Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Spatio-temporal variation on land use and land cover and its drivers under watershed management activities in Becho, Central Highlands of Ethiopia

Spatio-temporal variation on land use and land cover and its drivers under watershed management... Environmental & Socio-economic Studies DOI: 10.2478/environ-2022-0021 Environ. Socio.-econ. Stud., 2022, 10, 4: 22-34 ________________________________________________________________________________________________ Original article Spatio-temporal variation on land use and land cover and its drivers under watershed management activities in Becho, Central Highlands of Ethiopia 1, 2 2 3 Mahammed Endrias Hailu *, Mohammed Assen Ahmed , Temesgen Argaw Naji Ambo University, College of Agriculture and Veterinary Science, Department of Natural Resources Management, P.O. Box: 19, Ambo, Ethiopia Addis Ababa University, College of Social Sciences, Department of Geography and Environmental Studies, P.O. Box: 150249, Addis Ababa, Ethiopia Addis Ababa University, College of Development Studies, Department of food Security and Development, P.O. Box: 150249, Addis Ababa, Ethiopia E–mail address (corresponding author*): mahammede766@gmail.com ORCID iD: Mahammed Endrias: https://orcid.org/0000-0002-0170-4531; Temesgen Argaw Naji: https://orcid.org/0000-0002- 6489-6008 ______________________________________________________________________________________________________________________________________________ A B S T R A C T The study sought to understand the spatio-temporal variation of land use and land cover (LULC) and its drivers under watershed management activities in Becho district, in Ethiopia's Central Highlands. To gain a better comprehensio n of the subject, two micro watersheds were chosen to test the effectiveness of watershed management activities (treated -Shankur Tareqo and untreated-Mende Tufessa). LULC changes were detected using aerial photography (1973), and satellite images from Landsat 5 TM (1990), Spot 5 (2005), and Landsat 8 OLI (2021) obtained from the Ethiopian Geospatial Information Institute and the United States Geological Survey. In addition, key informant interviews (KII) and focus group discussions (FGD) were used to elicit LULC drivers. Between 1973 and 2021, the total area of woodland, and cultivated and rural settlement land LULC classes in the treated micro-watershed decreased by 24.65% (171.9 ha) and 7.34% (759 ha), respectively. Though, grassland, forestland, and barrenland increased by 30.83% (179.52 ha), 1% (3 ha), and 183.14% (755.28 ha), respectively. The overall area of LULC class for woodland, grassland, forestland, and cultivated and rural settlement decreased by 50.36% (316.16 ha), 41.23% (196.46 ha), 2.43% (11.85 ha), and 1.35% (138.6 ha), respectively, in the untreated micro-watershed, while barrenland increased by 175.86% (666.55 ha). According to KII and FGD, the drivers of LULC changes were identified as the expansion of cultivated land, population pressure, and government policy. According to the findings, local-scale watershed management activities was not as effective as expected. As a result, in order to achieve the desired outcome, the concerned stakeholders should reconsider how watershed management activities is undertaken. KEY WORDS: land use, land cover, watershed management activities, spatio-temporal variation, supervised classification, detection ARTICLE HISTORY: received 6 August 2022; received in revised form 19 November 2022; accepted 23 November 2022 ______________________________________________________________________________________________________________________________________________ 1. Introduction has expanded by 12 million km during the same time frame. The majority of the conversion of Land use and land cover dynamics are the main forests and woodlands to other land use categories driving forces underlying local, regional, and global has taken place in South America, Africa, and Asia. environmental changes (ABEBE ET AL., 2022). Due to a number of factors, such as climate change Almost 1.2 million km of forests and woodlands and population growth (GAREDEW ET AL., 2009), and 5.6 million km of grasslands and pastures East Africa has experienced the second-highest have been converted into various types of land rate of transformation in Africa. use over a long time period and on a worldwide According to ASSEN & NIGUSSIE (2009), each scale (GUAN ET AL., 2008). The area of agriculture region's specific physical and socioeconomic conditions determine LULC changes and their and the problem intensified in different parts of effects. For instance, due to the growth of the country, activities of natural resource agricultural land, the Gubeta-Arjo peasant conservation were primarily implemented in associations in central Ethiopia nearly completely selected watershed sites of the Ethiopian lost all of their forest cover between 1973 and 2006 Highlands, where land degradation was found to (GAREDEW ET AL., 2009). On the other hand, croplands be severe (AMEDE ET AL., 2001; BADEGE, 2001). 2 2 increased by a net 51%, from 403 km to 607 km , in Implementation of watershed management activities the Fincha'a watershed between 1957 and 2001, (WMA) brought some positive outcomes in those at the expense of forests and grazing areas (TEFERA & parts of Ethiopia, especially after the adoption of STERK, 2010). Additionally, in the Chirokella micro- integrated WMA in 2010 (GEBREHAWERIA ET AL., watershed in south eastern Ethiopia, the amount 2016; MELAKU ET AL., 2017). WMA induced at a of dense forest cover declined by nearly 80% watershed scale, for instance, caused a reduction between 1966 and 1996, whereas the area of of 25–38% in soil loss between 2011 and 2015 in moderately forested land was totally transformed the Gumara Maksegnit watershed (MELAKU ET AL., into other LULC categories. In the same micro- 2017), 64% of soil loss in the Agula watershed watershed, the number of cultivated and settlement (between 1990 and 2012) (FENTA ET AL., 2017a), lands increased by 62.8%, whereas the number of 47% of soil loss in the Central Highlands of Ethiopia shrubs and degraded land cover increased by (between 2007 and 2009) (ADIMASSU ET AL., 2012) 49.9% and 100%, respectively (ASSEN & NIGUSSIE, and 65–70% (16.5t ha-1 yr-1) of soil loss at country 2009). Land degradation has inevitably resulted level (FENTA ET AL., 2021). Moreover, watershed from this rapid change of grazing and forest land management induced coverage of dense forest into cultivated and degraded land (HURNI, 1993). (between 1965 and 2005) (ALEMAYEHU ET AL., 2009), Land degradation, prevalent in various and plantation and exclosure (between 2004 and agriculture-based developing countries, including 2009) (HAREGEWEYN ET AL., 2012) were significantly Ethiopia, is manifested by soil nutrient depletion increased in the northern part of the country. and loss of land productivity potential (TAMENE & These and other success stories caused the VLEK, 2008). In Ethiopia, where more than 90% of introduction of WMA in different parts of the country the population depends on agriculture for food through a top-down approach (GEBREHAWERIA ET and a livelihood, soil erosion and soil fertility decline AL., 2016). In line with this, Oromia regional state continue to be a significant problem (EBABU ET AL., announced WMA in 2010/2011 in all parts of the 2020). In Ethiopia, a high rate of soil erosion, region including the study area (OROMIA REGIONAL over-cultivation, improper land use, and poor soil NATIONAL STATE BUREAU OF AGRICULTURE AND management practices, among other things, are NATURAL RESOURCES, 2017). The specific place of Kiro considered to be the causes of land degradation in the Shankur Tareqo micro-watershed, was (AMEDE ET AL., 2001; BADEGE, 2001). In the Highlands, selected as a model area to implement different where inappropriate farming practices have been soil and water conservation measures as it used for centuries and where about 87% of the represents the hilly and gully areas of the district, nation's population lives, the problem of soil erosion which have suffered serious soil erosion for a and the associated reduction in soil fertility is worse long period (BECHO WOREDA NATURAL RESOURCES (EBABU ET AL., 2020). For instance, according to DEPARTMENT (BWNRD), 2021). With the start of BAI ET AL. (2008), about 26.33% (296,812 km ) of WMA, in Kiro and the Shankur Tareqo micro- Ethiopia’s land area was affected by land watershed as a whole, a better achievement has degradation, majorly caused by water-induced soil been observed on soil and water conservation erosion. About 23% of its soil erosion was activities even though different obstacles were contributed by cropland, which constituted about found. As a result, the Shankur Tareqo micro- 50% of the country, with a mean soil erosion loss watershed has been considered the subject of -1 -1 of 36.5 t ha yr (FENTA ET AL., 2021). Due to this, different forms of soil and water conservation Ethiopia annually loses about one billion tonnes activities since 2010/2011. of topsoil, costing about 30% of its gross Thus, in addition to the aforementioned factors domestic product (KIRUI & MIRZABAEV, 2014). (population growth and density, over-intensification To reverse the problem of land degradation of land use, farm size, land tenure status, and lack and achieve the sustainable use of its resources, of land use policies), LULC change-induced land optimistic land management practises have been degradation and land degradation-induced WMA implemented in Ethiopia at a watershed scale, are two sides of the same coin that have an focusing on the Northern Highlands since the 1970s impact on the LULC structure in the study area (AMEDE ET AL., 2001; BADEGE, 2001). As time passed and elsewhere. However, it is hard to find an article that was written on LULC change of the treated micro-watershed covers 12,327 ha and is located between and N district unlike that of the northern part of Ethiopia. Even the few existing articles about LULC changes latitude and to E longitude. (SHAWUL & CHAKMA, 2019; TEFFERA ET AL., 2018) Its elevation ranges between 2101 m. a.s.l. in the covering the watersheds were performed at the lower part, and 2795 m a.s.l. in the upper part, with a macroscale. However, the contribution of LULC slope gradient reaching 69%. It encompasses the change studies at such a large scale to land specific place of Kiro where a government-led WMA management options is less (GEBRELIBANOS & ASSEN, was implemented in collaboration with Shankur 2013). But studies at the micro-scale watershed, village residents on previously degraded communal which were uncommon in the study area, are land. The Kiro specific place covers approximately more effective. In order to devise suitable land 32 ha. It was viewed as a model site to disseminate management practises, strategies, and policies, a technologies and experiences in natural resource thorough analysis of LULC change brought on by conservation to the nearby watersheds. In this specific WMA and other driving variables within the place and the Shankur Tareqo watershed, about micro-watershed is thus urgent and essential. 606 m of gully rehabilitation with trees, 5 m of Hence, this study aimed to detect the spatio- gully rehabilitation with stones, 3.72 km of soil temporal variations of LULC and its driving factors bund, 1.5 km of the stone bund, and 20 micro-basins under watershed management in the Becho district were constructed between 2018 and 2019 (BWNRD, of central Ethiopia between 1973 and 2021. 