Land use and land cover change and system level analysis to guide sustainable intensification efforts in mixed crop- livestock farming system Getachew Tesfaye1, Degefie Tibebe1, Wuletawu Abera2, and Lulseged Tamene1 Author affiliation 1Alliance of Bioversity & CIAT, Addis Ababa, Ethiopia 2Alliance of Bioversity & CIAT, Accra, Ghana Published by Alliance of Bioversity International & International Center for Tropical Agriculture (CIAT) January, 2024 The Sustainable Intensification of Mixed Farming Systems Initiative aims to provide equitable, transformative pathways for improved livelihoods of actors in mixed farming systems through sustainable intensification within target agroecologies and socio-economic settings. Through action research and development partnerships, the Initiative will improve smallholder farmers' resilience to weather-induced shocks, provide a more stable income and significant benefits in welfare, and enhance social justice and inclusion for 13 million people by 2030. Activities will be implemented in six focus countries globally representing diverse mixed farming systems as follows: Ghana (cereal–root crop mixed), Ethiopia (highland mixed), Malawi: (maize mixed), Bangladesh (rice mixed), Nepal (highland mixed), and Lao People's Democratic Republic (upland intensive mixed/ highland extensive mixed). © 2023 This publication is licensed for use under the Creative Commons Attribution 4.0 International Licence - https://creativecommons.org/licenses/by/4.0. Unless otherwise noted, you are free to share (copy and redistribute the material in any medium or format), adapt (remix, transform, and build upon the material) for any purpose, even commercially, under the following conditions: ATTRIBUTION. The work must be attributed, but not in any way that suggests endorsement by the publisher or the author(s). https://www.cgiar.org/initiative/19-sustainable-intensification-of-mixed-farming-systems https://creativecommons.org/licenses/by/4.0 Contents Abbreviations and acronyms .................................................................................................... iv Abstract .............................................................................................................................................. v 1. Introduction ............................................................................................................................... 1 2. Materials and methods .......................................................................................................... 3 2.1. Descriptions of the study area ......................................................................................................... 3 2.2. Methodology.............................................................................................................................................. 4 3. Results .........................................................................................................................................7 3.1. Land use and land cover and trajectories ................................................................................ 7 3.2. LULC gain and loss over the period from 2013 to 2022 .................................................... 7 3.3. LULC change between 2013 to 2022 .......................................................................................... 10 4. Discussion ................................................................................................................................ 13 5. Implication of LULC change in the study area ............................................................ 15 6. Conclusion ................................................................................................................................ 18 References ....................................................................................................................................... 19 iv Abbreviations and acronyms ABC Alliance of Bioversity International and the International Center for Tropical Agriculture (CIAT) CSA Central Statistical Authority GEE Google Earth Engine GTPs Ground Truth Points LULC Land Use and Land Cover SIMFS Sustainable Intensification Mixed Farming System USGS United States Geological Survey v Abstract Changes in land use and land cover (LULC) are a major concern in Ethiopia. It has a significant impact on the environment, food and feed availability, and other ecosystem services and products for present and future generations. The effects of LULC change are particularly more pronounced in the highlands of Ethiopia, where the majority of the country’s cultivated land and livestock grazing occur, and competition between different land use decisions is a major concern. As a result, analysing spatial and temporal LULC changes and determining their primary causes is crucial for implementing effective land use planning, intervention targeting, and scaling out. The aim of this study was to analyse recent LULC changes in Basona Worena Woreda between 2013 and 2022 and identify the key drivers of change and their implications in the study area. The analysis was performed using Landsat 8 satellite imagery and random forest classification algorithms in the Google Earth Engine environment. Accordingly, agricultural land was identified as the dominant LULC type in the study area. It currently covers more than half of the total study area. Agricultural lands have expanded significantly, increasing from 42.3% in 2013 to 66.5% in 2022. Forest land and grassland, on the other hand, decreased from 30.3% and 21.2% in 2013 to 13.7% and 8.2% in 2022, respectively. Bareland coverage has expanded from 5% in 2013 to 9.3% in 2022. Thus, between 2013 and 2022, agricultural land and bareland have a net gain of 301.6 and 53.3 km2, respectively, whereas forestland and grassland have a net loss of 206.1 and 162 