Introduction

Land cover dynamic has the global concern of the twenty-first century, with the dramatic implication for human survival. Land cover change is the change in the physical as well as biological characteristics of land which is attributable to management including conversion of grazing and forest land into farming land, pollution and land degradation, removal of vegetation, and conversion to non-agricultural uses (Quentin et al. 2006; Prakasam 2010; Shiferaw 2011). Recently research on land use and land cover change detection has drawn attention of many researchers (Liang et al. 2002; Ayele et al. 2016). It affects biodiversity, hydrological cycle, land productivity and the sustainability of natural environment (Lupo et al. 2001). Continuous from the previous and in the coming years land use dynamics has been playing a wide role of driving force in alteration of the global environment (Baulies and Szejwach 1998).

The increasing change is alarming, and can have a huge implication on local, regional, national and global environment and consequently affect the food availability (Minale 2013). According to Reid et al. (2000), land use and land cover is continuously changing the surface of the earth. In the past few decades the conversion of forest and wood land, grass and pasture land into agricultural and pasture land has dramatically increasing in the tropics (Turner 1990). According to Reid et al. (2000), land cover change is accelerated by human activities and natural processes. Similarly, the change due to the complex interaction of various social, economic and biophysical situations following agricultural diversification, advancement in technology coupled with alarming rate of population pressure.

Minale and Rao (2011) and Shiferaw (2011), pointed out that associated population pressure found to be negative result on land use change. Soil erosion, land degradation, destruction of habitat and biodiversity; loss of endemic species due to you to out migration are resulted from land use dynamics (Meyer and Turner 1992). Even though many controversies on the factors of land cover dynamics, few research studies concluded that demographic factor is intensively accelerate to land use cover change (Mather and Needle 2000). Alarming rate of population dynamics, insecure land use right, lack of credit facilities and lack of market availability are some of socio-economic factors which facilitates for the change of land cover. For the poor those are living under subsistence farming has no other option other than natural resource. There was mismanagement of natural resource such as overgrazing, de-vegetation and expansion of agriculture into the marginal land as well as steep slope for the survival of their livelihoods (Grepperud 1996; Minale and Rao 2012a, b; Amare 2013; Asres et al. 2016). Despite the expansion of cultivation from sloping into steeper slope with inappropriate soil and water conservation measure, crop production is still lagging behind by 2.67 % annual population growth rate (Asres et al. 2016). Intense pressure on agricultural land, forest land and the availability of fuelwood in the sounding area in Ethiopia is the result of spatial and demographic changes; it exerts massive pressure on land use, agricultural productivity, and the use of ecosystem (Minale and Rao 2011).

In most parts of the world, particularly in developing countries agriculture is the livelihoods of the population in turn primarily the most driver of land use change. However, limited studies have been done on long term trend of land cover change (Goldewijk and Ramankutty 2004). For instance, in east Africa in the last 50 years, as the expense of other land use, there has been intensive expansion of agriculture into marginal land (Yitaferu 2007). Semiarid and sub humid areas were dominated by pasture land with widely scattered settlement and agricultural activity before 1950, but then after there has been a massive change of grazing land into mixed crop- livestock agriculture.

Interaction between various socio-economic conditions of the society, population pressure, physiographic feature, and land use type has resulted in land use change, the conceptual framework of which is presented in Fig. 1. Therefore, land use classification used to analyse the interaction between socio-economic and land use, which is contributed for the dynamics of land use and land cover change resulted from diversified and intensified agriculture and livestock population (Mendoza et al. 2002).

Fig. 1
figure 1

Adopted from FAO 2004; Shiferaw 2011

Conceptual framework on the driving force of land cover change and associated effect.

Full size image

The interaction between various anthropogenic and natural factors cause for land cover change (Fasona and Omojola 2005) and the utilization of this resource by human population in time and space (Clevers et al. 2004), analysing land use and land cover change at watershed and sub-watershed level using Landsat imagery and clearly identifying the rate and extent of land cover change is critically important input for the prioritization of natural resource management. To monitor land use and land cover dynamics, geospatial techniques has important role, therefore, geographic information system tools are used to grasp information about extent, rate and magnitude of land use cover change and disseminate accurate information (Carlson and Sánchez-Azofeifa 1999; Guerschman et al. 2003; Dezso et al. 2005).

As FAO (1986) cited in Asres et al. (2016), in the mid-1980s, around 27 Mha highland part of Ethiopia was significantly eroded as the same time around 14 Mha was seriously eroded. It concluded that more than 2 Mha of agricultural lands has reached at the point of zero return. Critically analysing the driving force for land cover dynamics of the past trend is important to understand the recent changes and predict for future alteration. A study of land covers dynamics and its driving force in time and space provides favourable foundation for the sustainability of natural resource systems, because it used to reflect the state of watershed. Therefore, land cover change and its driving forces are important for designing policies and strategies for the sustainable natural resource management and use. Even though different studies have under taken about the extent and status of clearing of forest, land cover change and soil erosion in many parts of Ethiopia, poorly documented about land cover dynamics and its driving force. Therefore, the main objective of the study was to identify land use and land cover dynamics and its driving force in the Beressa watershed in the last 31 years.