2021). The adjacent Mende Tufessa micro-watershed located between and N 2. Materials and methods latitude, and and E longitude, covering an area of 12,272 ha, was used as 2.1. Study area a control to check the effectiveness of the introduced WMA, due to the absence of government-led WMA The study was conducted in two adjacent micro- implemented in this micro-watershed. Thus, it watersheds, named Shankur Tareqo (treated) and was considered to be untreated. Mende Tufessa (untreated) in the Becho District, of the Central Highlands of Ethiopia (Fig. 1). The Fig. 1. Location map of Shankur Tareqo and Mende Tufessa watersheds, in Becho district The underlying geology in each micro-watershed of and , respectively. The mean consists of Nazareth series volcanoes, forming annual temperature and average annual rainfall parent materials for the vertisols that dominate are 16°C and 1127 mm, respectively (Fig. 2). The the landscape of the watersheds. About 90% of subsistence rain-fed farming system forms the the soils in the watersheds are vertisols, while main economic activity in both micro-watersheds. the remaining are occupied by luvisols, nitisols, Locally, maize (Zea mays), teff (Eragrostis tef), wheat and leptosols (ELIAS, 2016). (Triticum aestivum), and chickpea (Cicer arietinum) Both micro-watersheds are predominately are the major cultivated crops. Furthermore, oxen, grouped into the mid-altitude zone (ELIAS, 2016). cows, goats, sheep, hens, horses, mules, and According to long-term meteorological data (1990– donkeys are the most common animals found in 2019), these micro-watersheds were characterised by the micro-watersheds (BWNRD, 2021). mean maximum and minimum annual temperatures Fig. 2. Rainfall and temperature records of the study area based on the records of nearby Tulu Bolo station (1990–2019) from the national meteorological service agency, Ethiopia 2.2. Data acquisition, sources and processing Scanned aerial photos with a photogrammetric scanner at 1200 dots per inch (DPI) in tagged image The LULC change analysis of the study area file (TIF) format were independently orthorectified was carried out using a collection of aerial photos and registered to a coordinate system of WGS 1984 and various satellite imagery. Accordingly, the UTM Zone 37N using the Transverse Mercator Ethiopian Geospatial Information Institute (EGII) projection system with less than 0.85 root mean provided seven aerial photos for 1973 and a Spot 5 square error (RMSE) by the data providers, that image for 2005, as well as two sets of topographic is EGII. These rectified pieces of aerial photos maps at 1:50000 scale. The remaining Landsat were mosaicked using Erdas Imagine 2014. After Collection 1 level-1 data product images of 1990 calibrating at sensor radiance, radiometric correction and 2021, and digital elevation model were freely of Landsat 5 TM and Landsat 8 OLI was carried downloaded from the USGS website in December, out using the methods described by ZADBAGHER ET 2021 (https://earthexplorer.usgs.gov). Furthermore, AL. (2018) and PRIETO-AMPARÁN ET AL. (2016), the watershed boundaries were delineated using respectively, in Erdas Imagine 2014. All of these digital elevation model. The images were chosen images were reprojected to a WGS 84 into a UTM based on data availability (1973 aerial photos), Zone 37N coordinate system using a 1:50,000 anticipated major changes (1990), introduced topographic map as the base map to avoid any watershed management (2005) and year consistency projection distortion and displacement that might (2021). occur among those images. Finally, the study area was extracted using the this process because they had the same tonal boundary of watersheds on the mosaicked geo- value on the image. Therefore, rural settlements referenced aerial photos of 1973 and the satellite and cultivated land were amalgamated and called images. Investigation of shape, size, pattern, tone, "cultivated and rural settlement land”, while texture, shadow, and site association were used plantation and natural forest, which are commonly to identify LULC classes in mosaicked aerial found along riversides, were simply called photos of 1973. However, in order to improve "forestland". Using the existence of vegetation, delineation objectivity, interpretation repeatability, edaphic conditions and artificiality of cover as and processing efficiency, locally available LULC explained in the Land Cover Classification system types from satellite images were identified through (DI GREGORIO, 2005), five LULC types–barren land supervised classification (maximum likelihood (BL), forestland (FL), cultivated and rural settlement techniques) algorithms with the aid of Erdas Imagine land (CRSL), grassland (GL) and woodland (WL)– 2014 and ArcGIS 10.7. The indistinguishability of were identified in the studied micro-watersheds rural settlements and cultivated land, as well as (Table 1). natural forest and plantation, was recognised during Table 1. LULC types identified in the treated and untreated micro-watersheds LULC types Description BL Refers to areas of exposed stone, sand and soil or area with no vegetation or dominated by rock-out crops, eroded and degraded lands found along the flooding area of the local stream valleys, over gentle and steep mountain slopes CRSL Refers to areas covered with annual rainfed crops and scattered rural dwelling FL Refers to plantations, the disturbed natural patchy forest cover of mature trees and riparian vegetation GL Refers to areas covered with grasses WL Areas covered with sparse woody plants mixed with shrubs, bushes, and grasses. It also incorporates perennial cropland with 30-50% moderately stocked tree crown cover NB: WL=Woodland, GL=Grassland, FL=Forestland, CRSL=Cultivated and Rural Settlement land and BL=Barren land Furthermore, information that is not possible of the temporal changes of the LULC types and to extract from aerial photos or satellite images computed the extent of LULC conversion induced was obtained through open-ended guided questions by WMA and physical, socio-economic and policy used for focus group discussion (FGD) and key related factors. Thus, the rate of absolute variation of informant interview (KII) as well as from personal a single LULC category between two time points field observations. A total of three FGDs were was quantitatively determined using the held in both watersheds, comprising of 5 to 8 equations 1 and 2. Additionally, to have a participants selected from local elders, community complete picture of the conversion between leaders, and development agents. KII were also various LULC types, transition matrices were conducted with 12 (6 from each watershed) made for the different periods of analysis. purposefully selected local residents aged between 30 and 65 to gain further insight into issues 1 related to LULC changes perceived throughout their lives. 2 2.3. LULC spatio-temporal change detection methods where S is the absolute variation in the area of a certain land type in the study period; K is the Post classification change detection techniques, change rate of a land type with a positive value which reduced the possible effects of spectral representing an increasing rate and a negative resolution and sensor differences between the value representing a decreasing rate; Ua and U is multi-temporal image (ABERA ET AL., 2018) were the area of a land type at the beginning and the used to analyse the change in LULC for the end of the study period in hectares. periods 1973–1990, 1990–2005, 2005–2021, and 1973–2021. This method enabled an assessment 3. Results and discussion grassland, forestland, and barren land were the common LULC types identified in the watersheds. The LULC types and their spatial distribution Furthermore, the LULC transition matrix determined for Shankur Tareqo and Mende Tufessa micro- for the years between 1973 and 2021 are presented watersheds are shown in Figs. 3 and 4 as well as in Table 3. Table 2. Cultivated and rural settlement, woodland, Fig. 3. LULC distribution of Shankur Tareqo and Mende Tufessa micro-watershed for each year from 1973 to 2021 3.1. Cultivated and rural settlement land (CRSL) migration in the Gedalas watershed of the north- LULC pattern from 1973–2021 eastern Highlands of Ethiopia, which resulted in a decline in CRSL. MEKUYIE ET AL. (2018) also noted CRSL comprises the largest share of the a decrease in farmland between 1985 and 1995 watersheds throughout the study period (1973– as a result of the movement of pastoralists in 2021). The share of this LULC category in the response to climate change. In the later periods of treated micro-watershed accounted for 84%, 69%, analysis (1990–2005 and 2005–2021), CRSL LULC 72%, and 78% for the years 1973, 1990, 2005, classes showed an increasing trend. The highest -1 and 2021, respectively (Fig. 4). Except for the increment of 7.89%, with a rate of 43.8 ha yr , years 1973–1990, CRSL LULC types increased was observed in the years between 2005 and continuously (Table 2). In the first period of analysis 2021 (Table 2). This increment is the result of (1973–1990), a considerable reduction, by 17.48%, 753.5 ha of WL, 565.8 ha of BL, 469.9 ha of GL, of this LULC category was recorded due to rural- and 206.5 ha of FL conversion to CRSL between urban migration and Beherawi Witidrina , as per 2005 and 2021 (Table 3). In general, the CRSL KII and FGD. This was in line with the study of LULC classes decreased by ~7% over the study YESUPH & DAGNEW (2019), who found that between period, between 1973 and 2021, in the treated 1973 and 1986, there was a major rural out- micro-watershed. In the untreated micro-watershed, after a 1 decline from 83.93% in 1973 to 68.34% in 1990, It was the Derg regime's forced conscription of young male working-age groups into its wars with the Tigray People's CRSL showed a significant expansion from 69.35% Liberation Front, the Eritrean People's Liberation Front, and in 2005 to 82.78% in 2021 (Fig. 4). Between 2005 the Ogaden War. and 2021, the highest increment of 19.38%, with -1 a rate of 103.07 ha yr , was observed (Table 2). by ~7% throughout their study periods between CRSL expanded at the expense of 1023.1 ha of 2000 and 2017 in north west Ethiopia (WORKU ET WL, 631.7 ha of GL, 501.7 ha of BL, and 420.1 ha AL., 2021), ~1% between 1985 and 2017 in south of FL (Table 3). The primary reason for this west Ethiopia (FASIKA ET AL., 2019), ~14% increment, as mentioned by the stakeholders, was between 1985 and 2010 in the Abaya-Chamo population pressure. During the study period, basin (WOLDEYOHANNES ET AL., 2018), and ~24% CRSL in this micro-watershed decreased by 1.35%. between 1985–2015 in the Afar region (MEKUYIE The result of this study was in agreement with ET AL., 2018). the findings of those who found a decline of CRSL Fig. 4. Aerial coverage of LULC change in the Shankur Tareqo (A) and Mende Tufessa (B) micro-watersheds in 1973, 1990, 2005 and 2021 3.2. Grassland (GL) LULC pattern from 1973–2021 hence causing an increase in GL and WL LULC types during this time period. In contrast to this In 1973, GL covered approximately 5% of the period of analysis (1973–1990), aerial coverage treated micro-watershed, which increased to 13.14% of the GL LULC class decreased by 38.03% and in 1990. After that, GL showed continuous decline. 24.05% in the second (1990–2005) and third It declined from 8.14% in 2005 to 6.18% in 2021 (2005–2021) periods of analysis, respectively. (Fig. 4). Between 1973 and 1990, the GL LULC class The highest decline was recorded in the second -1 increased by 177.97% at a rate of 60.96 ha yr period (1990–2005) of analysis, which was (Table 2). This is the opposite side of a decline in attributed to a conversion of 758.3 ha, 298.7 ha, CRSL in which ~1293 ha of CRSL was converted 115.6 ha, and 72.9 ha of GL to CRSL, WL, BL, and to GL (Table 3). Thus, the source of shrinkage in FL, respectively (Table 3). Throughout the study CRSL was probably due to its part being taken period (1973–2021), a total of 312.6 ha (53.7%) over by GL that, over time, transformed into WL, of GL was converted to CRSL LULC types. Similarly, in an untreated micro-watershed, from population pressure (Table 2). The results the GL LULC class accounts for nearly 3.88% of of GAREDEW ET AL. (2009) and YESUPH & DAGNEW the total area in 1973 and 9.67% in 1990 (Fig. 4). (2019), found an increase in GL from 1973 to This showed an increment of GL which was 1986 and subsequent continuous reductions were attributed to a decline in CRSL following rural– perfectly in agreement with this study. Similar urban migration and Beherawi Witidrina as findings were also reported by DEGIFE ET AL., mentioned by the elders. However, in the later (2019) for the period of 1972–1992 in the Lake years, it gradually declined from 7.74% in 2005 Hawassa watershed. This LULC class decreased to 2.28% in 2021. The primary contraction took by 41% throughout the study period. This was in place from 2005 to 2021 by 70.52%, with a rate of line with the results of ASMAMAW ET AL. (2011) -1 41.87 ha yr (Table 2). Much of this LULC class and GEBRELIBANOS & ASSEN (2013) in the Gerado was changed into CRSL (631.7 ha, ~67%) due to and Hirmi Watersheds of the north-eastern and the high demand for agricultural land resulting northern highlands of Ethiopia, respectively. Table 2. Absolute variation in the area (S) and change rate (K) of LULC in the treated and untreated micro-watershed in different periods; increase (+) and decrease (-) Shankur Tareqo (Treated) micro-watershed LULC 1973-2021 1973-1990 1990-2005 2005-2021 Classes K (%) S (ha) K (%) S (ha) K (%) S (ha) K (%) S (ha) WL -24.65 240.36 34.47 311.9 33.26 -724.12 -57.95 -171.86 GL 30.83 1036.39 177.97 -615.64 -38.03 -241.23 -24.05 179.52 FL 1.09 391.59 141.25 -343.43 -51.35 -45.13 -13.87 3.03 CRSL -7.34 -1808.54 -17.48 348.72 4.08 700.74 7.89 -759.08 BL 183.14 147.85 35.85 299.65 53.49 307.78 35.79 755.28 Mende Tufessa (Untreated) micro-watershed LULC 1973-1990 1990-2005 2005-2021 1973-2021 Classes K (%) S (ha) K (%) S (ha) K (%) S (ha) K (%) S (ha) WL -50.36 438.51 69.85 380.72 35.71 -1135.39 -78.46 -316.16 GL -41.23 709.94 148.99 -236.42 -19.93 -669.98 -70.52 -196.46 FL -2.43 475.61 97.55 -321.17 -33.35 -166.29 -25.9 -11.85 CRSL -1.35 -1911.75 -18.57 124.05 1.48 1649.1 19.38 -138.6 BL 175.86 289.07 76.27 53.91 8.07 323.58 44.82 666.55 NB: WL=Woodland, GL=Grassland, FL=Forestland, CRSL=Cultivated and Rural Settlement land and BL=Barren land 3.3. Woodland (WL) LULC pattern from 1973–2021 throughout the study period (1973–2021), this LULC class exhibited a net loss of 24.65% (171.86 ha) The area of the WL class in the treated micro- in which the majority was transformed into CRSL watershed was comprised of 5.66%, 7.61%, 10.14%, (45.7%, 318.2 ha) (Table 3). and 4.26% in the years 1973, 1990, 2005, and