km2, respectively. The analysis has further indicated that a significant portion of the study areas (466 km2 or 37%) has experienced some form of LULC changes between the study periods. Between 2013 and 2022, 393.1 km2 of land from other LULC types were converted to agriculture, accounting for 31.5% of the total study area. The transition from forest and grassland to agricultural land constitutes the lion’s share of this, taking up 187.9 KM2 (48%) and 158 Km2(40%), respectively, of the total changes. A considerable portion of bareland has also been converted to agricultural land, accounting for 39.8 KM2 or 10% of the total changes. Agricultural land, grassland, and forest land (101 KM2 or 8% of the total study area) was turned into bareland. A total of 40 KM2 of forestland was also converted to grassland. In contrast, a considerable amount of land, primarily from grasslands and agricultural lands, was converted to forestlands. Rapid population growth, land scarcity, low productivity, poverty, poor agricultural and grassland management practice, and other factors have been recognized as important drivers of LULC change in Ethiopia’s highlands. With the business-as- usual scenario, the expansion of agricultural land and land degradation concerns caused by the conversion of LULC types will likely worsen. Thus, expansion of agricultural land and land degradation are the most critical and ongoing concerns of the study area that must be addressed if future land use systems are to continue vi delivering the necessary food, feed, and other ecosystem services sustainably. Thus, system level sustainable interventions are suggested as a way maintaining the ecosystem balance and halt the expansion of agriculture and land degradation in the study area. 1 1. Introduction In recent times, human-caused land use and land cover (LULC) changes are increasingly impacting our planet and impeding the sustainability of obtaining products and ecosystem services from the environment (Rindfuss et al., 2004; Verburg and Neumann, 2011). Many of the global forests and grasslands and other types of vegetation are being converted to agricultural land and other LULC types because of the rapid growth of the population, deforestation, land degradation, overuse of grazing land, etc. (Geist & Lambin, 2002). Such destructive and massive LULC changes have a significant impact on the environment, availability of food, feed, and other ecosystem services and products for the present and future generations (Radwan et al., 2021). In Ethiopia, LULC change is a crucial issue, and the country has experienced a substantial shift in LULC during the last century (Binyam, 2015). The LULC change and its adverse effects are particularly apparent in Ethiopia's highlands (> 1500 m a.s.l), where approximately 95% and 65% of the country's cultivated land and livestock grazing, respectively, happen (Anon, 1988; Ermia, 1986). Because of the mixed crop-livestock farming systems commonly practiced, competition between different land use decisions is a major concern in Ethiopia's highlands. For example, the majority of scholars (e.g., Tefera, 2011; Tesfaye et al., 2014; Meles et al., 2008; Mekuria et al., 2018; Anteneh, 2022; Wogderes, 2014) argue that agricultural land expansion is increased from time to time at the expense of natural vegetation and grazing land, whereas a few others (e.g., Eleni et al. 2013; Worku et al. 2021) are claimed that agricultural lands have been converted to woodlots. While others (e.g., Gebrekidan, & Tibebe, 2012; Gashaw & Dinkayoh, 2015) are also arguing that forests, woodlands, and shrublands are converting to agricultural and grazing lands in the highlands of Ethiopia. Thus, analysing spatial and temporal LULC changes and identifying their primary causes is critical for implementing effective land use planning and intervention targeting and scaling out in mixed farming system. As a result, numerous scholars have assessed historical LULC changes and their causes by combining remote sensing satellite datasets (mostly landsat imageries available for free since 1973) and other empirical findings (e.g., Bewket and Sterk, 2005; Tegene, 2002; Hurni et al., 2005; Kindu et al., 2015; Temesgen et al., 2013; Gebresamuel and Singh, 2010) in the highlands of Ethiopia. The majority of these research have found that the trend of LULC change is concerning and that the conversion of one LULC type to the other is complex and driven by many interconnected factors (Bewket, 2002; Kindu et al., 2015). Despite the fact that it varies from time to time and location to location, many authors have boldly stated that majority of the LULC changes in the highlands of Ethiopia were primarily caused by anthropogenic 2 activities such as rapid population growth, agricultural land expansion, deforestation, overgrazing, and improper land use and management, which resulted in accelerated soil erosion and degradation (Tesfaye et al., 2014; Eleni et al., 2013; Tegene, 2002; Bewket, 2002; Tefera, 2011; Gebrehiwot et al. 2021; Hssein et al., 2021). Although previous studies have produced crucial information that can be utilized to guide policy- and decision makers, and alleviate or prevent further changes in natural resources, a number of them have investigated historical LULC changes over a relatively wider temporal period. However, in addition to the historical changes in LULC change, it is critical to capture the current LULC changes and trajectories, as well as their links with the drivers, in order to set alternatives for sustainable land management and design proper land use systems. Examining recent changes in LULC and how they evolve could provide useful information of current LULC conditions and may aid in predicting what the future LULC will be if we continue with business-as-usual scenario. This could help in providing evidence to decision- makers and targeting interventions in light of future changes. The aim of this research is to analyse recent land use and land cover and to comprehend both the magnitude and spatial changes over time. We also aimed to identify the major drivers of change and their implications in the study area and propose interventions under the theme of sustainable intensification that could help to address many of the current and future evolving economic, social, and environmental problems associated with changes in land use and land cover. This study was carried out in Basona Worena Woreda, one of the Sustainable Intensification Mixed Framing System (SI-MFS) sites in Ethiopia. 