Spatio-temporal Dynamics, Drivers, Effects and Environmental Implications of Land use/land Cover Changes in Ambo Town and Its Watershed, West Shoa Zone, Oromia Regional State, Ethiopia

Background: Land use/land cover change in urban watersheds of developing countries like Ethiopia is claimed to be a consequence of complex interaction of different actors, driving forces, and land itself. It is asserted to result in the degradation of natural vegetation and signicant increases in impervious surfaces. The purpose of the study was to analyze spatio-teporal changes in land use/ land cover in Huluka watershed where Ambo town is situated and examine their drivers and effects with environmental implications. Results: The overall increase of urban built-up area, cultivated land, and bare land use/land cover type with 351%, 105%, and 41.9% respectively between the year 1979 and 2017 implies the increase in ooding disaster risk in the watershed as such land use/land cover types exacerbate the run-off conditions in the watershed. Infrastructural expansion, agricultural expansion, increased demand for fuel wood and wood for construction, local environmental factors, local biophysical drivers, and local Social events were identied as proximate drivers of land use/land cover changes in the study area. Demographic factors, economic factors, technological factors, policy and institutional factors and cultural factors were conrmed as the underlying drivers of land use/land cover change in the watershed. Increased ooding risk, increased soil erosion; increased sedimentation into the lake (Dendi lake) and rivers (Huluka, Awaro, Debis, Boji, Bolo, Aleltu, Karkaro, and Korke), decrease in soil fertility resulting from ooding risk, and change in climatic parameters (decrease in annual rainfall and increase in heat during dry season) were claimed as the negative effects of land use/land cover change in the study area. Conclusions: Practice of appropriate land use planning and management in the watershed, appropriate environmental impact assessment (EIA), and proper planning and management of socio-cultural, economic, and environmental development are of paramount importance to promote sustainable development in the watershed. perceptions on the causes and negative effects of land cover change in their area. Qualitative methods of data collection (focus group discussions, key informant interview, and personal observation) were employed to collect data on local residents’ perceptions on proximate and underlying causes of land use/land cover change in their area. The qualitative analysis of drivers of land use/land cover changes is presented hereunder. Landsat TM ETM + imagery for the periods 1979, 1984, 2009, 2017 were used. The analysis of images involved the key steps of pre-processing; post-processing, overlaying and change detection and creation of maps of land use/ land cover change from Landsat TM and ETM + imagery for the period, 1979–2017.


Background
Land use/land cover change in developing countries like Ethiopia is claimed to be a consequence of complex interaction of different actors, driving forces, and the land itself (Zeleke,2000;Bewket,2003 Land conversion due to residential development, economic growth and transportation belongs to the most serious environmental pressures on landscapes worldwide, in particular in urbanized areas (Nuissl et al., 2009;Wheater & Evans, 2009;Adebayo et al., 2010;Santato et al., 2013). Individual ecosystem services that are affected by land use transition include the production of food, regulation of energy and matter ows, water supply, supply of recreational space, biodiversity or natural aesthetic values (Nuissl et al., 2009;Birkinshaw et al., 2010;Berggren et al., 2013;Santato et al., 2013). Unless policy strategies and planning instruments attempt to address the problem of land use transition as an ongoing process (rather than de ning xed land use patterns as their ultimate goal), they will neglect possible feedback dynamics and thus can hardly be successful (Ebi & Semenza, 2008;Nuissl et  Understanding the dynamics of land use/land cover changes with their drivers and effects contributes immensely for promoting sustainable urban and rural development in the watershed and its basin. Hence, this study analyzed change in land use/ land cover in Huluka watershed where Ambo town is situated and explore local perceptions on its causes and effects. To adress these objectives, four research questions were answered: 1) What is the composition of the land use/land cover types in Ambo town's watershed for the study period? 2)What are the proportions of these land use/land cover over the study period? 3) What is the rate of change for land use/land cover in Ambo town's watershed for the study period? 4)What are the drivers and effects of land use/land cover changes in the study area? 2. Methodology

Description of the Study Area
Pellic Vertisos were identi ed as the types of soils in the watershed. The rainfall of the area is con rmed to be bimodal, with unpredictable short rains from March to April and the main season ranging over June to September. The highest mean total annual rainfall of the watershed over 32 years  was a rmed to be 1181 mm while the lowest was 1036 mm (Fig. 2).