In the untreated micro-watershed, GL LULC 2021, respectively (Fig. 4). This showed a gain of WL accounted for 5.12%, 8.69%, 11.79%, and 2.54% in the first (1973–1990) and second (1990–2005) of the total area in 1973, 1990, 2005, and 2021, periods of analysis, but a decline in the third period respectively (Fig. 4). The WL LULC class in this of analysis (2005–2021) (Table 2). The highest watershed increased by 69.85% and 35.71% in (34.47%) rise of the WL LULC class was detected the periods of analysis from 1973 to 1990 and in the first period of analysis (1973–1990) with a from 1990 to 2005, respectively. According to KII -1 rate of 14.14 ha yr due to the conversion of and FGD participants, the socialist regime's policy CRSL (806 ha, 7.8%) to WL. Thus, the increase in of rehabilitating degraded land in the 1980s WL area is presumably attributed to gains from CRSL through exclosure of hillsides and tree-planting due to the reasons mentioned below. In contrast, efforts led to an increase in the WL class in both from 2005 to 2021, WL area coverage decreased micro-watersheds throughout the first and -1 by 57.95% at a rate of 45.26 ha yr . However, second periods of analysis. In contrast, it declined by 78.46% in the third period (2005–2021) of Ethiopia was overthrown in 1991, and the expansion analysis (Table 2). In the overall study period of CRSL, as mentioned by KII and FGD, may all be (1973–2021), WL LULC dwindled by 50.36% in responsible for the overall decline in WL in both the untreated micro-watershed. This loss was micro-watersheds. This is consistent with research largely associated with its conversion to a CRSL done in other regions of Ethiopia (YESUPH & (467 ha, 74.5%) (Table 3). DAGNEW, 2019). However, the overall decline in The need for more firewood due to population the treated micro-watershed was significantly growth, the redistribution of land to ex-soldiers lower, which may have been caused by the who have returned since the military regime in government's adoption of WMA. Table 3. LULC conversion matrices in different periods in each watershed (area in ha) Shankur Tareqo (Treated) Mende Tufessa (Untreated) Changed Changed 1973- 1990- 2005- 1973- 1973- 1990- 2005- 1973- from to 1990 2005 2021 2021 1990 2005 2021 2021 ha ha ha ha ha ha ha ha WL 51.9 125.7 147.7 284.1 82.7 184.9 90.3 57 GL 114.5 29.1 91.8 29.6 72.9 22.3 33.5 11.5 FL 79.2 76.2 125.2 19.8 69.1 240.9 170.7 47.1 Woodland (WL) CRSL 427.2 681.9 753.5 318.2 375.4 589.9 1023.1 467 BL 23.6 23.1 127.5 44.5 27.1 27.4 127.3 44.5 Total 696.4 936 1245.7 696.2 627.1 1065.4 1444.9 627.1 WL 26.2 298.7 67.5 16.5 36.3 229 35 12.2 GL 114 370.3 313.1 205.9 111.4 275.1 128.5 76 FL 19.6 72.9 19.7 8 49.9 86.3 19.6 34.7 Grassland (GL) CRSL 380.7 758.3 469.9 312.6 254.7 542.5 631.7 309.6 BL 41.4 115.6 132 39 23.6 52.1 134.7 43.5 Total 581.9 1616 1002.1 582 475.9 1185 949.5 476 WL 23.7 149.5 58.3 10 78.1 182.2 34.5 17.2 GL 35.9 9.8 17.9 11.8 48.3 9.3 7.9 6.6 FL 23.9 81.6 17.6 29.9 89.4 232 136.1 52.1 Forestland (FL) CRSL 185.1 413.3 206.5 198.4 255.4 525.9 420.1 378.7 BL 8.2 13.1 23.8 26.7 16 12.4 43.4 32.6 Total 276.8 667.3 324.1 276.8 487.1 961.7 642 487.2 WL 806 622.7 196.6 202.5 841.7 786.4 126.9 214.9 GL 563.6 94.7 180 1293 516.8 252 496.6 920.1 FL 524.7 81 98.3 216.3 725.7 78.5 123.5 331.3 CRSL CRSL 7240 6680 7576 8523.7 7216.9 6423.5 7569.2 8729.2 BL 464.5 620.9 749 892.3 575.9 522.2 583.8 828.4 Total 10329 8521 8872 10331.4 10280.3 8374.1 8498.2 10283.9 WL 27.7 49.1 54.9 11.6 26.1 62.5 24.4 9.6 GL 58.1 76.1 86.4 17.4 31.5 79.1 15.3 5.5 FL 19.1 12.7 19 5.1 27 4.4 25.2 9.4 Barren land (BL) CRSL 285.7 335.7 565.8 215.7 269.7 414.5 501.7 259.1 BL 20.7 84.7 131.2 161.8 23.9 106 153.5 94.9 Total 411.4 558.3 857.3 411.7 378.2 666.5 720.1 378.5 NB: CRSL=Cultivated and Rural Settlement land 3.4. Forestland (FL) LULC pattern from 1973–2021 revealed a declining rate of increase in the treated micro-watershed after the implementation of WMA, In 1973 and 1990, around 2.25% and 5.43% which was also affirmed through FGD. It increased of treated micro-watershed were classified as FL, by 183.14% over the entire study period (1973– respectively (Fig. 4). This indicates a 141.25% 2021), much of which came from CRSL classes increase in FL cover class over 17 years, with a rate (892.3 ha, 8.6%) (Table 3). -1 of 23.03 ha yr in the first period of analysis The spatial coverage of the BL class in the (1973–1990), compared to a decrease in the second untreated micro-watershed comprised of 3.09%, (1990–2005) and third (2005–2021) periods of 5.45%, 5.88%, and 8.52% in 1973, 1990, 2005, analysis (Table 2). The increment in the first period and 2021 respectively (Fig. 4). The rate of increment of analysis was ascribed to the replantation of the BL LULC class in the first (1973–1990) and programme announced and implemented by the third (2005–2021) periods of analysis was higher socialist regime that preserved the remnant for the untreated compared to the treated micro- indigenous trees and expanded forest cover. watershed (Table 2). However, in the overall Moreover, politically affiliated rural-urban migration, period (1973–2021) of analysis, the rate of BL which abandoned CRSL and served to regenerate increment was higher for the treated micro- remnant trees, was also mentioned as a cause by watershed. As shown in (Table 3), the treated the elders. This was similar to the result of a micro-watershed still had the highest conversion reported increment of forestland cover between rate of CRSL to BL (8.6%), compared to 8.06% in 1970 and 1980 in the Rib watershed (MOGES & BHAT, the untreated one. This confirms that, in the 2017). Conversely, in the second (1990–2005) and treated micro-watersheds, further efforts will be third (2005–2021) periods of analysis, the FL class needed to rehabilitate the degraded land through exhibited a reduction of 51.35% and 13.85%, soil and water conservation activities. According respectively, as a consequence of its conversion to the discussion with the KII, this was one of the to CRSL (61.9%, 413.3 ha and 63.7%, 206.5 ha) factors that led to the selection of the treated (Table 3). The increased demand for cropland and micro- watershed for rehabilitation activities. the use of trees for construction purposes were the apparent causes of reduction of the FL cover. 3.6. Major drivers of LULC change In the untreated micro-watershed, FL increased from 3.97% in 1973 to 7.85% by 1990. In the 3.6.1. Expansion of cultivated and rural latter period, it gradually declined from 5.23% in settlement land 2005 to 3.88% by 2021 (Fig. 4). In the overall study period (1973–2021) of analysis, FL class in Except for the first period of analysis (1973– the treated micro-watershed showed an increment 1990), the CRSL class was shown to have -1 of 1.09% with a rate of 0.06 ha yr while it significant expansion at the expense of WL, GL, decreased by 2.43% in the untreated one. This and FL in both watersheds. In the first period of shows an improvement of FL LULC class in the analysis (1973–1990), CRSL classes surprisingly treated micro-watershed, owing primarily to declined in both micro-watersheds from where replantation activities in the Kiro specific location they were in 1973. This finding sharply contrasts and the expansion of eucalyptus plantation, as with that of REID ET AL. (2000), who reported a revealed by KII and FGD participants. Thus, this significant decrease in cultivated land in Gullele would be a small benefit detected in the treated (27%), in Gerangera (50%) and in Kumbi (67%) micro-watershed due to the introduction of WMA. between 1973 and 1987 in the Gibe valley (south Similar findings were also observed in north- western Ethiopia). BEWKET & STERK (2002) also eastern Ethiopia (ASMAMAW ET AL., 2011) and reported a shrinking of farmland by 2% between northern Ethiopia (GEBRELIBANOS & ASSEN, 2013). 1982 and 1998 in the Chemoga Watershed, Ethiopia. According to KII and FGD participants, the socialist 3.5. Barren land (BL) LULC pattern from 1973–2021 regime's shift of land ownership from landlord to peasant association (which prevented some farmers In the treated micro-watershed, BL accounts from cultivating land previously owned by the for 3.35%, 4.55%, 6.98%, and 9.48% of the total landlord due to widespread fear), Beherawi area in 1973, 1990, 2005, and 2021, respectively Witidrina, and rural-urban migration were (Fig. 4). The percentage of each increment was mentioned as reasons for change. 35.85%, 53.49%, and 35.79% in the first (1973– In the second (1990–2005) and third (2005– 1990), second (1990–2005), and third (2005–2021) 2021) periods of analysis, the CRSL LULC class periods of analysis, respectively (Table 2). This increased at the expense of other LULC classes due to a high demand for cereal crops resulting of livelihood security had forced farmers to use from population pressure within both micro- the woodlands to cope with recurrent household watersheds (Table 3). This increment in CRSL shocks in the Central highlands of Ethiopia, which was similar to the findings of SHIFERAW ET AL. further modified the LULC of the area. Therefore, (2001) and TEFERA (2011), who reported an it can be concluded that population growth has increment of cropland in Borena Woreda (South driven the expansion of CRSL in the studied Wollo) and Nono Woreda (Central Ethiopia) due watersheds at the expense of other LULC classes, to rapid population growth. However, the lower which have ended up being degraded, as observed in rate of increment in CRSL class was recorded for other parts of the country. the treated micro-watershed for the third period (2005–2021) of analysis (Table 2). This was due 3.6.3. Policy changes related factor to the introduction of WMA. Besides, as per the information gained from KII and FGD participants, A change in policy is a perceptible departure the low-rate of increment of CRSL in the treated from the intended path of action that had been micro-watershed was due to a lack of additional put in place to solve a perceived societal issue farmland, rural-urban migration of the young (MICHAELS ET AL., 2006). Policy change follows a working age group, and resettlement programmes predetermined process that involves problem undertaken by Oromia regional state in 2005. identification, adoption of a specific policy, Therefore, CRSL change caused LULC change in implementation of the policy, and evaluation of the studied micro-watersheds. the policy (MICHAELS ET AL., 2006). The last step is crucial in identifying the effect of the policy on 3.6.2. Demographic related factors the environment. In line with this, long-term residents of the area were requested to explain Demographic factors are one of the underlying the policy-related factors responsible for the causes that brought LULC change in different parts change in LULC. Accordingly, the informants cited of the country (ABERA ET AL., 2018; GEBRELIBANOS the policy changes of 1974 and 1991 as the main & ASSEN, 2013), which is implicated in the study drivers of LULC transformation in the study area. area also. According to the CSA's population census In accordance with the well-known motto "Land and projection (CENTRAL STATISTICAL AGENCY, 2007, to the Tiller," which was elicited by KII and FGD 2013, 2021), the rural population of the district participants, the first policy change nationalized increased rapidly from a total of 44,382 in 1994 the land and the natural resources on it, resulting to 82,737 in 2021. This showed an increase in the in FL and WL being owned by the government. rural population of more than 86.6% between The second policy change resulted in the extensive 1994 and 2021. To some extent, an increase in destruction of vegetated areas as portions of the human population in the Highlands has also formerly government-owned tracts of land were caused an increase in the population of livestock given to private investors and distributed among (DUBALE, 2001). Such rapid population growth, smallholder farmers. In order to bring about combined with an increase in the livestock LULC change in the watersheds under research, population in the area, has put pressure on the the second policy modification has the biggest area's limited land resources by increasing influence. demand for food, fuel, construction materials, According to KII and FGD, much of the expansion grazing land, and permanent settlement land. of CRSL, BL, and loss of FL LULC classes were These issues were also confirmed through KII found after the downfall of the socialist regime. and FGD participants as well as the quantitative Before the downfall, the number of people residing data presented in Fig. 3, suggesting a positive in the watershed was low due to rural–urban relationship between population growth and migration and Beherawi Witidrina, which created expansion of the cultivation land at the expense surplus land. Moreover, they also stated that the of other land use types. This was in line with a socialist government planted trees in a finding by ZELEKE & HURNI (2001), who reported a mountainous area that extended from the hilly decline in forest cover in central Gojjam from areas of Shankur and Mende kebeles to Dilella 27% to 0.3% between 1957 and 1995 due to