3 2. Materials and methods 2.1. Descriptions of the study area Basona Werena Woreda is found in the North Shewa Zone of the Amhara region in Ethiopia and geographically located between the latitude of 90 38’ - 9o 41’00’’ N and longitude of 39 0 30’ 00’’-390 32’00’’E (Figure 1). It covers a total area of 1215.29- kilometre squares. The topography of Basona Woreda is largely mountainous with escarpments covered predominantly with reddish-brown soil. Figure 1 Location map Basona Worena Woreda of North Shewa Zone of Amhara region, Ethiopia. Based on figures reported by the Central Statistical Agency (CSA, 2007), the woreda has a total population of 120, 930 (CSA, 2007). The livelihood of the majority of the population is dependent on agriculture and agricultural-related activities. The long- term mean rainfall analysis has revealed that Basona woreda receives rainfall ranging from 890 to 1310.38 mm. More than 70% of the annual rainfall in the study woredas are received during the main rainfall season (referred to as Kiremt) that occurs between June to September. Thus, most of the crop cultivation in the study site takes place during this season. The woreda incorporates kola, Woyne Dega and Dega agro-ecological zones. Mixed crop-livestock is the dominant farming system in 4 the study area where farmers cultivate a variety of cereal such as Barley (Hordeum vulgare), Wheat (Tiriticum aestivum), and Teff (Eragrostis tef ZUCC.) and pulse crops such as Faba Bean (Vicia faba L.) and Field Pea (Pisum sativum L.) in the woreda. Different types of forest are found in the woreda including Tasmanian blue gum (Eucalyptus globulus), River red gum (Eucalyptus camaldulensis), Mexican cypress (Cupressus lusitanica), Tree health (Erica arborea), Lucky-bean tree (Erythrina abyssinica), wild olive (Olea europaea subs. cuspidata), African pencil-cedar (Juniperus procera), African redwood (Hagenia abyssinica), Cordia Africana (c. abyssinica), Strangler Fig (Ficus thonningii), Umbrella thorn (Acacia abyssinica), Ethiopian rose (Rosa abyssinica), and Woolly caper bush (Capparis tomentosa), Cape ash (Ekebergia capensis), African Wild Date (Phoenix reclinate), and Outeniqua yellowwood (Podocarpus falcatus) (Demalo,2014;Abayneh,2021; Kuria et al.,2014). 2.2. Methodology To quantify the extent and patterns of recent LULC changes in the study area, USGS Landsat 8 collection 2 Tier was utilized, which contains a collection of landsat scenes with the highest data quality. It is provided with a spatial and temporal resolution of 30m and 16 days and has eleven spectral bands. Landsat 8 datasets are available since 2013 and was accessed through the Google Earth Engine (GEE) environment. We applied a simple composite algorithm method provided by GEE to create a cloud-free landsat composites. Under the GEE environment, a Java based LULC classification script was developed. A pixel- based supervised classification system using random-forest classification algorithm was adopted to classify the LULC types of the study area. A total of 638 ground truthing points (GTPs) were collected from very high-resolution images available in Google Earth Pro for the period of 2021 and were used to train and validate the classification algorithm. The classification model was trained with 80% of the GTPS and validated with the remaining 20%. Six land use and land cover types were defined in the study area based on the knowledge of the study area and visual interpretation of satellite imagery, including Agricultural land, Bareland, Forestland, Settlement, Grassland, and Water body. A GTPs were collected for each of the six LULC classes. Table 1 contains descriptions of each land use and land cover type. Overall accuracy and kappa coefficient statistics were computed to assess the accuracy of the classified map (Congalton & Green, 2009). To determine the LULC trajectories, a LULC map of the study area for each year from 2013 to 2022 were produced. Then it was estimated by calculating the cumulative net changes in the total area of each land use and land cover types at each year to the initial period of 2013. 5 Figure 2 Schematic diagrams show the flow of land use and land cover change analysis in the study area. Table 1 Land use land cover class and its descriptions Land use and land cover class Descriptions Agricultural land An area of land ploughed for growing rain feed or irrigated crops. This category includes areas currently under crops, fallow, and land under preparation. Characterized by no trees if present extremely scattered. Bareland Areas under degraded grasses and shrubs. It also includes areas without any vegetation coverage Forest land An area of land comprised of forests, woodlands and it includes both the natural as well as plantation tree species. Grassland An area of land covered by grasses, bushes, and much closed and scattered shrubs water body An area of land covered by surface water Settlement Any built up areas in the study area. It includes urban centres, traditional huts, churches, health service offices etc 6 The gross gain, loss, and net changes as well as the LULC change transition were calculated for the period between 2013 and 2022. The LULC change driving factors were gathered from earlier similar studies in the Ethiopian highlands considering that they have comparable mixed crop-livestock farming system that significantly determine the land use decisions. Accordingly, a brief explanation for the implications of the recent LULC change observed in the study area, as well as potential responses or sustainable intensification interventions to mitigate the present and future impacts of LULC, has been provided. Figure 2 provides the schematic diagram of the flow of methodologies used in the study. 