Data Types and sources of Data
Four period of satellite images were used to conduct this study. To this end, Landsat TM and ETM + imagery for the periods 1979, 1984, 2009, and 2017 were used. These years were chosen based on the purpose of the study and the purpose of the study was to understand the changes in land use/land cover and local residents' perceptions on its cause and negative effects in the watershed. 1979, 1984, and 2009 were chosen to see the dynamics of land use/land cover change over four decades. To  Digital map on shape le with the scale of 1:50,000 from Ethiopian Mapping Authority was used as supporting spatial data for delineating the boundary of the study watershed. Global positioning system (GPS) points collected during eld observation were used to collect GCP (ground control point) to successfully undertake the image classi cation. To this end, 300 sample-training sites were used in each year from ancillary data like high-resolution Google Imagery while 300 sample training sites were used from eld observation for the year 2017. Other sources of data included: Central Statistical Authority (CSA), Ethiopian Mapping Agency (EMA), Landsat website of www.glovis.USGS.gov, urban and rural communities in Huluka watershed, urban planners of Ambo town, and land use planners in Huluka watershed for secondary data types and perception related primary data types.

Methods of Data Collection
To collect relevant data to analyze the dynamics of land use/land cover change in the watershed for the periods considered, online Satellite Imagery (Monkkonen, 2008; Gondo & Zibabgwe, 2010) was employed. In addition, personal eld observation, focus group discussion, and key informant interviews were employed to get additional primary data. Accordingly, six focus group discussions, and fteen key informant interviews were undertaken to collect qualitative data from the local communities in the watershed based on the purpose of the study.

Methods of Data analysis
Based on prior knowledge of the study area, data collected from the local communities in the watershed, characteristics of Landsat images, ancillary data like Google Earth and eld observation, seven land use/land cover classes (Table 1) were used for image classi cation and land use/land cover change analysis. Arc GIS 10.1 software and ERDAS IMAGINE 9.1 software (Huang et al., 2007;Monkkonen, 2008; Gondo & Zibabgwe, 2010) were employed for the intended image classi cation, land use/land cover change detection and mapping. While ERDAS IMAGINE 9.1 software was employed to classify images and detect change over time, ArcGIS10.1 software was employed for geospatial analysis of the classi ed images and developing maps of land-use/land cover change.

Cultivated land
Areas used for rain fed crop production and scattered rural settlements.
urban Builtup area Areas occupied by urban residential houses, buildings and industrial uses.

Bush/shrub land
Land with shrubs and bushes, scattered small trees mixed with grasses.
Bare land Areas with no or very little vegetation cover and characterized with shallow and rocky surface along the ooding area of the local stream valleys, over gentle and steep mountain slopes.

Grassland
Land predominately covered with grasses, forbs, grassy areas used for communal grazing.

Water
Areas covered by Lake in the catchment permanently The analysis of images involved the key steps of pre-processing; post-processing, overlaying and change detection and creation of maps of land use/ land cover change from Landsat TM imagery for the period,