towns on the land taken from the previous population pressure. Forest cover declined from landlord. This was partly responsible for the 34% to 3% in the country between 1950 and slight increase in GL, WL, and FL in the first 1980 due to population pressure-induced factors period of analysis (1973–1990). (ASSEN & NIGUSSIE, 2009). Moreover, GAREDEW ET AL. (2009) reported that population-induced lack It is the lowest administrative unit in Ethiopia The Ethiopian People’s Revolutionary Democratic Acknowledgements Front government that succeeded the socialist The authors wish to acknowledge Addis Ababa University for regime in 1991 subsequently advocated state its financial support and the USGS for providing free Landsat ownership of land whereby only usufruct rights data. In addition, the authors would like to acknowledge the local are bestowed upon landholders. According to communities, Becho district administrators, and development FGD participants, this policy shift encourages the agents for their valuable support during the data collection process by providing and collecting relevant information. private sector and individual farmers to engage in free and uncontrolled exploitation of FL in order References to have more cultivated land, particularly by clearing previously planted trees. It can, therefore, Abebe M.T., Degefu M.A., Assen M., Legass A. 2022. Dynamics be concluded that government policy has played of land use/land cover: implications on environmental a significant role in LULC change in the study area. resources and human livelihoods in the Middle Awash The results from ABERA ET AL. (2018), ASMAMAW Valley of Ethiopia. Environmental Monitoring and Assessment, ET AL. (2011), and GEBRELIBANOS & ASSEN (2013) in 194, 11: 833. Abera W., Assen M., Satyal P. 2018. Spatio-temporal land the Wanka, Gerado, and Hirmi Watersheds of the use/cover dynamics and its implication for sustainable northwest, north-eastern, and northern highlands of land use in Wanka watershed, northwestern highlands of Ethiopia, respectively, were similar to these. Ethiopia. Modeling Earth Systems and Environment, 5, 2: 571–581. Adimassu Z., Mekonnen K., Yirga C., Kessler A. 2012. Effect of 4. Conclusions Soil Bunds on Runoff, Soil and Nutrient Losses, and Crop Yield in the Central Highlands of Ethiopia. Land Based on mosaicked aerial photos and Landsat Degradation & Development, 25, 6: 554–564. images, LULC changes were calculated for the Alemayehu F., Taha N., Nyssen J., Girma A., Zenebe A., treated and untreated micro-watersheds of the Behailu M., Deckers S., Poesen J. 2009. The impacts of watershed management on land use and land cover central highlands of Ethiopia using in-depth dynamics in Eastern Tigray (Ethiopia). Resources, analysis. The result indicated that CRSL in both Conservation and Recycling, 53, 4: 192–198. watersheds declined between 1973 and 1990, Amede T., Geta E., Takele B. 2001. Reversing the degradation and they experienced upsurges later. From 1973 of arable land in Ethiopian Highlands, Managing African Soils, Vol. 23. to 1990, WL, GL, and FL grew in both watersheds, Asmamaw L.B., Mohammed A.A., Lulseged T.D. 2011. Land whereas CRSL decreased. Due to the growth of BL use/cover dynamics and their effects in the Gerado in the chosen micro-watersheds, FL, WL, and GL catchment, northeastern Ethiopia. International Journal decreased in the following years. These LULC of Environmental Studies, 68, 6: 883–900. types declined at a slower rate in the treated Assen M., Nigussie T. 2009. Land use/cover changes between 1966 and 1996 in Chirokella micro-watershed, Southeastern micro-watershed. This lower rate of decline was Ethiopia. East African Journal of Sciences, 3, 1: 1–8. attributed to the WMA performed there. However, Badege B. 2001. Deforestation and Land Degradation in the significant variation between the two micro- Ethiopian Highlands: A Strategy for Physical Recovery. watersheds was not observed. Thus, this implicates Northeast African Studies, 8, 1: 7–26. Bai Z., Dent D.L., Olsson L., Schaepman M.E. 2008. Global the ineffectiveness of the WMA. Because of this, assessment of land degradation and improvement: 1. as the land loses its protective natural cover as a identification by remote sensing: ISRIC-World Soil Information. result of the growth of agricultural land, population Bewket W., Sterk G. 2002. Farmers' participation in soil and pressure, and other government policy-induced water conservation activities in the Chemoga Watershed, causes, the examined watersheds become more Blue Nile basin, Ethiopia. Land Degradation & Development, 13, 3: 189–200. and more vulnerable to land degradation. Thus, BWNRD. 2021. Socio-economic and Natural Resource Report the patterns of LULC changes in the examined Becho Woreda of 2020. Becho Woreda Natural Resource micro-watersheds have nevertheless been greatly Department, Becho. influenced by an interaction of physical, socio- Central Statistical Agency. 2007. Population and Housing Census 2007: Oromia statistical, 1084. Central Statistical economic, and policy issues. In order to identify Agency, Addis Ababa. observed impediments, previously deployed WMA Central Statistical Agency. 2013. Population Projection of should be reviewed. Furthermore, the chosen Ethiopia for All Regions At Wereda Level from 2014– micro-watersheds should adopt a suitable 2017, 118. Central Statistical Agency, Addis Ababa. additional watershed management plan based on Central Statistical Agency. 2021. Population Projection Wereda as of July 2021: 39. Central Statistical Agency, the knowledge gained from the prior WMA. Addis Ababa. Reforestation of the watershed's degraded areas, Degife A., Worku H., Gizaw S., Legesse A. 2019. Land use land protection and enhancement of the current FL cover dynamics, its drivers and environmental implications cover, and other successful soil and water in Lake Hawassa Watershed of Ethiopia. Remote Sensing Applications: Society and Environment, 14: 178–190. conservation techniques should be included in this effort to prevent additional land degradation. Di Gregorio A. 2005. Land cover classification system: in the evolution of watershed management. Environmental classification concepts and user manual: LCCS: Food & Management, 38, 6: 983–992. Agriculture Organization. Moges D.M., Bhat H.G. 2017. Integration of geospatial Dubale P. 2001. Soil and Water Resources and Degradation technologies with RUSLE for analysis of land use/cover Factors Affecting their Productivity in the Ethiopian change impact on soil erosion: case study in Rib Highland Agro-ecosystems. International Conference on watershed, north-western highland Ethiopia. Environmental African Development Archives. Earth Sciences, 76, 22: 1–14. Ebabu K., Tsunekawa A., Haregeweyn N., Adgo E., Meshesha Oromia Regional National State Bureau of Agriculture and D.T., Aklog D., Masunaga T., Tsubo M., Sultan D., Fenta Natural Resources. 2017. Effort of Integrated Watershed A.A., Yibeltal M. 2020. Exploring the variability of soil Management Towards Reversing Land Degradation. Oromia properties as influenced by land use and management Regional National State Bureau of Agriculture and practices: A case study in the Upper Blue Nile basin, Natural Resources. Ethiopia. Soil and Tillage Research, 200: 104614. Prieto-Amparán J., Pinedo-Alvarez A., Villarreal-Guerrero F., Elias E. 2016. Soils of the Ethiopian Highlands: Geomorphology and Pinedo-Alvarez C., Morales-Nieto C., Manjarrez-Domínguez C. Properties The Netherlands.: CASCAPE Project, ALTERA, 2016. Past and Future Spatial Growth Dynamics of Wageningen University and Research Centre (Wageningen Chihuahua City, Mexico: Pressures for Land Use. ISPRS UR). International Journal of Geo-Information, 5, 12: 1–19. Fasika A., Motuma T., Gizaw T. 2019. Land Use Land Cover Reid R.S., Kruska R.L., Muthui N., Taye A., Wotton S., Wilson Change Trend and Its Drivers in Somodo Watershed C.J., Mulatu W. 2000. Land-use and land-cover dynamics South Western, Ethiopia. African Journal of Agricultural in response to changes in climatic, biological and socio- Research, 14, 2: 102–117. political forces: the case of southwestern Ethiopia. Fenta A.A., Tsunekawa A., Haregeweyn N., Tsubo M., Yasuda Landscape Ecology, 15, 4: 339–355. H., Kawai T., Ebabu K., Berihun M.L., Belay A.S., Sultan D. Shawul A.A., Chakma S. 2019. Spatiotemporal detection of 2021. Agroecology-based soil erosion assessment for land use/land cover change in the large basin using better conservation planning in Ethiopian river basins. integrated approaches of remote sensing and GIS in the Environmental Research, 195: 110786. Upper Awash basin, Ethiopia. Environmental Earth Sciences, Fenta A.A., Yasuda H., Haregeweyn N., Belay A.S., Hadush Z., 78, 5: 1–13. Gebremedhin M.A., Mekonnen G. 2017a. The dynamics of Shiferaw B., Holden S., Aune J. 2001. Population Pressure urban expansion and land use/land cover changes using and Land Degradation in the Ethiopian Highlands: A Bio- remote sensing and spatial metrics: the case of Mekelle Economic Model with Endogenous Soil Degradation. In: City of northern Ethiopia. International Journal of Remote Heerink, N., van Keulen, H., Kuiper, M. (eds) Economic Policy Sensing, 38, 14: 4107–4129. and Sustainable Land Use: Recent Advances in Quantitative Garedew E., Sandewall M., Soderberg U., Campbell B.M. 2009. Analysis for Developing Countries. Heidelberg: 73–92. Land-use and land-cover dynamics in the central rift valley of Tamene L., Vlek P.L.G. 2008. Soil Erosion Studies in Northern Ethiopia. Environmental Management, 44, 4: 683–694. Ethiopia. In: A.K. Braimoh, P.L.G. Vlek (Eds.) Land Use and Gebrehaweria G., Dereje A.A., Girmay G., Giordano M., Langan S. Soil Resources. Springer Netherlands, Dordrecht: 73-100. 2016. An assessment of integrated watershed management Tefera B., Sterk G. 2010. Land management, erosion problems in Ethiopia. IWMI Working Paper (170). and soil and water conservation in Fincha’a watershed, Gebrelibanos T., Assen M. 2013. Land use/land cover western Ethiopia. Land Use Policy, 27, 4: 1027–1037. dynamics and their driving forces in the Hirmi watershed Tefera M.M. 2011. Land-use/land-cover dynamics in Nonno and its adjacent agro-ecosystem, highlands of Northern District, Central Ethiopia. Journal of Sustainable Development Ethiopia. Journal of Land Use Science, 10, 1: 81–94. in Africa 13, 1: 123–140. Guan D., Gao W., Watari K., Fukahori H. 2008. Land use change Teffera Z.L., Li J., Debsu T.M., Menegesha B.Y. 2018. Assessing of Kitakyushu based on landscape ecology and Markov land use and land cover dynamics using composites of model. Journal of Geographical Sciences, 18, 4: 455–468. spectral indices and principal component analysis: A Haregeweyn N., Fikadu G., Tsunekawa A., Tsubo M., case study in middle Awash subbasin, Ethiopia. Applied Meshesha D.T. 2012. The dynamics of urban expansion Geography, 96: 109–129. and its impacts on land use/land cover change and Woldeyohannes A., Cotter M., Kelboro G., Dessalegn W. 2018. small-scale farmers living near the urban fringe: A case Land Use and Land Cover Changes and Their Effects on study of Bahir Dar, Ethiopia. Landscape and Urban the Landscape of Abaya-Chamo Basin, Southern Ethiopia. Planning, 106, 2: 149–157. Land, 7, 1: 1–17. Hurni H. 1993. Land degredation, famine, and land resource Worku T., Mekonnen M., Yitaferu B., Cerdà A. 2021. scenarios in Ethiopia. Conversion of crop land use to plantation land use, Kirui O.K., Mirzabaev A. 2014. Economics of land degradation northwest Ethiopia. Trees, Forests and People, 3: 100044. in Eastern Africa. ZEF Working Paper Series. Yesuph A.Y., Dagnew A.B. 2019. Land use/cover Mekuyie M., Jordaan A., Melka Y. 2018. Land-use and land- spatiotemporal dynamics, driving forces and cover changes and their drivers in rangeland-dependent implications at the Beshillo catchment of the Blue Nile pastoral communities in the southern Afar Region of Ethiopia. Basin, North Eastern Highlands of Ethiopia. Environmental African Journal of Range & Forage Science, 35, 1: 33–43. Systems Research, 8, 1: 1–30. Melaku N.D., Renschler C.S., Holzmann H., Strohmeier S., Zadbagher E., Becek K., Berberoglu S. 2018. Modelling land Bayu W., Zucca C., Ziadat F., Klik A. 2017. Prediction of use/land cover change using remote sensing and soil and water conservation structure impacts on runoff geographic information systems: a case study of the and erosion processes using SWAT model in the northern Seyhan Basin, Turkey. Journal of Environmental Monitoring Ethiopian highlands. Journal of Soils and Sediments, 18, 4: and Assessment, 190, 8: 1–15. 1743–1755. Zeleke G., Hurni H. 2001. Implications of Land Use and Land Michaels S., Goucher N.P., McCarthy D. 2006. Policy windows, Cover Dynamics for Mountain Resource Degradation in policy change, and organizational learning: watersheds the Northwestern Ethiopian Highlands. Mountain Research and Development, 21, 2: 184–191. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Environmental & Socio-economic Studies de Gruyter

Spatio-temporal variation on land use and land cover and its drivers under watershed management activities in Becho, Central Highlands of Ethiopia

Download PDF
13 pages

Loading next page...