7 3. Results 3.1. Land use and land cover and trajectories In this study, an overall classification accuracy and kappa coefficient of 80% and 0.7122, respectively, was achieved. The estimated LULC classification accuracy is within the acceptable limit (Congalton, 1991). As can be seen in Table 2, agricultural land was the dominant LULC type in the study area, followed by forestland and grasslands. However, from 2018 to 2021, bareland overtakes forestland and grasslands as the second most dominant LULC type, only to be supressed by forestland again in 2022. Except for 2013 (it is estimated about 42.3%), an agricultural land has covered more than half of the total study area during all time periods considered in the study, with maximum coverage of 67.9% in the period of 2018. In comparison to the initial year in 2013, agricultural land and bareland in general increased, while forest and grasslands declined. This is nicely depicted in Figure 3, which shows the cumulative net change in total area of each LULC type in the study area between 2013 and 2022. Table 2 Area coverage of each LULC types from 2013 to 2022 (KM2) in Basona Worena Woreda. Superscripts indicate percentage of each LULC types in each year. LULC types 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Agriculture 527.542.3 660.453 846.267.9 691.4 55.4 829.3 66.5 847.0 67.9 766.161.4 839.3 67.3 775.3 62.2 829.166.5 Bareland 62.25 73.15.9 71.95.8 65.75.3 85.66.9 164.113.2 177.814.3 182.714.7 192.515.4 115.69.3 Forestland 377.530.3 300.224.1 193.815.5 281.022.5 201.316.1 143.311.5 149.512 104.48.4 117.49.4 171.413.7 Settlement 2.50.2 3.00.2 7.30.6 3.50.3 10.00.8 40.63.3 109.28.8 58.64.7 114.39.2 16.11.3 Grassland 264.421.2 192.715.5 116.69.3 193.015.5 107.98.6 19.81.6 9.20.7 24.72 12.21 102.28.2 Water 13.01 17.71.4 11.40.9 12.61 13.11.1 32.32.6 35.22.8 37.43 35.42.8 12.71 Overall Classification Accuracy=80%; Overall Kappa Statistics=0.7122 using Landsat 8 acquired in 2021 The trajectories of agricultural land showed an irregular pattern, as seen in Figure 3. The forest and grassland, on the other hand, had either minimal changes or stable at the beginning and then were drastically reduced after 2016 and exhibit the behaver of retaining back at the year of 2022. Similarly, bareland and settlements were relatively stable until 2017, when they began to rise steadily. The water class, on the other hand, remains reasonably consistent during the study period. 3.2. LULC gain and loss over the period from 2013 to 2022 Figures 4 and 5 depicted the LULC maps of the study area for 2013 and 2022, respectively. According to the map from 2013 (Figure 4), the study area was dominated by agricultural land and was primarily located in the central and western parts of the study area. The eastern and southern parts of the study area were occupied by forest and grasslands. 8 Figure 3 Time-series of the cumulative net change in total area of each LULC type in the study area between 2013 and 2022. Figure 4 Land use land cover map of the study area in 2013 This appears to follow the topographic patters of the study area, with vegetation (including forest and grasslands) covering the higher elevations and agricultural land primarily covering the lower elevations. However, as shown by the 2022 LULC map in Figure 5, this tendency is no longer observed, and agricultural lands are 9 expanding to high elevation areas of the study site as well. During the analysis period, it has been observed that the expansion of agricultural lands was significant, and it increases from 42.3% in 2013 to 66.5% in 2022 (Table 2). However, forest land and grassland exhibited a negative change and it decreased from 30.3% and 21.2% in 2013 to 13.7% and 8.2% in 2022, respectively (Table 2 and Figures 4 and 5). On the other hand, bareland coverage has expanded from 5% in 2013 to 9.3% in 2022. Although very small, settlements have also showed a positive change (0.2% in 2013 to 1.3% in 2022) (Table 2 and Figures4 and 5). While water LULC class remains constant over the two periods. Figure 5 Land use land cover map of the study area in 2022 These changes are also presented in Figure 6, which provides the total gross gain and loss of each LULC type, as well as their net changes between 2013 and 2022. Apart from the water class, all of the other LULC types showed a substantial gross gain or loss, indicating that many locations in the study area have undergone either an expansion or reduction of these LULC types. According to the same figure, agricultural land and bareland have a net gain of 301.6 and 53.3 km2, respectively. While the net loss of forestland and grassland is 206.1 and 162 km2, respectively. 