition and config­
Cover Classification using Google Earth. American Journal uration in the Beressa watershed, Blue Nile basin of
of Environmental Protection, 4, 193. Ethiopian Highlands: Historical and future exploration.
Tolessa, T., Senbeta, F., & Kidan, M. (2017). The impact of land Heliyon, 6(9), 1–17. https://doi.org/10.1016/j.heliyon.
use/land cover change on ecosystem services in the cen­ 2020.e04859
tral highlands of Ethiopia. Ecosystem Services, 23, 47–54. Yuan, F., Bauer, M. E., Heinert, N. J., & Holden, G. R. (2005).
https://doi.org/10.1016/j.ecoser.2016.11.010 Multi-level land cover mapping of the Twin Cities
Tsegaye, et al (2010). Land-use/cover dynamics in Northern (Minnesota) metropolitan area with multi-seasonal
Afar rangelands, Ethiopia. Agriculture, Ecosystems & Landsat TM/ETM+ Data. Geo-Carto International, 20(2),
Environment, 139, 174–180. 5–14. https://doi.org/10.1080/10106040508542340
Tsehaye, G., & Mohammed, A. (2015). Land use land cover Zeleke, G., & Hurni, H. (2001). Implications of land use and
dynamics and their driving forces in the Hirmi watershed land cover dynamics for mountain resource degradation
and its adjacent agro ecosystem highlands of Northern in the Northwestern Ethiopian Highlands. Mountain
Ethiopia. Journal of Land Use Science, 10(1), 81–94. https:// Research and Development, 21(2), 184–191.
doi.org/10.1080/1747423X.2013.845614 Zewdie, W., & Csaplovies, E. (2017). Remote Sensing based
Tucker, C. J., Grant, D. M., & Dykstra, J. D. (2004). NASA’s multi-temporal land cover classification and change
Global Orthorectified Landsat data set. Photogrammetric detection in northwestern Ethiopia. European Journal of
Engineering and Remote Sensing, 70(3), 313–322. https:// Remote Sensing, 48(1), 121–139. https://doi.org/10.5721/
doi.org/10.14358/PERS.70.3.313 EuJRS20154808