 
/lp/de-gruyter/spatio-temporal-variation-on-land-use-and-land-cover-and-its-drivers-gBdX6qjn5S

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
de Gruyter
Copyright
© 2022 Mahammed Endrias Hailu et al., published by Sciendo
ISSN
2354-0079
eISSN
2354-0079
DOI
10.2478/environ-2022-0021
Publisher site
See Article on Publisher Site

Abstract

Environmental & Socio-economic Studies DOI: 10.2478/environ-2022-0021 Environ. Socio.-econ. Stud., 2022, 10, 4: 22-34 ________________________________________________________________________________________________ Original article Spatio-temporal variation on land use and land cover and its drivers under watershed management activities in Becho, Central Highlands of Ethiopia 1, 2 2 3 Mahammed Endrias Hailu *, Mohammed Assen Ahmed , Temesgen Argaw Naji Ambo University, College of Agriculture and Veterinary Science, Department of Natural Resources Management, P.O. Box: 19, Ambo, Ethiopia Addis Ababa University, College of Social Sciences, Department of Geography and Environmental Studies, P.O. Box: 150249, Addis Ababa, Ethiopia Addis Ababa University, College of Development Studies, Department of food Security and Development, P.O. Box: 150249, Addis Ababa, Ethiopia E–mail address (corresponding author*): mahammede766@gmail.com ORCID iD: Mahammed Endrias: https://orcid.org/0000-0002-0170-4531; Temesgen Argaw Naji: https://orcid.org/0000-0002- 6489-6008 ______________________________________________________________________________________________________________________________________________ A B S T R A C T The study sought to understand the spatio-temporal variation of land use and land cover (LULC) and its drivers under watershed management activities in Becho district, in Ethiopia's Central Highlands. To gain a better comprehensio n of the subject, two micro watersheds were chosen to test the effectiveness of watershed management activities (treated -Shankur Tareqo and untreated-Mende Tufessa). LULC changes were detected using aerial photography (1973), and satellite images from Landsat 5 TM (1990), Spot 5 (2005), and Landsat 8 OLI (2021) obtained from the Ethiopian Geospatial Information Institute and the United States Geological Survey. In addition, key informant interviews (KII) and focus group discussions (FGD) were used to elicit LULC drivers. Between 1973 and 2021, the total area of woodland, and cultivated and rural settlement land LULC classes in the treated micro-watershed decreased by 24.65% (171.9 ha) and 7.34% (759 ha), respectively. Though, grassland, forestland, and barrenland increased by 30.83% (179.52 ha), 1% (3 ha), and 183.14% (755.28 ha), respectively. The overall area of LULC class for woodland, grassland, forestland, and cultivated and rural settlement decreased by 50.36% (316.16 ha), 41.23% (196.46 ha), 2.43% (11.85 ha), and 1.35% (138.6 ha), respectively, in the untreated micro-watershed, while barrenland increased by 175.86% (666.55 ha). According to KII and FGD, the drivers of LULC changes were identified as the expansion of cultivated land, population pressure, and government policy. According to the findings, local-scale watershed management activities was not as effective as expected. As a result, in order to achieve the desired outcome, the concerned stakeholders should reconsider how watershed management activities is undertaken. KEY WORDS: land use, land cover, watershed management activities, spatio-temporal variation, supervised classification, detection ARTICLE HISTORY: received 6 August 2022; received in revised form 19 November 2022; accepted 23 November 2022 ______________________________________________________________________________________________________________________________________________ 1. Introduction has expanded by 12 million km during the same time frame. The majority of the conversion of Land use and land cover dynamics are the main forests and woodlands to other land use categories driving forces underlying local, regional, and global has taken place in South America, Africa, and Asia. environmental changes (ABEBE ET AL., 2022). Due to a number of factors, such as climate change Almost 1.2 million km of forests and woodlands and population growth (GAREDEW ET AL., 2009), and 5.6 million km of grasslands and pastures East Africa has experienced the second-highest have been converted into various types of land rate of transformation in Africa. use over a long time period and on a worldwide According to ASSEN & NIGUSSIE (2009), each scale (GUAN ET AL., 2008). The area of agriculture region's specific physical and socioeconomic conditions determine LULC changes and their and the problem intensified in different parts of effects. For instance, due to the growth of the country, activities of natural resource agricultural land, the Gubeta-Arjo peasant conservation were primarily implemented in associations in central Ethiopia nearly completely selected watershed sites of the Ethiopian lost all of their forest cover between 1973 and 2006 Highlands, where land degradation was found to (GAREDEW ET AL., 2009). On the other hand, croplands be severe (AMEDE ET AL., 2001; BADEGE, 2001). 2 2 increased by a net 51%, from 403 km to 607 km , in Implementation of watershed management activities the Fincha'a watershed between 1957 and 2001, (WMA) brought some positive outcomes in those at the expense of forests and grazing areas (TEFERA & parts of Ethiopia, especially after the adoption of STERK, 2010). Additionally, in the Chirokella micro- integrated WMA in 2010 (GEBREHAWERIA ET AL., watershed in south eastern Ethiopia, the amount 2016; MELAKU ET AL., 2017). WMA induced at a of dense forest cover declined by nearly 80% watershed scale, for instance, caused a reduction between 1966 and 1996, whereas the area of of 25–38% in soil loss between 2011 and 2015 in moderately forested land was totally transformed the Gumara Maksegnit watershed (MELAKU ET AL., into other LULC categories. In the same micro- 2017), 64% of soil loss in the Agula watershed watershed, the number of cultivated and settlement (between 1990 and 2012) (FENTA ET AL., 2017a), lands increased by 62.8%, whereas the number of 47% of soil loss in the Central Highlands of Ethiopia shrubs and degraded land cover increased by (between 2007 and 2009) (ADIMASSU ET AL., 2012) 49.9% and 100%, respectively (ASSEN & NIGUSSIE, and 65–70% (16.5t ha-1 yr-1) of soil loss at country 2009). Land degradation has inevitably resulted level (FENTA ET AL., 2021). Moreover, watershed from this rapid change of grazing and forest land management induced coverage of dense forest into cultivated and degraded land (HURNI, 1993). (between 1965 and 2005) (ALEMAYEHU ET AL., 2009), Land degradation, prevalent in various and plantation and exclosure (between 2004 and agriculture-based developing countries, including 2009) (HAREGEWEYN ET AL., 2012) were significantly Ethiopia, is manifested by soil nutrient depletion increased in the northern part of the country. and loss of land productivity potential (TAMENE & These and other success stories caused the VLEK, 2008). In Ethiopia, where more than 90% of introduction of WMA in different parts of the country the population depends on agriculture for food through a top-down approach (GEBREHAWERIA ET and a livelihood, soil erosion and soil fertility decline AL., 2016). In line with this, Oromia regional state continue to be a significant problem (EBABU ET AL., announced WMA in 2010/2011 in all parts of the 2020). In Ethiopia, a high rate of soil erosion, region including the study area (OROMIA REGIONAL over-cultivation, improper land use, and poor soil NATIONAL STATE BUREAU OF AGRICULTURE AND management practices, among other things, are NATURAL RESOURCES, 2017). The specific place of Kiro considered to be the causes of land degradation in the Shankur Tareqo micro-watershed, was (AMEDE ET AL., 2001; BADEGE, 2001). In the Highlands, selected as a model area to implement different where inappropriate farming practices have been soil and water conservation measures as it used for centuries and where about 87% of the represents the hilly and gully areas of the district, nation's population lives, the problem of soil erosion which have suffered serious soil erosion for a and the associated reduction in soil fertility is worse long period (BECHO WOREDA NATURAL RESOURCES (EBABU ET AL., 2020). For instance, according to DEPARTMENT (BWNRD), 2021). With the start of BAI ET AL. (2008), about 26.33% (296,812 km ) of WMA, in Kiro and the Shankur Tareqo micro- Ethiopia’s land area was affected by land watershed as a whole, a better achievement has degradation, majorly caused by water-induced soil been observed on soil and water conservation erosion. About 23% of its soil erosion was activities even though different obstacles were contributed by cropland, which constituted about found. As a result, the Shankur Tareqo micro- 50% of the country, with a mean soil erosion loss watershed has been considered the subject of -1 -1 of 36.5 t ha yr (FENTA ET AL., 2021). Due to this, different forms of soil and water conservation Ethiopia annually loses about one billion tonnes activities since 2010/2011. of topsoil, costing about 30% of its gross Thus, in addition to the aforementioned factors domestic product (KIRUI & MIRZABAEV, 2014). (population growth and density, over-intensification To reverse the problem of land degradation of land use, farm size, land tenure status, and lack and achieve the sustainable use of its resources, of land use policies), LULC change-induced land optimistic land management practises have been degradation and land degradation-induced WMA implemented in Ethiopia at a watershed scale, are two sides of the same coin that have an focusing on the Northern Highlands since the 1970s impact on the LULC structure in the study area (AMEDE ET AL., 2001; BADEGE, 2001). As time passed and elsewhere. However, it is hard to find an article that was written on LULC change of the treated micro-watershed covers 12,327 ha and is located between and N district unlike that of the northern part of Ethiopia. Even the few existing articles about LULC changes latitude and to E longitude. (SHAWUL & CHAKMA, 2019; TEFFERA ET AL., 2018) Its elevation ranges between 2101 m. a.s.l. in the covering the watersheds were performed at the lower part, and 2795 m a.s.l. in the upper part, with a macroscale. However, the contribution of LULC slope gradient reaching 69%. It encompasses the change studies at such a large scale to land specific place of Kiro where a government-led WMA management options is less (GEBRELIBANOS & ASSEN, was implemented in collaboration with Shankur 2013). But studies at the micro-scale watershed, village residents on previously degraded communal which were uncommon in the study area, are land. The Kiro specific place covers approximately more effective. In order to devise suitable land 32 ha. It was viewed as a model site to disseminate management practises, strategies, and policies, a technologies and experiences in natural resource thorough analysis of LULC change brought on by conservation to the nearby watersheds. In this specific WMA and other driving variables within the place and the Shankur Tareqo watershed, about micro-watershed is thus urgent and essential. 606 m of gully rehabilitation with trees, 5 m of Hence, this study aimed to detect the spatio- gully rehabilitation with stones, 3.72 km of soil temporal variations of LULC and its driving factors bund, 1.5 km of the stone bund, and 20 micro-basins under watershed management in the Becho district were constructed between 2018 and 2019 (BWNRD, of central Ethiopia between 1973 and 2021. 