10 Figure 6 Total area of gains and losses of the different LULC types in the study area between 2013 and 2022 3.3. LULC change between 2013 to 2022 The LULC change map in Figure 7 provides the major LULC changes in the study area between 2013 to 2022. While Figure 8 summarises the transitions that occurred between the five main LULC types. According to the analysis, the total area of land that converted to and persisted between 2013 and 2022 was 781 km2 (63%) and 466 km2 (37%), respectively. This could imply that a significant portion of the study areas has experienced some form of LULC changes between the study periods. The dominant LULC change, as seen in Figures 7 and 8, was changes to agricultural land, followed by changes to grassland and bareland. Between 2013 and 2022, 393.1 km2 of land from other LULC types were converted to agriculture, accounting for 31.5% of the total study area. The transition from forest and grassland to agricultural land constitutes the lion’s share of this, taking up 187.9 km2 (48%) and 158 km2(40%), respectively, of the total changes. A considerable portion of bareland LULC type has also been converted to agricultural land, accounting for 39.8 km2 or 10% of the total changes. Agricultural land, grassland, and forest land (101 km2 or 8% of the total study area) was turned into bareland. Indeed, the conversion of agricultural land to bareland is large and contribute more than half of this changes. 11 Figure 7 Major land use land cover changes captured in the study area for the period from 2013 to 2022 Figure 8 Schematic representation of LULC transition between 2013 and 2022. The transitions provided in Km2, and the actual values are given in the square. As minor transition was estimated, the water class was note not included. 12 During the study period, a large amount of LULC types primarily from grasslands (26.8 km2) and agricultural lands (16.6 km2), were converted to forestlands. Similarly, about 55 km2 of grasslands were established at the expense of other LULC types. Whereas conversion of forestland to grassland has contributed the most (40 km2). 13 4. Discussion The land use and land cover change in mixed crop-livestock farming system has a direct impact on the availability of food and feed, as well as other products and services to humans and livestock. Land use change, both the spatial and temporal changes, is important for determining the land use and land cover patterns (Agidew and Singh, 2017). In this study, a recent LULC change analysis for the period between 2013 and 2022 was performed using Landsat 8 satellite imageries and random forest classification algorithms in the Google Earth engine environment. Although some irregular patterns in the trends of the LULC types in the study area, agricultural land and bareland have shown an increasing trend while forestland and grassland have decreasing over the study period. Some of the irregularity in the patterns of the LULC types across the study period could be explained by the different interventions practiced in the study area at different periods. Overall, agricultural land was found to be the most dominant LULC type of the study area (covering 66% of the total study area in 2022), followed by forestland. It is not surprising to anticipate such a coverage for agricultural land in Basona Worena woreda, given that the bulk of the people are farmers and rely significantly on farming and farm products for living. The analysis revealed that the majority of the LULC type conversion is multidirectional, despite the fact that the area of transition from one to the other LULC types differ greatly due to the different land use decisions made by various actors. Overall, a significant portion of the study area (65%) has been experiencing different forms of LULC changes over the study period. As a result, changes from other LULC types to agricultural land being the most prominent changes identified in the study area. A comparable finding has been reported by Agidew and Singh (2017), Gashaw et al. (2014) and Bewket (2002) in the highlands of Ethiopia. Agricultural land was expanded mostly at the expense of forest land, grassland, and bareland. The transition from forestland and grassland to agriculture was particularly significant, resulting in a net loss of forest and grassland from 2013 to 2022. This expansion of agricultural land in the study area might be related to population growth and low agricultural productivity, which leads to the requirement for additional land to meet farmer’s food and feed demands (Agidew and Singh, 2017; Gebrehiwot et al., 2021). A similar justification for agricultural land expansion in the Andit Tid watershed in the North Shewa zone has been provided by Gessesse et al. (2017). On the other hand, Shiferaw (2011) and Gebrehiwot et al. (2021) argued that limited off-farm employment opportunity in Ethiopia’s highlands have led to farmers, rural youths, and other landless people to widely engaged in the clearing of forest land and further expanding agricultural into sloppy areas, grasslands, bareland, and other LULC types. Agricultural land, on the other hand, has been converted into forest and grasslands. This could be because of the massive 14 land restoration and afforestation program implemented by the government and other development partners at various times, in which they plant different tree seedlings and fodder grasses/shrubs over degraded agricultural and pasturelands (Woldemariam et al., 2021).The conversion of agricultural land to forest land could also be explained by the fact that nowadays framers are converting their farmlands to plantations in order to get cash, more firewood’s, shade and year round fodder for their livestock’s (Eleni et al. 2013; Worku et al. 2021). Bareland, conversely, has shown a considerable net gain from 2013 to 2022, and a significant area of agriculture, grassland, and forest land has been converted to bareland. This could be associated to sever soil erosion in agricultural lands in the study area, overgrazing of grasslands, and deforestation for firewood and timber production needs. The conversion of agricultural land to bareland was particularly significant, implying that a significant fractions of agricultural lands in the study area are experiencing widespread soil erosion and degradation, rendering the land less productive and unsuitable for cultivation. This could be due to the fact that the majority of agricultural lands in the study area are located on steep slopes, and