1979-2017.
The Pre-processing step involved geo-referencing the landsat images, radiometric correction, layer stacking, resolution merge, image enhancement, and adding vector information from administrative boundary of the study area. The post-processing step involved image classi cation which was undertaken using hybrid classi cation methods involving both unsupervised and supervised techniques among different classi cation algorithms. To this end, maximum likelihood was used for supervised classi cation by taking ground control points for seven major land use land cover classes. These land use/land cover (LULC) types were identi ed with the help of visual interpretation elements and the different re ection characteristics of the feature in the satellite images of 1979, 1984, 2009 and 2017. In other words, the supervised classi cation involved selecting pixels that represents land cover classes that were recognized by the researcher. Accuracy assessment was also undertaken in the post-processing step. Accordingly, accuracy assessment was carried out to verify to what extent the produced classi cation is compatible with what actually exists on the ground (Anderson et al., 1976;Congalton, 1991). All the output maps for the study period (1979,1984,2009, and 2017) ful lled the required standard (which is a minimum of 85% accuracy).
The overlaying and change detection step involved: interpretation of change in land use and land cover change in various years and analysis of its implication for environmental and management issues like ooding risk in Ambo town and its watershed. In other words, post-classi cation method was employed for the change detection. This technique helps to generate change matrix where different transfers from one land use/ land cover types to another can be visually observed. Accordingly, change metrics for detecting land use and land cover change were constructed between 1979 and 1984, between 1984 and 2009, between 2009 and 2017, and between 1979 and 2017 through pixel-to-pixel comparisons. The classi ed data were then taken in to ArcGIS 9.1 environment to calculate area for each land use/ land cover type and produce the land use/land cover maps of the area. These land cover maps were compared pixel by pixel with the nal results showing both change-no-change information as well as 'from to' land cover change information. The land use and land cover change detection was made using the following formula (Fura, 2013): Where: Δ = land use and land cover change (%), A1 = amount of land use/ land cover type in year 1, A2 = amount of land use and land cover type in year 2.
The nal step was creation of maps of land use/land cover change. Accordingly, the analysis of images ended with the creation of maps of land-use and land cover change from Landsat TM imagery for the period, 1979-2017.
To examine the drivers and effects of land use/land cover changes, qualitative data collected through document review, key informant interview, focus group discussion, and personal observation were analyzed with the help of thematic analysis, content analysis and narrations. The major criteria underpinning the combined application of the aforementioned qualitative methods of data analysis were: transparency, maximizing validity, maximizing reliability, comparative analysis, and re exive approach in the process of analysis. The major steps involved in the process included: 1. Reading and annotating transcripts; 2. Identify themes; Developing a coding scheme; and 4. Coding the data.

Results
This section presents results on the composition of land use/land cover in Huluka Watershed, analysis of land use change dynamics, and local residents' perceptions on the causes and negative effects of land use/land cover change in their area.

Composition of Land Use/Land Cover in Huluka
Watershed for the Study Period Forest area, cultivated land area, urban built-up area, bush/shrub land area, bare land area, grassland area, and water area were identi ed as the seven types of land use/land cover in the watershed.
The overall decline of bush and shrub land, forest, grassland, and water land use/land cover type with 68.2%, 59.3%, 32.7%, and 5.1% respectively between 1979 and 2017 is bad news for the watershed as the potential of the watershed to contribute in mitigating ooding disaster risk decline with the decline of these land use/land cover types. Moreover, the overall increase of urban built-up area, cultivated land, and bare land use/land cover type with 351%, 105%, and 41.9% respectively between the year 1979 and 2017 implies the increase in ooding disaster risk in the watershed as such land use/land cover types exacerbate the run-off conditions in the watershed. Discussants of the focus group discussion were asked to opine on the causal linkage between infrastructure expansion and land use/ land cover change in Huluka watershed. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). Accordingly, infrastructural expansion was perceived as one of the proximate causes of land use/land cover change in Huluka watershed. It was also possible to observe current expansion of infrastructures like construction of houses for Ambo University expansion ( Figure 5). One of the key informants said "Infrastructural development is relevant for us as we need roads, telecommunication, and electri cation. However, any infrastructural development in this watershed must not result in degradation of land resources in this watershed."(Key informant one, Huluka Watershed). Moreover, discussants of the focus group discussion appreciated the expansion of infrastructure in Huluka watershed. However, they suggested that expansion of any infrastructure in Huluka Watershed should be preceded by appropriate environmental impact assessments to mitigate the negative impacts of infrastructural expansion. In other words, failure to institutionalize environmental impact assessment in projects of infrastructural expansions like roads and construction of houses for different purposes was claimed to result in irreversible deforestation and degradation of natural resources in Huluka Watershed.