2021). The adjacent Mende Tufessa micro-watershed located between and N 2. Materials and methods latitude, and and E longitude, covering an area of 12,272 ha, was used as 2.1. Study area a control to check the effectiveness of the introduced WMA, due to the absence of government-led WMA The study was conducted in two adjacent micro- implemented in this micro-watershed. Thus, it watersheds, named Shankur Tareqo (treated) and was considered to be untreated. Mende Tufessa (untreated) in the Becho District, of the Central Highlands of Ethiopia (Fig. 1). The Fig. 1. Location map of Shankur Tareqo and Mende Tufessa watersheds, in Becho district The underlying geology in each micro-watershed of and , respectively. The mean consists of Nazareth series volcanoes, forming annual temperature and average annual rainfall parent materials for the vertisols that dominate are 16°C and 1127 mm, respectively (Fig. 2). The the landscape of the watersheds. About 90% of subsistence rain-fed farming system forms the the soils in the watersheds are vertisols, while main economic activity in both micro-watersheds. the remaining are occupied by luvisols, nitisols, Locally, maize (Zea mays), teff (Eragrostis tef), wheat and leptosols (ELIAS, 2016). (Triticum aestivum), and chickpea (Cicer arietinum) Both micro-watersheds are predominately are the major cultivated crops. Furthermore, oxen, grouped into the mid-altitude zone (ELIAS, 2016). cows, goats, sheep, hens, horses, mules, and According to long-term meteorological data (1990– donkeys are the most common animals found in 2019), these micro-watersheds were characterised by the micro-watersheds (BWNRD, 2021). mean maximum and minimum annual temperatures Fig. 2. Rainfall and temperature records of the study area based on the records of nearby Tulu Bolo station (1990–2019) from the national meteorological service agency, Ethiopia 2.2. Data acquisition, sources and processing Scanned aerial photos with a photogrammetric scanner at 1200 dots per inch (DPI) in tagged image The LULC change analysis of the study area file (TIF) format were independently orthorectified was carried out using a collection of aerial photos and registered to a coordinate system of WGS 1984 and various satellite imagery. Accordingly, the UTM Zone 37N using the Transverse Mercator Ethiopian Geospatial Information Institute (EGII) projection system with less than 0.85 root mean provided seven aerial photos for 1973 and a Spot 5 square error (RMSE) by the data providers, that image for 2005, as well as two sets of topographic is EGII. These rectified pieces of aerial photos maps at 1:50000 scale. The remaining Landsat were mosaicked using Erdas Imagine 2014. After Collection 1 level-1 data product images of 1990 calibrating at sensor radiance, radiometric correction and 2021, and digital elevation model were freely of Landsat 5 TM and Landsat 8 OLI was carried downloaded from the USGS website in December, out using the methods described by ZADBAGHER ET 2021 (https://earthexplorer.usgs.gov). Furthermore, AL. (2018) and PRIETO-AMPARÁN ET AL. (2016), the watershed boundaries were delineated using respectively, in Erdas Imagine 2014. All of these digital elevation model. The images were chosen images were reprojected to a WGS 84 into a UTM based on data availability (1973 aerial photos), Zone 37N coordinate system using a 1:50,000 anticipated major changes (1990), introduced topographic map as the base map to avoid any watershed management (2005) and year consistency projection distortion and displacement that might (2021). occur among those images. Finally, the study area was extracted using the this process because they had the same tonal boundary of watersheds on the mosaicked geo- value on the image. Therefore, rural settlements referenced aerial photos of 1973 and the satellite and cultivated land were amalgamated and called images. Investigation of shape, size, pattern, tone, "cultivated and rural settlement land”, while texture, shadow, and site association were used plantation and natural forest, which are commonly to identify LULC classes in mosaicked aerial found along riversides, were simply called photos of 1973. However, in order to improve "forestland". Using the existence of vegetation, delineation objectivity, interpretation repeatability, edaphic conditions and artificiality of cover as and processing efficiency, locally available LULC explained in the Land Cover Classification system types from satellite images were identified through (DI GREGORIO, 2005), five LULC types–barren land supervised classification (maximum likelihood (BL), forestland (FL), cultivated and rural settlement techniques) algorithms with the aid of Erdas Imagine land (CRSL), grassland (GL) and woodland (WL)– 2014 and ArcGIS 10.7. The indistinguishability of were identified in the studied micro-watersheds rural settlements and cultivated land, as well as (Table 1). natural forest and plantation, was recognised during Table 1. LULC types identified in the treated and untreated micro-watersheds LULC types Description BL Refers to areas of exposed stone, sand and soil or area with no vegetation or dominated by rock-out crops, eroded and degraded lands found along the flooding area of the local stream valleys, over gentle and steep mountain slopes CRSL Refers to areas covered with annual rainfed crops and scattered rural dwelling FL Refers to plantations, the disturbed natural patchy forest cover of mature trees and riparian vegetation GL Refers to areas covered with grasses WL Areas covered with sparse woody plants mixed with shrubs, bushes, and grasses. It also incorporates perennial cropland with 30-50% moderately stocked tree crown cover NB: WL=Woodland, GL=Grassland, FL=Forestland, CRSL=Cultivated and Rural Settlement land and BL=Barren land Furthermore, information that is not possible of the temporal changes of the LULC types and to extract from aerial photos or satellite images computed the extent of LULC conversion induced was obtained through open-ended guided questions by WMA and physical, socio-economic and policy used for focus group discussion (FGD) and key related factors. Thus, the rate of absolute variation of informant interview (KII) as well as from personal a single LULC category between two time points field observations. A total of three FGDs were was quantitatively determined using the held in both watersheds, comprising of 5 to 8 equations 1 and 2. Additionally, to have a participants selected from local elders, community complete picture of the conversion between leaders, and development agents. KII were also various LULC types, transition matrices were conducted with 12 (6 from each watershed) made for the different periods of analysis. purposefully selected local residents aged between 30 and 65 to gain further insight into issues 1 related to LULC changes perceived throughout their lives. 2 2.3. LULC spatio-temporal change detection methods where S is the absolute variation in the area of a certain land type in the study period; K is the Post classification change detection techniques, change rate of a land type with a positive value which reduced the possible effects of spectral representing an increasing rate and a negative resolution and sensor differences between the value representing a decreasing rate; Ua and U is multi-temporal image (ABERA ET AL., 2018) were the area of a land type at the beginning and the used to analyse the change in LULC for the end of the study period in hectares. periods 1973–1990, 1990–2005, 2005–2021, and 1973–2021. This method enabled an assessment 3. Results and discussion grassland, forestland, and barren land were the common LULC types identified in the watersheds. The LULC types and their spatial distribution Furthermore, the LULC transition matrix determined for Shankur Tareqo and Mende Tufessa micro- for the years between 1973 and 2021 are presented watersheds are shown in Figs. 3 and 4 as well as in Table 3. Table 2. Cultivated and rural settlement, woodland, Fig. 3. LULC distribution of Shankur Tareqo and Mende Tufessa micro-watershed for each year from 1973 to 2021 3.1. Cultivated and rural settlement land (CRSL) migration in the Gedalas watershed of the north- LULC pattern from 1973–2021 eastern Highlands of Ethiopia, which resulted in a decline in CRSL. MEKUYIE ET AL. (2018) also noted CRSL comprises the largest share of the a decrease in farmland between 1985 and 1995 watersheds throughout the study period (1973– as a result of the movement of pastoralists in 2021). The share of this LULC category in the response to climate change. In the later periods of treated micro-watershed accounted for 84%, 69%, analysis (1990–2005 and 2005–2021), CRSL LULC 72%, and 78% for the years 1973, 1990, 2005, classes showed an increasing trend. The highest -1 and 2021, respectively (Fig. 4). Except for the increment of 7.89%, with a rate of 43.8 ha yr , years 1973–1990, CRSL LULC types increased was observed in the years between 2005 and continuously (Table 2). In the first period of analysis 2021 (Table 2). This increment is the result of (1973–1990), a considerable reduction, by 17.48%, 753.5 ha of WL, 565.8 ha of BL, 469.9 ha of GL, of this LULC category was recorded due to rural- and 206.5 ha of FL conversion to CRSL between urban migration and Beherawi Witidrina , as per 2005 and 2021 (Table 3). In general, the CRSL KII and FGD. This was in line with the study of LULC classes decreased by ~7% over the study YESUPH & DAGNEW (2019), who found that between period, between 1973 and 2021, in the treated 1973 and 1986, there was a major rural out- micro-watershed. In the untreated micro-watershed, after a 1 decline from 83.93% in 1973 to 68.34% in 1990, It was the Derg regime's forced conscription of young male working-age groups into its wars with the Tigray People's CRSL showed a significant expansion from 69.35% Liberation Front, the Eritrean People's Liberation Front, and in 2005 to 82.78% in 2021 (Fig. 4). Between 2005 the Ogaden War. and 2021, the highest increment of 19.38%, with -1 a rate of 103.07 ha yr , was observed (Table 2). by ~7% throughout their study periods between CRSL expanded at the expense of 1023.1 ha of 2000 and 2017 in north west Ethiopia (WORKU ET WL, 631.7 ha of GL, 501.7 ha of BL, and 420.1 ha AL., 2021), ~1% between 1985 and 2017 in south of FL (Table 3). The primary reason for this west Ethiopia (FASIKA ET AL., 2019), ~14% increment, as mentioned by the stakeholders, was between 1985 and 2010 in the Abaya-Chamo population pressure. During the study period, basin (WOLDEYOHANNES ET AL., 2018), and ~24% CRSL in this micro-watershed decreased by 1.35%. between 1985–2015 in the Afar region (MEKUYIE The result of this study was in agreement with ET AL., 2018). the findings of those who found a decline of CRSL Fig. 4. Aerial coverage of LULC change in the Shankur Tareqo (A) and Mende Tufessa (B) micro-watersheds in 1973, 1990, 2005 and 2021 3.2. Grassland (GL) LULC pattern from 1973–2021 hence causing an increase in GL and WL LULC types during this time period. In contrast to this In 1973, GL covered approximately 5% of the period of analysis (1973–1990), aerial coverage treated micro-watershed, which increased to 13.14% of the GL LULC class decreased by 38.03% and in 1990. After that, GL showed continuous decline. 24.05% in the second (1990–2005) and third It declined from 8.14% in 2005 to 6.18% in 2021 (2005–2021) periods of analysis, respectively. (Fig. 4). Between 1973 and 1990, the GL LULC class The highest decline was recorded in the second -1 increased by 177.97% at a rate of 60.96 ha yr period (1990–2005) of analysis, which was (Table 2). This is the opposite side of a decline in attributed to a conversion of 758.3 ha, 298.7 ha, CRSL in which ~1293 ha of CRSL was converted 115.6 ha, and 72.9 ha of GL to CRSL, WL, BL, and to GL (Table 3). Thus, the source of shrinkage in FL, respectively (Table 3). Throughout the study CRSL was probably due to its part being taken period (1973–2021), a total of 312.6 ha (53.7%) over by GL that, over time, transformed into WL, of GL was converted to CRSL LULC types. Similarly, in an untreated micro-watershed, from population pressure (Table 2). The results the GL LULC class accounts for nearly 3.88% of of GAREDEW ET AL. (2009) and YESUPH & DAGNEW the total area in 1973 and 9.67% in 1990 (Fig. 4). (2019), found an increase in GL from 1973 to This showed an increment of GL which was 1986 and subsequent continuous reductions were attributed to a decline in CRSL following rural– perfectly in agreement with this study. Similar urban migration and Beherawi Witidrina as findings were also reported by DEGIFE ET AL., mentioned by the elders. However, in the later (2019) for the period of 1972–1992 in the Lake years, it gradually declined from 7.74% in 2005 Hawassa watershed. This LULC class decreased to 2.28% in 2021. The primary contraction took by 41% throughout the study period. This was in place from 2005 to 2021 by 70.52%, with a rate of line with the results of ASMAMAW ET AL. (2011) -1 41.87 ha yr (Table 2). Much of this LULC class and GEBRELIBANOS & ASSEN (2013) in the Gerado was changed into CRSL (631.7 ha, ~67%) due to and Hirmi Watersheds of the north-eastern and the high demand for agricultural land resulting northern highlands of Ethiopia, respectively. Table 2. Absolute variation in the area (S) and change rate (K) of LULC in the treated and untreated micro-watershed in different periods; increase (+) and decrease (-) Shankur Tareqo (Treated) micro-watershed LULC 1973-2021 1973-1990 1990-2005 2005-2021 Classes K (%) S (ha) K (%) S (ha) K (%) S (ha) K (%) S (ha) WL -24.65 240.36 34.47 311.9 33.26 -724.12 -57.95 -171.86 GL 30.83 1036.39 177.97 -615.64 -38.03 -241.23 -24.05 179.52 FL 1.09 391.59 141.25 -343.43 -51.35 -45.13 -13.87 3.03 CRSL -7.34 -1808.54 -17.48 348.72 4.08 700.74 7.89 -759.08 BL 183.14 147.85 35.85 299.65 53.49 307.78 35.79 755.28 Mende Tufessa (Untreated) micro-watershed LULC 1973-1990 1990-2005 2005-2021 1973-2021 Classes K (%) S (ha) K (%) S (ha) K (%) S (ha) K (%) S (ha) WL -50.36 438.51 69.85 380.72 35.71 -1135.39 -78.46 -316.16 GL -41.23 709.94 148.99 -236.42 -19.93 -669.98 -70.52 -196.46 FL -2.43 475.61 97.55 -321.17 -33.35 -166.29 -25.9 -11.85 CRSL -1.35 -1911.75 -18.57 124.05 1.48 1649.1 19.38 -138.6 BL 175.86 289.07 76.27 53.91 8.07 323.58 44.82 666.55 NB: WL=Woodland, GL=Grassland, FL=Forestland, CRSL=Cultivated and Rural Settlement land and BL=Barren land 3.3. Woodland (WL) LULC pattern from 1973–2021 throughout the study period (1973–2021), this LULC class exhibited a net loss of 24.65% (171.86 ha) The area of the WL class in the treated micro- in which the majority was transformed into CRSL watershed was comprised of 5.66%, 7.61%, 10.14%, (45.7%, 318.2 ha) (Table 3). and 