soil and water conservation practice is relatively limited, and soil erosion is a key problem that rendered agricultural fields lees productive (Tesfaye et al., 2023). In general, the conversion of other LULC types (particularly agricultural land, forest land, and grassland) to bareland may indicate the presence of critical land degradation problem in the study area, which may have a negative impact on food security and socio-economic stability. Conversely, a considerable amount of forest land has been converted to grasslands, which could be ascribed to the clearance of forest lands for the purpose of grazing areas for their livestock. A similar finding in the highlands of Ethiopia has been reported (e.g., Gebrekidan, & Tibebe, 2012; Gashaw & Dinkayoh, 2015). As a mixed crop-livestock farming system common in the study area, farmers are rearing various sizes of livestock’s and they are always in need of feed for their livestock’s, and they could convert forestlands to grassland to satisfy these demands. 15 5. Implication of LULC change in the study area Land use and land cover change has an important effect on the supply of food for the growing population and its adverse impact could compromise the sustainability and conservation of key ecosystem functions. Thus, identifying and studying the key drivers of land conversion and its implication has a paramount importance to prevent or reduce the present and future impacts of LULC change. An appropriate response to LULC conversion can be achieved by attempting to control the impacts of the drivers or by devising solutions for adapting agricultural ecosystem services to the consequences of the changes (Lambin et al., 1999). There are numerous approaches to developing solutions to the consequences of change drivers. These are accomplished by using innovative and cost-effective technologies, strategies, and institutional structures that encourage intensification, diversification, and regulation (Lambin et al., 1999). Based on earlier LULC change study conducted in Ethiopia’s highlands, the primary driving factors identified by Gebrehiwot et al. (2021) and Hssein et al. (2021) were considered to have caused and accelerated LULC change in the stud area. Figure 9 depicts the relationship between LULC change and its key drivers as well as their implications, in the study area. Figure 9 The relationship between land use and land cover changes as well as its implications and potential drivers in the study area. The conversion of forest land, grassland, and bareland to agricultural land was found to be the major changes in the study area between 2013 to 2022 (Figures 7 & 8). This implies that agricultural lands are expanding at the expenses of forestland, grassland, and bareland because of many interlinked factors including the rapid population growth, low productivity of lands, shortage of land, etc (Gebrehiwot et al., 2021). Farmers need additional land to produce more and meet their demands. Expansion of agricultural land, especially on steep slopes and bareland, may exacerbate soil erosion and degradation (Tegene, 2022; Amede et al., 2001). The 16 expansion of agricultural lands over grasslands may also result in a lack of feed for livestock, which may reduce livestock productivity and availability of draft animals for land preparation. According to Mekuria et al. (2018), the intensive conversion of grassland to cropland and plantations has been one of the key reasons of feed shortage in the highland mixed farming system. The LULC change analysis, on the other hand, found that there is an extensive LULC conversion from agricultural land, forest land, and grassland to bareland. A considerable amount of the forest land has also been converted to grasslands. These conversions to bareland and the conversion of forest land to grassland are derived by the major causes of land degradation in Ethiopian highlands which includes deforestation & forest degradation in the forest lands, overgrazing in the grasslands, soil erosion and lack of proper soil conservation measures in the agricultural lands, among other things (Hssein et al. 2021). Rapid conversion of forest and grassland to bareland could expose fertile soil to massive soil erosion and degradation (Tegene 2002; Maitima et al. 2009). Thus, expansion of agricultural land and land degradation are the most critical and ongoing concerns that must be addressed if future land use systems are to continue delivering the necessary food, feed, and other ecosystem services in a sustainable manner. Therefore, limiting agricultural land expansion and reverting and reducing land degradation could be key interventions to alleviate the adverse effects of LULC conversion in the study area. With the business-as-usual scenario, the expansion of agricultural land and land degradation concerns caused by the conversion of LULC types will likely worsen. In Figure 10, we proposed a part of the solution under the theme of sustainable intensification that might be applied on the ground to reverse the current LULC change trend, mitigate the future risks, and assure sustainable development and food security in the study area. We cannot address the risk of LULC conversion through providing a few simple solutions implemented by piecemeal since the major driving factors are numerous, complex, and interconnected. As a result, our solution should be as many as the driving factors, well-integrated, and require the involvement of various actors and stakeholders. The proposed intervention diagram in Figure 10 tries to show the different components of the solutions, and the components are proposed based on their direct contribution towards addressing the problem as detected by the LULC change in the study area. Agricultural expansion in the study area might be hindered by ensuring or enhancing crop and livestock productivity, reducing population growth, and engaging rural youths, women, and other landless individuals in off-farm activities. Population pressure is one of the key driving factors of LULC change in the highlands of Ethiopia (Gebrehiwot et al., 2021). Given that the bulk of the Ethiopian population is young (Megquier and Belohlav, 2014) (and we expect a similar situation in the study area) we can anticipate increasing pressure on various LULC types. 