Agricultural Expansion:
Discussants were asked to opine on the causal linkage between agricultural expansion and land use/ land cover change. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). Discussants agreed that agricultural expansion is one of the causal factors for land use/land cover change in the watershed. The same concern was also raised by the key informants as all of them claimed that agricultural expansion is one of the causes for land use/land cover change in the watershed. It was also possible to observe current expansion of agriculture to forest areas in the watershed ( Figure 6).

Increased Demand for Fuel Wood and Wood for Construction:
Discussants were asked to opine on the causal linkage between increased demands for fuel wood and wood and land use/land cover change in Huluka watershed. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). The discussants of the focus group discussion agreed that the increase in demand for fuel wood and wood for construction is one of the causal factors for land use/land cover change in Huluka watershed. The same concern was also raised by the key informants as all of them claimed that increase in demand for fuel wood and wood for construction in their area has resulted in change in land use/land cover. It was also possible to observe current increased demand for fuel wood and wood for construction in Ambo town and Huluka watershed ( Figure 7).

Local Environmental Factors:
Discussants were asked to opine on the causal linkage between local environmental factors like soil quality and topography and Land Use/ Land Cover Change. Similar perceptions were received across the three streams of the watershed (upstream, midstream, and low stream). Participants of the focus group discussion agreed that the soil type of the watershed contributes much for land use/land cover change in the watershed due to its vulnerability to change. Moreover, hilly areas in the watershed were perceived to contribute much for land use/land cover change in Huluka Watershed. The same concern was also raised by the key informants interviewed as all of them claimed that local environmental factors (soil type and topography) are among the causes for land use/land cover in Huluka watershed. It was also possible to observe the physical impact of local environmental factors on land use/land cover in Huluka watershed (Figure 8).

Local Biophysical Drivers:
Participants of the focus group discussion were asked to opine on the causal linkage between local biophysical drivers like ooding and land use/ land cover change. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). Discussants agreed that that ooding disaster risk is a major biophysical driver for land use/land cover change in Huluka Watershed. For instance, many of the agricultural land areas in the watershed were reported to be affected by ooding risk resulting in soil erosion, decrease in soil fertility, and decrease in agricultural production. In addition, re risk on forests and grasses was reported as one of the local biophysical factors contributing for land use/land cover change in Huluka Watershed. Furthermore, prevalence of drought during dry season of the year was reported to affect growing plants in the watershed. The same concern was also raised by the key informants interviewed as all of them attested that local biophysical drivers are among the causes of land use/ land cover change in Huluka watershed. It was also possible to observe local biophysical drivers contributing for land use/land cover changes in Huluka watershed ( Figure 9).

Local Social Events:
Discussants were asked to opine on the causal linkage between local social events and Land Use/ Land Cover Change in Huluka Watershed. Similar perceptions were not explored across the three streams of the watershed (upstream, midstream, and low stream). For instance, no local social events were claimed by the discussants to be associated with land use/land cover change in both upstream and mid stream of the watershed. However, sudden displacement and abrupt policy shifts in their locality were reported as local social events contributing for land use/land cover change in low stream of Huluka Watershed. In relation to the aforementioned local social events in the low stream of the watershed, the sudden displacements of farming households associated with the expansion of Ambo University and abrupt policy shifts like land lease policy were reported to contribute for land use/land cover change in their area. The same concern was also raised by the key informants interviewed as all of them from the low stream of the watershed noted that local social events have contributed for land use/land cover change in their area.