4.26% in the years 1973, 1990, 2005, and In the untreated micro-watershed, GL LULC 2021, respectively (Fig. 4). This showed a gain of WL accounted for 5.12%, 8.69%, 11.79%, and 2.54% in the first (1973–1990) and second (1990–2005) of the total area in 1973, 1990, 2005, and 2021, periods of analysis, but a decline in the third period respectively (Fig. 4). The WL LULC class in this of analysis (2005–2021) (Table 2). The highest watershed increased by 69.85% and 35.71% in (34.47%) rise of the WL LULC class was detected the periods of analysis from 1973 to 1990 and in the first period of analysis (1973–1990) with a from 1990 to 2005, respectively. According to KII -1 rate of 14.14 ha yr due to the conversion of and FGD participants, the socialist regime's policy CRSL (806 ha, 7.8%) to WL. Thus, the increase in of rehabilitating degraded land in the 1980s WL area is presumably attributed to gains from CRSL through exclosure of hillsides and tree-planting due to the reasons mentioned below. In contrast, efforts led to an increase in the WL class in both from 2005 to 2021, WL area coverage decreased micro-watersheds throughout the first and -1 by 57.95% at a rate of 45.26 ha yr . However, second periods of analysis. In contrast, it declined by 78.46% in the third period (2005–2021) of Ethiopia was overthrown in 1991, and the expansion analysis (Table 2). In the overall study period of CRSL, as mentioned by KII and FGD, may all be (1973–2021), WL LULC dwindled by 50.36% in responsible for the overall decline in WL in both the untreated micro-watershed. This loss was micro-watersheds. This is consistent with research largely associated with its conversion to a CRSL done in other regions of Ethiopia (YESUPH & (467 ha, 74.5%) (Table 3). DAGNEW, 2019). However, the overall decline in The need for more firewood due to population the treated micro-watershed was significantly growth, the redistribution of land to ex-soldiers lower, which may have been caused by the who have returned since the military regime in government's adoption of WMA. Table 3. LULC conversion matrices in different periods in each watershed (area in ha) Shankur Tareqo (Treated) Mende Tufessa (Untreated) Changed Changed 1973- 1990- 2005- 1973- 1973- 1990- 2005- 1973- from to 1990 2005 2021 2021 1990 2005 2021 2021 ha ha ha ha ha ha ha ha WL 51.9 125.7 147.7 284.1 82.7 184.9 90.3 57 GL 114.5 29.1 91.8 29.6 72.9 22.3 33.5 11.5 FL 79.2 76.2 125.2 19.8 69.1 240.9 170.7 47.1 Woodland (WL) CRSL 427.2 681.9 753.5 318.2 375.4 589.9 1023.1 467 BL 23.6 23.1 127.5 44.5 27.1 27.4 127.3 44.5 Total 696.4 936 1245.7 696.2 627.1 1065.4 1444.9 627.1 WL 26.2 298.7 67.5 16.5 36.3 229 35 12.2 GL 114 370.3 313.1 205.9 111.4 275.1 128.5 76 FL 19.6 72.9 19.7 8 49.9 86.3 19.6 34.7 Grassland (GL) CRSL 380.7 758.3 469.9 312.6 254.7 542.5 631.7 309.6 BL 41.4 115.6 132 39 23.6 52.1 134.7 43.5 Total 581.9 1616 1002.1 582 475.9 1185 949.5 476 WL 23.7 149.5 58.3 10 78.1 182.2 34.5 17.2 GL 35.9 9.8 17.9 11.8 48.3 9.3 7.9 6.6 FL 23.9 81.6 17.6 29.9 89.4 232 136.1 52.1 Forestland (FL) CRSL 185.1 413.3 206.5 198.4 255.4 525.9 420.1 378.7 BL 8.2 13.1 23.8 26.7 16 12.4 43.4 32.6 Total 276.8 667.3 324.1 276.8 487.1 961.7 642 487.2 WL 806 622.7 196.6 202.5 841.7 786.4 126.9 214.9 GL 563.6 94.7 180 1293 516.8 252 496.6 920.1 FL 524.7 81 98.3 216.3 725.7 78.5 123.5 331.3 CRSL CRSL 7240 6680 7576 8523.7 7216.9 6423.5 7569.2 8729.2 BL 464.5 620.9 749 892.3 575.9 522.2 583.8 828.4 Total 10329 8521 8872 10331.4 10280.3 8374.1 8498.2 10283.9 WL 27.7 49.1 54.9 11.6 26.1 62.5 24.4 9.6 GL 58.1 76.1 86.4 17.4 31.5 79.1 15.3 5.5 FL 19.1 12.7 19 5.1 27 4.4 25.2 9.4 Barren land (BL) CRSL 285.7 335.7 565.8 215.7 269.7 414.5 501.7 259.1 BL 20.7 84.7 131.2 161.8 23.9 106 153.5 94.9 Total 411.4 558.3 857.3 411.7 378.2 666.5 720.1 378.5 NB: CRSL=Cultivated and Rural Settlement land 3.4. Forestland (FL) LULC pattern from 1973–2021 revealed a declining rate of increase in the treated micro-watershed after the implementation of WMA, In 1973 and 1990, around 2.25% and 5.43% which was also affirmed through FGD. It increased of treated micro-watershed were classified as FL, by 183.14% over the entire study period (1973– respectively (Fig. 4). This indicates a 141.25% 2021), much of which came from CRSL classes increase in FL cover class over 17 years, with a rate (892.3 ha, 8.6%) (Table 3). -1 of 23.03 ha yr in the first period of analysis The spatial coverage of the BL class in the (1973–1990), compared to a decrease in the second untreated micro-watershed comprised of 3.09%, (1990–2005) and third (2005–2021) periods of 5.45%, 5.88%, and 8.52% in 1973, 1990, 2005, analysis (Table 2). The increment in the first period and 2021 respectively (Fig. 4). The rate of increment of analysis was ascribed to the replantation of the BL LULC class in the first (1973–1990) and programme announced and implemented by the third (2005–2021) periods of analysis was higher socialist regime that preserved the remnant for the untreated compared to the treated micro- indigenous trees and expanded forest cover. watershed (Table 2). However, in the overall Moreover, politically affiliated rural-urban migration, period (1973–2021) of analysis, the rate of BL which abandoned CRSL and served to regenerate increment was higher for the treated micro- remnant trees, was also mentioned as a cause by watershed. As shown in (Table 3), the treated the elders. This was similar to the result of a micro-watershed still had the highest conversion reported increment of forestland cover between rate of CRSL to BL (8.6%), compared to 8.06% in 1970 and 1980 in the Rib watershed (MOGES & BHAT, the untreated one. This confirms that, in the 2017). Conversely, in the second (1990–2005) and treated micro-watersheds, further efforts will be third (2005–2021) periods of analysis, the FL class needed to rehabilitate the degraded land through exhibited a reduction of 51.35% and 13.85%, soil and water conservation activities. According respectively, as a consequence of its conversion to the discussion with the KII, this was one of the to CRSL (61.9%, 413.3 ha and 63.7%, 206.5 ha) factors that led to the selection of the treated (Table 3). The increased demand for cropland and micro- watershed for rehabilitation activities. the use of trees for construction purposes were the apparent causes of reduction of the FL cover. 3.6. Major drivers of LULC change In the untreated micro-watershed, FL increased from 3.97% in 1973 to 7.85% by 1990. In the 3.6.1. Expansion of cultivated and rural latter period, it gradually declined from 5.23% in settlement land 2005 to 3.88% by 2021 (Fig. 4). In the overall study period (1973–2021) of analysis, FL class in Except for the first period of analysis (1973– the treated micro-watershed showed an increment 1990), the CRSL class was shown to have -1 of 1.09% with a rate of 0.06 ha yr while it significant expansion at the expense of WL, GL, decreased by 2.43% in the untreated one. This and FL in both watersheds. In the first period of shows an improvement of FL LULC class in the analysis (1973–1990), CRSL classes surprisingly treated micro-watershed, owing primarily to declined in both micro-watersheds from where replantation activities in the Kiro specific location they were in 1973. This finding sharply contrasts and the expansion of eucalyptus plantation, as with that of REID ET AL. (2000), who reported a revealed by KII and FGD participants. Thus, this significant decrease in cultivated land in Gullele would be a small benefit detected in the treated (27%), in Gerangera (50%) and in Kumbi (67%) micro-watershed due to the introduction of WMA. between 1973 and 1987 in the Gibe valley (south Similar findings were also observed in north- western Ethiopia). BEWKET & STERK (2002) also eastern Ethiopia (ASMAMAW ET AL., 2011) and reported a shrinking of farmland by 2% between northern Ethiopia (GEBRELIBANOS & ASSEN, 2013). 1982 and 1998 in the Chemoga Watershed, Ethiopia. According to KII and FGD participants, the socialist 3.5. Barren land (BL) LULC pattern from 1973–2021 regime's shift of land ownership from landlord to peasant association (which prevented some farmers In the treated micro-watershed, BL accounts from cultivating land previously owned by the for 3.35%, 4.55%, 6.98%, and 9.48% of the total landlord due to widespread fear), Beherawi area in 1973, 1990, 2005, and 2021, respectively Witidrina, and rural-urban migration were (Fig. 4). The percentage of each increment was mentioned as reasons for change. 35.85%, 53.49%, and 35.79% in the first (1973– In the second (1990–2005) and third (2005– 1990), second (1990–2005), and third (2005–2021) 2021) periods of analysis, the CRSL LULC class periods of analysis, respectively (Table 2). This increased at the expense of other LULC classes due to a high demand for cereal crops resulting of livelihood security had forced farmers to use from population pressure within both micro- the woodlands to cope with recurrent household watersheds (Table 3). This increment in CRSL shocks in the Central highlands of Ethiopia, which was similar to the findings of SHIFERAW ET AL. further modified the LULC of the area. Therefore, (2001) and TEFERA (2011), who reported an it can be concluded that population growth has increment of cropland in Borena Woreda (South driven the expansion of CRSL in the studied Wollo) and Nono Woreda (Central Ethiopia) due watersheds at the expense of other LULC classes, to rapid population growth. However, the lower which have ended up being degraded, as observed in rate of increment in CRSL class was recorded for other parts of the country. the treated micro-watershed for the third period (2005–2021) of analysis (Table 2). This was due 3.6.3. Policy changes related factor to the introduction of WMA. Besides, as per the information gained from KII and FGD participants, A change in policy is a perceptible departure the low-rate of increment of CRSL in the treated from the intended path of action that had been micro-watershed was due to a lack of additional put in place to solve a perceived societal issue farmland, rural-urban migration of the young (MICHAELS ET AL., 2006). Policy change follows a working age group, and resettlement programmes predetermined process that involves problem undertaken by Oromia regional state in 2005. identification, adoption of a specific policy, Therefore, CRSL change caused LULC change in implementation of the policy, and evaluation of the studied micro-watersheds. the policy (MICHAELS ET AL., 2006). The last step is crucial in identifying the effect of the policy on 3.6.2. Demographic related factors the environment. In line with this, long-term residents of the area were requested to explain Demographic factors are one of the underlying the policy-related factors responsible for the causes that brought LULC change in different parts change in LULC. Accordingly, the informants cited of the country (ABERA ET AL., 2018; GEBRELIBANOS the policy changes of 1974 and 1991 as the main & ASSEN, 2013), which is implicated in the study drivers of LULC transformation in the study area. area also. According to the CSA's population census In accordance with the well-known motto "Land and projection (CENTRAL STATISTICAL AGENCY, 2007, to the Tiller," which was elicited by KII and FGD 2013, 2021), the rural population of the district participants, the first policy change nationalized increased rapidly from a total of 44,382 in 1994 the land and the natural resources on it, resulting to 82,737 in 2021. This showed an increase in the in FL and WL being owned by the government. rural population of more than 86.6% between The second policy change resulted in the extensive 1994 and 2021. To some extent, an increase in destruction of vegetated areas as portions of the human population in the Highlands has also formerly government-owned tracts of land were caused an increase in the population of livestock given to private investors and distributed among (DUBALE, 2001). Such rapid population growth, smallholder farmers. In order to bring about combined with an increase in the livestock LULC change in the watersheds under research, population in the area, has put pressure on the the second policy modification has the biggest area's limited land resources by increasing influence. demand for food, fuel, construction materials, According to KII and FGD, much of the expansion grazing land, and permanent settlement land. of CRSL, BL, and loss of FL LULC classes were These issues were also confirmed through KII found after the downfall of the socialist regime. and FGD participants as well as the quantitative Before the downfall, the number of people residing data presented in Fig. 3, suggesting a positive in the watershed was low due to rural–urban relationship between population growth and migration and Beherawi Witidrina, which created expansion of the cultivation land at the expense surplus land. Moreover, they also stated that the of other land use types. This was in line with a socialist government planted trees in a finding by ZELEKE & HURNI (2001), who reported a mountainous area that extended from the hilly decline in forest cover in central Gojjam from areas of Shankur and Mende kebeles to Dilella 