17 Figure 10 The flows and linkages of different SI interventions to tackle the negative consequences of land use land cover changes in the mixed crop-livestock farming systems. Thus, a proactive measure to population management, such as awareness creation and family planning programs, is necessary. There should be a strategy in place to widen off-farm employment opportunities to engage rural youths, women, and other landless people. Increasing crop and livestock productivity could play an important role in reducing the expansion of agricultural lands. Farmers, however, requires improved technologies, agricultural inputs, and extension services to ensure productivity. Since agricultural land is expanding at the expenses of grasslands in the study area, the supply of feed for livestock could be a critical problem and so an alternative option, including dry season fodder production, should be established to ensure livestock productivity in the study area. Land degradation, on the other hand, could be reversed and reduced by lowering pressures on the land resources, such as reducing or avoiding deforestation via introducing sustainable use and management of forests as well as law enforcements, reducing soil erosion, introducing controlled and proper use of grasslands (I, e., rotational grazing, managing proper size of livestock’s, reduce livestock trampling etc), and afforestation and reforestation of degraded lands. 18 6. Conclusion In this study, a recent LULC change analysis and its implications for the period between 2013 and 2022 were performed. The results showed that agricultural land and bareland increased while forestland and grassland have decreasing over the study period. During the study period, a significant portion of the study area (65%) has been experiencing different forms of LULC changes. The most noticeable changes detected in the study area were transitions from other LULC types to agricultural land, followed by transitions to bareland. Expansion of agricultural land and land degradation are the most critical and ongoing concerns that must be addressed if future land use systems are to continue providing the necessary food, feed, and other ecosystem services in a sustainable manner. According to previous similar studies in Ethiopia’s highland, the majority of these changes are mostly caused by rapid population growth, low agricultural productivity, shortage of land, limited off-farm employment opportunity, deforestation, poor management of farmlands and grazing areas etc. The expansion of agricultural land and land degradation concerns induced by the conversion of LULC types are likely worsen under the business-as-usual scenario. Hence, appropriate, and well-integrated strategies that limit agricultural land expansion and reverting and reducing land degradation could be key interventions to alleviate the adverse effects of LULC conversion in the study area. 19 References • Abayneh, I. (2021). Ethnobotanical study of medicinal plants used in indigenous people in Lemo woreda, Hadiya zone, SNNPR, Ethiopia. Addis Ababa University (Msc Thesis). • Agidew, A.A., & singh, K.N. (2017). The implications of land use and land cover changes for rural household food insecurity in the Northeastern highlands of Ethiopia: the case of the Teleyayen sub-watershed. Agric & Food Secur 6:56. • Anon. (1988). National Atlas of Ethiopia, 76 pp. Ethiopian Mapping Authority, Addis Ababa • Anteneh, M. (2022). Analysis of Land Use/Land Cover Change and Its Implication on Natural Resources of the Dedo Watershed, Southwest Ethiopia. Scientific World Journal. 2022: 6471291. • Bewket, W. (2002). Land cover dynamics since the 1950s in Chemoga watershed, Blue Nile Basin, Ethiopia. Mountain Research and Development, 22, 263–269. • Bewket, W. (2002). Land cover dynamics since the 1950s in Chemoga watershed, Blue Nile Basin, Ethiopia. Mountain Research and Development, 22, 263–269. • Bewket, W., Sterk, G. (2005). Dynamics in land cover and its effect on stream flow in the Chemoga watershed, Blue Nile basin, Ethiopia. Hydrological Processes 19(2), 445-458. • Binyam, A. (2015). The Effect of Land Use Land Cover Change on Land Degradation in the Highlands of Ethiopia. Journal of Environment and Earth Science, Vol.5, No.1, 2224-3216. • Binyam, A., Efrem, G., Zewudu, E., Habtemariam, K. (2015) Land use and land cover changes and associated driving forces in north-western lowlands of Ethiopia. Int Res J Agr Sci Soil Sci 5(1):28–44. • Central Statistical Agency (CSA). (2007). Compilation of Economic Statistics. Addis Ababa, Ethiopia • Congalton, R., & Green, K. (2009). Assessing the accuracy of remotely sensed data: Principles and practices (2nd ed.). Florida: Lewis Press210p • Congalton, R.G. (1991). A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens. Environ. 37, 35–46. • Demalo, F.T. (2014). Farmers’ perception on erosion problem and adoption of soil and water conservation structures: The case of Ojoje watershed Doyogena Woreda, Kambata Tambaro Zone, Ethiopia. Addis Ababa University (Msc Thesis). • Eleni, Y., Wolfgang, W., Michael, E., Dagnachew, L., Gunter, B. (2013). Identifying LULC dynamics in the Koga catchment, Ethiopia, from multi-scale data, and implications for environmental change. ISPRS Int J Geo-Inform 2(2):302–323 • Gashaw, T., Bantider, A., & Mahari, A. (2014). Population dynamicsand land use/land cover changes in Dera District, Ethiopia. Global Journal of Biology Agriculture & HealthSciences, 3, 137–140. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9514936/ 20 • Gebrehiwot, T., Teklewold, H., Devenish, A., Jellason, P. N. (2023). Unpicking the socioecological drivers and impacts of agricultural expansion in Ethiopia. Accessed at https://www.sentinel gcrf.org/sites/sentinel/files/resources/2021- 11/Ethiopia%20RS%20briefing_FINAL.pdf on 30 October 2023. • Gebresamuel, G., Singh, B.R., Dick, O. (2010). Land-use changes and their impacts on soil degradation and surface runoff of two catchments of Northern Ethiopia. Acta Agric. Scand, Section B. Plant Soil Sci. 60, 211-226. • Geist, H. J., & Lambin, E. F. (2002). Proximate causes and underlying driving forces of tropical deforestation. BioScience, 52(2), 143–150 • Gessesse, A.T., Chanie, T., Feyisa, T., Jemal, A. (2017). Impact Assessment Of Land Use /Land Cover Change On Soil Erosion And Rural Livelihood In Andit Tid Watershed, North Shewa,Ethiopia. Asia Pac. J. Energy Environ. Volume 4, No 2. • Hurni, H., Tato, K., Zeleke, G., (2005). The implications of changes in population, land use,and land management for surface runoff in the upper Nile basin area of Ethiopia. Mt. Res. Dev. 25, 147–154. • Hussein, A. (2021). Major Causes and Effects of Land Degradation and Rehabilitation Practices in Ethiopia: Systematic Review. J Earth Sci Clim Change, 12:11. • Kindu, M., Schneider, T., Teketay, D., Knoke, T. (2015). Drivers of land use/land cover changes in Munessa Shashemene landscape of the south-central highlands of Ethiopia. Environ. Monit. Assess. 187. • Kuria, A., Lamond, G., Pagella, T., Gebrekirstos, A., Hadgu, K., Sinclair., F. (2014). Local knowledge of farmers on opportunities and constraints to sustainable intensification of crop-livestock-trees mixed systems in Basona Woreda, Amhara Region, Ethiopian Highlands. International Livestock Research Institute. • Lambin, E.F., Baulies, X., Bockstael, N., Fischer, G., Krug, T., Leemans, R., Moran,E.F., Rindfuss, R.R., Sato, Y., Skole, D., Turner II, B.L., Vogel, C. (1999). Land- Use and Land-Cover Change (LUCC) Implementation Strategy. International Geosphere-Biosphere Programme (IGBP) and International Human Dimensions Programme (IHDP) on Global Environmental Change. IGBP report 48. IHDP report 10. • Megquier, S., Belohlav, K. (2014). ETHIOPIA’S KEY: YOUNG PEOPLE AND THE DEMOGRAPHIC DIVIDEND. Policy Brief, 2014. https://www.prb.org/wp- content/uploads/2015/01/ethiopia-demographic-dividend policybriefa.pdf accessed on 14/12/2023 • Mekuria, W., , Mekonnen, K., Thorne, P., Bezabih, M., Tamene, L., Wuletawu, A. (2018). Competition for land resources: driving forces and consequences in crop-livestock production systems of the Ethiopian highlands. Ecological Processes 7:30. • Meles, K., Epema, G.F., van Bruggen, A.H.C .(2008). Temporal and spatial changes in land use patterns and biodiversity in relation to farm productivity at multiple scales in Tigray, Ethiopia (PhD Thesis). Wageningen University, Wageningen, The Netherlands https://www.prb.org/wp-content/uploads/2015/01/ethiopia-demographic-dividend-policybriefa.pdf%20accessed%20on%2014/12/ https://www.prb.org/wp-content/uploads/2015/01/ethiopia-demographic-dividend-policybriefa.pdf%20accessed%20on%2014/12/ https://www.prb.org/wp-content/uploads/2015/01/ethiopia-demographic-dividend-policybriefa.pdf%20accessed%20on%2014/12/ 21 • Radwan, T.M., Blackburn, G.A., Whyatt, J. D., Peter M. Atkinson, P.M. (2021). Global land cover trajectories and transitions. Scientific Reports Nature, 11:12814. • Rindfuss, R. R., Walsh, S. J., Turner, B. L., Fox, J. & Mishra, V. (2004). Developing a science of land change: Challenges and methodological issues. Proc. Natl. Acad. Sci. U. S. A. 101, 13976–13981. • Tefera, M. M. (2011). Land-use/land-cover dynamics in Nonno District, Central Ethiopia. Journal of Sustainable Development in Africa, 13, 123–141. • Tegene, B. (2002). Land-cover/land-use changes in the derekolli catchment of the South Welo Zone of Amhara Region, Ethiopia. Eastern Africa Social Science Research Review,18, 1–20. • Temesgen, H., Nyssen, J., Zenebe, A., Haregeweyn, N., Kindu, M., Lemenih, M., Haile, M. (2013). Ecological succession and land use changes in a lake retreat area (Main Ethiopian Rift Valley). J. Arid Environ. 91, 53–60. • Tesfaye, G., Tibebe, D., Ebrahim, M., Tilaye, A., Abera, W., Tamene, L. (2023). Biophysical characterization: Sustainable Intensification Mixed Farming system (SI-MFS) sites in Ethiopia. Addis Ababa (Ethiopia) Bioversity International and International Center of Tropical Agriculture. 35 p. • Tesfaye, S., Guyassa, E., Joseph, R.A., Birhane, E., Taye, G. (2014). Land Use and Land Cover Change, and Woody Vegetation Diversity in Human Driven Landscape of Gilgel Tekeze Catchment, Northern Ethiopia. International Journal of Forestry Research vol. 2014, Article ID 614249, 10 pages. • Verburg, P. H., Neumann, K. & Nol, L. (2011). Challenges in using land use and land cover data for global change studies. Glob. Change Biol. 17, 974–989. • Woldemariam, T., Gronkiewicz, M., Will Anderson, W. (2023). How Local Data Is Accelerating Land Restoration in Ethiopia. Available at https://www.wri.org/insights/local-data-accelerates-land-restoration-ethiopia (accessed on 14/12/2023) • Worku, T., Mekonnen, M., Yitaferu, B., Artemi Cerdà, A. (2021). Conversion of crop land use to plantation land use, northwest Ethiopia. Trees, Forests and People, Vol.3. 100044. https://www.wri.org/profile/tesfay-woldemariam https://www.wri.org/profile/mary-gronkiewicz https://www.wri.org/profile/will-anderson https://www.wri.org/insights/local-data-accelerates-land-restoration-ethiopia https://www.sciencedirect.com/journal/trees-forests-and-people https://www.sciencedirect.com/journal/trees-forests-and-people cgiar.org/initiative/mixed-farming-systems https://www.cgiar.org/initiative/mixed-farming-systems/