Demographic Factors:
Participants of the focus group discussion were asked to opine on the causal linkage between demographic factors and land use/land cover change. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). Discussants agreed that natural increase and migration into the area are demographic factors contributing for land use/land cover change in Huluka Watershed. For instance, increase in human population in the watershed was claimed by the discussants as one of the drivers for degradation of land resources in Huluka Watershed. The same concern was also raised by the key informants interviewed as all of them con rmed that demographic factors are among the major causes of land use/land cover change in their watershed. It was also possible to observe the undesirable situation of Dendi Crater Lake at the upstream of the watershed attributable to increase in human population in the watershed ( Figure 10).

Economic Factors:
Participants of the focus group discussion were asked to opine on the causal linkage between economic factors and land use/land cover change. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). The discussants agreed that market growth and commercialization, urban expansion, and price increase are the economic factors contributing for land use/ land cover change in Huluka watershed. For instance, market growth and commercialization was claimed by the discussants to encourage production of high value agricultural products like vegetables. Urban expansion was also claimed to encourage conversion of land use from agriculture to nonagricultural land use. Price increase was also claimed as one of the economic factors causing land use/land cover change in the watershed ( Figure 11).

Technological Factors:
Participants of the focus group discussion were asked to opine on the causal linkage between technological factors and land use/land cover change in Huluka watershed. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). Discussants agreed that technological factors like agro technical change (intensi cation) and agricultural production factors contributes for land use/ land cover change in Huluka watershed. For instance, contour farming, using improved crop varieties, and crop rotation were claimed by the discussants as agro-technical change factors contributing for land use/land cover change in upstream of Huluka Watershed ( Figure   12). The same concern was also raised by the key informants interviewed as all of them con rmed that technological factors are among the factors causing land use/land cover change in Huluka watershed. It was also possible to observe the in uence of technological factors on land use/land cover change at upstream of the watershed.

Policy and Institutional Factors:
Participants of the focus group discussion were asked to opine on the causal linkage between policy and institutional factors and land use/land cover change. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). The discussants agreed that policy and institutional factors like formal policy and property right contribute for land use/ land cover change in Huluka Watershed. For instance, expansion of agriculture to forest areas and resulting deforestation and ooding risk in the watershed was claimed by the discussants to be highly associated with weak forest conservation policy and weak forest management institutional framework in the watershed. The same concern was also raised by the key informants interviewed as all of them contended that policy and institutional factors contributes for land use/land cover change in their watershed.

Cultural Factors:
Participants of the focus group discussion were asked to opine on the causal linkage between cultural factors and land use/land cover change. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). The discussants agreed that cultural factor like public attitudes and beliefs and individual and household behavior in the watershed contribute for land use/ land cover change in Huluka Watershed. For instance, change in public attitudes and beliefs towards natural forest conservation in the watershed were claimed by the discussants to be resulted in land use/land cover change in the watershed. In other words, expansion of agricultural land to forest areas is currently much encouraged and practiced by local people in the watershed. The same concern was also raised by the key informants who also con rmed that cultural factors are among the major factors for causing land use/land cover change in Huluka watershed.