27% to 0.3% between 1957 and 1995 due to towns on the land taken from the previous population pressure. Forest cover declined from landlord. This was partly responsible for the 34% to 3% in the country between 1950 and slight increase in GL, WL, and FL in the first 1980 due to population pressure-induced factors period of analysis (1973–1990). (ASSEN & NIGUSSIE, 2009). Moreover, GAREDEW ET AL. (2009) reported that population-induced lack It is the lowest administrative unit in Ethiopia The Ethiopian People’s Revolutionary Democratic Acknowledgements Front government that succeeded the socialist The authors wish to acknowledge Addis Ababa University for regime in 1991 subsequently advocated state its financial support and the USGS for providing free Landsat ownership of land whereby only usufruct rights data. In addition, the authors would like to acknowledge the local are bestowed upon landholders. According to communities, Becho district administrators, and development FGD participants, this policy shift encourages the agents for their valuable support during the data collection process by providing and collecting relevant information. private sector and individual farmers to engage in free and uncontrolled exploitation of FL in order References to have more cultivated land, particularly by clearing previously planted trees. It can, therefore, Abebe M.T., Degefu M.A., Assen M., Legass A. 2022. Dynamics be concluded that government policy has played of land use/land cover: implications on environmental a significant role in LULC change in the study area. resources and human livelihoods in the Middle Awash The results from ABERA ET AL. (2018), ASMAMAW Valley of Ethiopia. Environmental Monitoring and Assessment, ET AL. (2011), and GEBRELIBANOS & ASSEN (2013) in 194, 11: 833. Abera W., Assen M., Satyal P. 2018. Spatio-temporal land the Wanka, Gerado, and Hirmi Watersheds of the use/cover dynamics and its implication for sustainable northwest, north-eastern, and northern highlands of land use in Wanka watershed, northwestern highlands of Ethiopia, respectively, were similar to these. Ethiopia. Modeling Earth Systems and Environment, 5, 2: 571–581. Adimassu Z., Mekonnen K., Yirga C., Kessler A. 2012. Effect of 4. Conclusions Soil Bunds on Runoff, Soil and Nutrient Losses, and Crop Yield in the Central Highlands of Ethiopia. Land Based on mosaicked aerial photos and Landsat Degradation & Development, 25, 6: 554–564. images, LULC changes were calculated for the Alemayehu F., Taha N., Nyssen J., Girma A., Zenebe A., treated and untreated micro-watersheds of the Behailu M., Deckers S., Poesen J. 2009. The impacts of watershed management on land use and land cover central highlands of Ethiopia using in-depth dynamics in Eastern Tigray (Ethiopia). Resources, analysis. The result indicated that CRSL in both Conservation and Recycling, 53, 4: 192–198. watersheds declined between 1973 and 1990, Amede T., Geta E., Takele B. 2001. Reversing the degradation and they experienced upsurges later. From 1973 of arable land in Ethiopian Highlands, Managing African Soils, Vol. 23. to 1990, WL, GL, and FL grew in both watersheds, Asmamaw L.B., Mohammed A.A., Lulseged T.D. 2011. Land whereas CRSL decreased. Due to the growth of BL use/cover dynamics and their effects in the Gerado in the chosen micro-watersheds, FL, WL, and GL catchment, northeastern Ethiopia. International Journal decreased in the following years. These LULC of Environmental Studies, 68, 6: 883–900. types declined at a slower rate in the treated Assen M., Nigussie T. 2009. Land use/cover changes between 1966 and 1996 in Chirokella micro-watershed, Southeastern micro-watershed. This lower rate of decline was Ethiopia. East African Journal of Sciences, 3, 1: 1–8. attributed to the WMA performed there. However, Badege B. 2001. Deforestation and Land Degradation in the significant variation between the two micro- Ethiopian Highlands: A Strategy for Physical Recovery. watersheds was not observed. Thus, this implicates Northeast African Studies, 8, 1: 7–26. Bai Z., Dent D.L., Olsson L., Schaepman M.E. 2008. Global the ineffectiveness of the WMA. Because of this, assessment of land degradation and improvement: 1. as the land loses its protective natural cover as a identification by remote sensing: ISRIC-World Soil Information. result of the growth of agricultural land, population Bewket W., Sterk G. 2002. Farmers' participation in soil and pressure, and other government policy-induced water conservation activities in the Chemoga Watershed, causes, the examined watersheds become more Blue Nile basin, Ethiopia. Land Degradation & Development, 13, 3: 189–200. and more vulnerable to land degradation. Thus, BWNRD. 2021. Socio-economic and Natural Resource Report the patterns of LULC changes in the examined Becho Woreda of 2020. Becho Woreda Natural Resource micro-watersheds have nevertheless been greatly Department, Becho. influenced by an interaction of physical, socio- Central Statistical Agency. 2007. Population and Housing Census 2007: Oromia statistical, 1084. Central Statistical economic, and policy issues. In order to identify Agency, Addis Ababa. observed impediments, previously deployed WMA Central Statistical Agency. 2013. Population Projection of should be reviewed. Furthermore, the chosen Ethiopia for All Regions At Wereda Level from 2014– micro-watersheds should adopt a suitable 2017, 118. Central Statistical Agency, Addis Ababa. additional watershed management plan based on Central Statistical Agency. 2021. Population Projection Wereda as of July 2021: 39. Central Statistical Agency, the knowledge gained from the prior WMA. Addis Ababa. Reforestation of the watershed's degraded areas, Degife A., Worku H., Gizaw S., Legesse A. 2019. Land use land protection and enhancement of the current FL cover dynamics, its drivers and environmental implications cover, and other successful soil and water in Lake Hawassa Watershed of Ethiopia. Remote Sensing Applications: Society and Environment, 14: 178–190. conservation techniques should be included in this effort to prevent additional land degradation. Di Gregorio A. 2005. Land cover classification system: in the evolution of watershed management. Environmental classification concepts and user manual: LCCS: Food & Management, 38, 6: 983–992. Agriculture Organization. Moges D.M., Bhat H.G. 2017. Integration of geospatial Dubale P. 2001. Soil and Water Resources and Degradation technologies with RUSLE for analysis of land use/cover Factors Affecting their Productivity in the Ethiopian change impact on soil erosion: case study in Rib Highland Agro-ecosystems. International Conference on watershed, north-western highland Ethiopia. Environmental African Development Archives. Earth Sciences, 76, 22: 1–14. Ebabu K., Tsunekawa A., Haregeweyn N., Adgo E., Meshesha Oromia Regional National State Bureau of Agriculture and D.T., Aklog D., Masunaga T., Tsubo M., Sultan D., Fenta Natural Resources. 2017. Effort of Integrated Watershed A.A., Yibeltal M. 2020. Exploring the variability of soil Management Towards Reversing Land Degradation. Oromia properties as influenced by land use and management Regional National State Bureau of Agriculture and practices: A case study in the Upper Blue Nile basin, Natural Resources. Ethiopia. Soil and Tillage Research, 200: 104614. Prieto-Amparán J., Pinedo-Alvarez A., Villarreal-Guerrero F., Elias E. 2016. Soils of the Ethiopian Highlands: Geomorphology and Pinedo-Alvarez C., Morales-Nieto C., Manjarrez-Domínguez C. Properties The Netherlands.: CASCAPE Project, ALTERA, 2016. Past and Future Spatial Growth Dynamics of Wageningen University and Research Centre (Wageningen Chihuahua City, Mexico: Pressures for Land Use. ISPRS UR). International Journal of Geo-Information, 5, 12: 1–19. Fasika A., Motuma T., Gizaw T. 2019. Land Use Land Cover Reid R.S., Kruska R.L., Muthui N., Taye A., Wotton S., Wilson Change Trend and Its Drivers in Somodo Watershed C.J., Mulatu W. 2000. Land-use and land-cover dynamics South Western, Ethiopia. African Journal of Agricultural in response to changes in climatic, biological and socio- Research, 14, 2: 102–117. political forces: the case of southwestern Ethiopia. Fenta A.A., Tsunekawa A., Haregeweyn N., Tsubo M., Yasuda Landscape Ecology, 15, 4: 339–355. H., Kawai T., Ebabu K., Berihun M.L., Belay A.S., Sultan D. Shawul A.A., Chakma S. 2019. Spatiotemporal detection of 2021. Agroecology-based soil erosion assessment for land use/land cover change in the large basin using better conservation planning in Ethiopian river basins. integrated approaches of remote sensing and GIS in the Environmental Research, 195: 110786. Upper Awash basin, Ethiopia. Environmental Earth Sciences, Fenta A.A., Yasuda H., Haregeweyn N., Belay A.S., Hadush Z., 78, 5: 1–13. Gebremedhin M.A., Mekonnen G. 2017a. The dynamics of Shiferaw B., Holden S., Aune J. 2001. Population Pressure urban expansion and land use/land cover changes using and Land Degradation in the Ethiopian Highlands: A Bio- remote sensing and spatial metrics: the case of Mekelle Economic Model with Endogenous Soil Degradation. In: City of northern Ethiopia. International Journal of Remote Heerink, N., van Keulen, H., Kuiper, M. (eds) Economic Policy Sensing, 38, 14: 4107–4129. and Sustainable Land Use: Recent Advances in Quantitative Garedew E., Sandewall M., Soderberg U., Campbell B.M. 2009. Analysis for Developing Countries. Heidelberg: 73–92. Land-use and land-cover dynamics in the central rift valley of Tamene L., Vlek P.L.G. 2008. Soil Erosion Studies in Northern Ethiopia. Environmental Management, 44, 4: 683–694. Ethiopia. In: A.K. Braimoh, P.L.G. Vlek (Eds.) Land Use and Gebrehaweria G., Dereje A.A., Girmay G., Giordano M., Langan S. Soil Resources. Springer Netherlands, Dordrecht: 73-100. 2016. An assessment of integrated watershed management Tefera B., Sterk G. 2010. Land management, erosion problems in Ethiopia. IWMI Working Paper (170). and soil and water conservation in Fincha’a watershed, Gebrelibanos T., Assen M. 2013. Land use/land cover western Ethiopia. Land Use Policy, 27, 4: 1027–1037. dynamics and their driving forces in the Hirmi watershed Tefera M.M. 2011. Land-use/land-cover dynamics in Nonno and its adjacent agro-ecosystem, highlands of Northern District, Central Ethiopia. Journal of Sustainable Development Ethiopia. Journal of Land Use Science, 10, 1: 81–94. in Africa 13, 1: 123–140. Guan D., Gao W., Watari K., Fukahori H. 2008. Land use change Teffera Z.L., Li J., Debsu T.M., Menegesha B.Y. 2018. Assessing of Kitakyushu based on landscape ecology and Markov land use and land cover dynamics using composites of model. Journal of Geographical Sciences, 18, 4: 455–468. spectral indices and principal component analysis: A Haregeweyn N., Fikadu G., Tsunekawa A., Tsubo M., case study in middle Awash subbasin, Ethiopia. Applied Meshesha D.T. 2012. The dynamics of urban expansion Geography, 96: 109–129. and its impacts on land use/land cover change and Woldeyohannes A., Cotter M., Kelboro G., Dessalegn W. 2018. small-scale farmers living near the urban fringe: A case Land Use and Land Cover Changes and Their Effects on study of Bahir Dar, Ethiopia. Landscape and Urban the Landscape of Abaya-Chamo Basin, Southern Ethiopia. Planning, 106, 2: 149–157. Land, 7, 1: 1–17. Hurni H. 1993. Land degredation, famine, and land resource Worku T., Mekonnen M., Yitaferu B., Cerdà A. 2021. scenarios in Ethiopia. Conversion of crop land use to plantation land use, Kirui O.K., Mirzabaev A. 2014. Economics of land degradation northwest Ethiopia. Trees, Forests and People, 3: 100044. in Eastern Africa. ZEF Working Paper Series. Yesuph A.Y., Dagnew A.B. 2019. Land use/cover Mekuyie M., Jordaan A., Melka Y. 2018. Land-use and land- spatiotemporal dynamics, driving forces and cover changes and their drivers in rangeland-dependent implications at the Beshillo catchment of the Blue Nile pastoral communities in the southern Afar Region of Ethiopia. Basin, North Eastern Highlands of Ethiopia. Environmental African Journal of Range & Forage Science, 35, 1: 33–43. Systems Research, 8, 1: 1–30. Melaku N.D., Renschler C.S., Holzmann H., Strohmeier S., Zadbagher E., Becek K., Berberoglu S. 2018. Modelling land Bayu W., Zucca C., Ziadat F., Klik A. 2017. Prediction of use/land cover change using remote sensing and soil and water conservation structure impacts on runoff geographic information systems: a case study of the and erosion processes using SWAT model in the northern Seyhan Basin, Turkey. Journal of Environmental Monitoring Ethiopian highlands. Journal of Soils and Sediments, 18, 4: and Assessment, 190, 8: 1–15. 1743–1755. Zeleke G., Hurni H. 2001. Implications of Land Use and Land Michaels S., Goucher N.P., McCarthy D. 2006. Policy windows, Cover Dynamics for Mountain Resource Degradation in policy change, and organizational learning: watersheds the Northwestern Ethiopian Highlands. Mountain Research and Development, 21, 2: 184–191.

Journal

Environmental & Socio-economic Studies – de Gruyter

Published: Dec 1, 2022

Keywords: land use; land cover; watershed management activities; spatio-temporal variation; supervised classification; detection

There are no references for this article.