Effects of Land Use/ Land Cover Change on Residents' Livelihoods in Huluka Watershed
Participants of the focus group discussions were asked to opine on the effects of land use/land cover change on their livelihoods in Huluka Watershed. Similar perceptions were explored across the three streams of the watershed (upstream, midstream, and low stream). Accordingly, the discussants claimed the following negative effects: increased ooding risk; increased soil erosion; increased sedimentation into the lake (Dendi lake) and rivers ( Huluka, Awaro, Debis, Boji, Bolo, Aleltu, Karkaro, and Korke); decrease in soil fertility resulting from ooding risk; and change in climatic parameters (decrease in annual rainfall and increase in heat during dry season). The same concern was also raised by all the key informants interviewed who contended that the land use/land cover change in Huluka watershed has negatively affected their livelihood underpinned by degradation of natural resources.
The participants of the focus group discussion were further asked to opine on the reciprocal cause-effect relationship between ooding risk and land use/land cover change in Huluka Watershed. The discussants a rmed that ooding risk in their watershed is highly linked to land use/land cover change as deforestation aggravate ooding caused by other factors. In other words, land areas without forest cover were claimed by the discussants to be much vulnerable to ooding. On the other hand, ooding risk in Huluka watershed was claimed to contribute for land use/land cover change in the watershed. Hence, reciprocal cause-effect relationship was claimed by the discussants to exist between ooding and land use/land cover change in Huluka Watershed. The same concern was also raised by the key informants interviewed in the watershed.
Participants of the focus group discussion were also further asked to suggest effective adaptation measures for land use/land cover change in Huluka Watershed. Similar suggestions were explored across the three streams of the watershed (upstream, midstream, and low stream).
Accordingly, the following adaptation measures were suggested by the discussants: Compost preparation and use; practicing crop rotation; effective planning to rehabilitate degraded lands through structural and non-structural soil and water conservation measures; strengthening the ongoing community-based soil and water conservation practices; institutionalizing appropriate environmental impact assessment into any local development projects; planting appropriate tree species and management on established soil and conservation structures; and establishing appropriate institutional framework for forest and other natural resources' management.
Participants of the focus group discussion were also further asked to opine on the strengths of the current community-based soil and water conservation measures practiced in their watershed. Similar perceptions on the strengths were explored across the three streams of the watershed (upstream, midstream, and low stream). Accordingly, the discussants identi ed the following strengths: increased soil fertility; healing of degraded land areas; conserved soil and water resources; reduced negative impacts of ooding risk on farmers' livelihoods; increased yield from crop production; and improved social capital among local farmers and between farmers and agricultural development professionals.
Participants of the focus group discussion were also further asked to opine on the weaknesses of the current community-based soil and water conservation measures practiced in their Watershed. Similar perceptions on the weaknesses were explored across the three streams of the watershed (upstream, midstream, and low stream). Accordingly, the discussants identi ed the following weaknesses: poorly supervised and managed soil and water conservation measures; appropriate tree species were not planted on the established soil and water conservation structures; rehabilitation of degraded lands was not considered in the soil and water conservation measures practiced; Some soil and water conservation measures practiced had negative downstream impacts due to poor management; established soil and water conservation structures were not protected from interference of livestock; and non-structural soil and water conservation measures (land use planning, and awareness creation and education were not well integrated into the current community-based soil and water conservation practices).

Discussion
The decrease in the forest land use/land cover type by 59. The increase in cultivated land use/land cover type by 105.3% over 38 years (1979-2017)

Conclusions
The purpose of the study was to analyze change in land use/ land cover in Huluka watershed where Ambo town is situated and explore local perceptions on its causes and negative effects. Increased ooding risk, increased soil erosion, increased sedimentation into the lake and rivers, decrease in soil fertility resulting from ooding risk, and change in climatic parameters (decrease in annual rainfall and increase in heat during dry season) were identi ed as the perceived major negative effects of land use/land cover change in the watershed.
To promote sustainable local development in the watershed, the following recommendations are forwarded: Appropriate land use planning and management in the watershed should be practiced to reduce the negative effects of land use/land cover change in the watershed.
Appropriate environmental impact assessment (EIA) should be conducted prior to initiating any local development projects; Appropriate institutional framework for integrated watershed management should be established for proper planning and management of the watershed development activities; Compost preparation and use, and crop rotation should be encouraged to improve the productivity of soil resources in the watershed; Rehabilitation of degraded lands through structural and non-structural soil and water conservation measures should be properly planned and managed; The ongoing community-based integrated watershed management should be strengthened; Appropriate tree species should be planted and managed on the established soil and water conservation structures; Local socio-cultural development in the watershed should be properly planned and managed to promote sustainable socio-cultural development in the watershed; and Local economic development in the watershed should be properly planned and managed to promote sustainable economic development in the watershed.

Declarations
Ethics approval and consent to participate