Materials and methods

Study area

The watershed chosen for the present study lies between 39°37′E–39°32′E and 9°40′N–9°41′N. In administrative terms, it is located Basona District, North Showa zone of Amhara regional state. Situated 180 km distance north east of capital city, Addis Ababa, the watershed forms parts of northern central highlands of Ethiopia which is the parts of Abay basin. The area of the study watershed is 213.2 km2. The watershed is characterized by diverse topographic conditions like mountainous, dissected terrain with steep slope. The elevation ranges from 2200 to 3600 m a.s.l. The annual average temperature of the area is 19.7 °C; the annual maximum and minimum amount of rainfall is 1083.3 and 698.5 mm respectively. The most common types of soil are cambisols (locally called Abolse), vertisols (Merere), Andosols, Fluvisols and Regosols. Mixed crop-livestock is production system of the area perhaps the only source of livelihoods for the majority of the population. Barely, wheat, horse beans, field peas, lentils and chick peas are commonly growing crops. It is characterized by traditional, rainfed, labor intensive and subsistence oriented or hand to mouth. Cattle and sheep are the dominant types of livestock, but goats, equines, and chickens are also common. Because of rainfall dependent farming practice, farmers are always worried about the duration and intensity of rainfall.

Data source and methods of analysis

Land cover data The source for land and land cover dynamics was freely downloaded Landsat imagery from http://earthexplore.usgs.gov. The detail of satellite data area presented in Table 1. Landsta5 and landsat8, path/row of 168/53 with 30 m spatial resolution was acquired on 17/12/1984, 25/01/1999, and 23/12/2015. The imagery was processed using ArcGIS10.2.2 and ERDAS IMAGE14 software. Initially images were converted into Universal Transfer Mercator and geo-referenced to a datum in which Ethiopia has selected by WGS-84. To improve the image quality, it was enhanced using histogram equalizations. Then land use and land cover change detection of the study watershed was analysed for the last 31 years. To classify Landsat image supervised classification was used; initially more than 250 signature extractions was used in order to convert images into thematic land cover class.

Table 1 Materials and their source used for the study
Full size table

Before actual identifying the land cover change detection, Thematic Mapper was geo-referenced, transformed and enhanced. To reduce the resolution difference of Thematic Mapper images, using nearest neighbour re-sampling techniques the image was re-sampled into the same size. The topography of the study area was defined by DEM which is used to describe the elevation of points for the given area at a specific spatial resolution. In addition, sub-basin parameters including slope, slope length, slope width and stream network was obtained from the digital elevation models (DEM). The various steps developed and used to analyse, quantify and interpret the map are presented in Fig. 2.

Fig. 2
figure 2

General methodology for the classification of land use and land cover class

Full size image

Lastly six land use and land cover class were identified for the watershed and land use and land cover change were determined in Table 2. To get additional information about the long term experience of land use and land cover change practice in the watershed focus group discussion and informal interview have been conducted. For the discussion and in-depth interview elder peoples were purposively selected as they assumed to have better history of information about the trend of land use and land cover change.

Table 2 Descriptions of land use land cover class
Full size table

After having classified images, the geographical extent in terms of hectares for land use and land cover class was calculated for each mentioned time periods and the extent of change in land use type within and between time periods was compared. The change of different land use and land cover class has been performed using both ArcGIS10.2.2 and ERDAS IMAGINE14 and finally using the following formula calculation has been employed to know the rate of change hectare/year and percentage share of each class in the studied time periods;

$$\Delta {\text{A}}\left( \% \right) = \frac{{{\text{At}}2 - {\text{At}}1}}{{{\text{At}}1}} \times 100$$
(1)

where, ΔA (%) = percentage change in the area of land use and land cover class type between initial time At1 and time period At2 At1 = area of land use and land cover type at initial time At2 = area of land use and land cover type at final timeAs stated by Abate (2011), the rate of change of land use and land cover type was calculated by the following formula:

$${\text{R}}\Delta \left( {\frac{\text{ha}}{\text{year}}} \right)\frac{{{\text{Z}} - {\text{X}}}}{\text{W}}$$
(2)

where: R\(\Delta\) = rate of change Z = recent area of land use land cover type in ha X = previous area of land use land cover type in ha W = time interval between Z and X in years.

Result and discussion

Land use and land cover map

Red, green and blue band combinations were used to display stacked images in the standard color composite. Spectral class combination to display images frequently varies (Trotter 1998). For the analysis False Color composite bands 2, 3 and 4 used.

In order to meet the requirement of food demand, cultivated land has increased in all parts of the world at the expense of other land use class such as forest, bush/shrubs, barren land and grazing land particularly in developing countries in which majority of inhabitants are depending on agriculture for their survival (Lambin et al. 2003). In the present study similar trend in the watershed has been found where farm land and settlement area have increased over time in all the analysis time periods. Substantial land use change has been observed in the study area since 1970s. To show clearly this substantial change through over time due to various use and the overall change was presented in the year 1984, 1999 and 2015 (Figs. 3a, b, c, 4, 5, 6, 7; Table 3).

Fig. 3
figure 3

Land use and land cover map of Beressa watershed. a LULC of Beresa Watershed in 1984. b LULC of Beresa Watershed in 1999. c LULC of Beresa Watershed in 2015

Full size image
Fig. 4
figure 4

Land use and land cover types and area coverage of the study watershed in hectare from 1984 to 2015

Full size image
Fig. 5
figure 5

Percentage share of land use and land cover class of the study watershed from 1984 to 2015

Full size image
Fig. 6
figure 6

Change in Land use and land cover class of the study watershed in hectare from 1984 to 2015

Full size image
Fig. 7
figure 7

Percentage change of Land use land cover class of the study watershed from 1984 to 2015

Full size image
Table 3 Land use land and cover change of Beressa Watershed during 1984–2015
Full size table

Land use and land cover dynamics

In the present study six classes of land use and land cover were presented namely farm land, barren land, forest land, grazing land, settlement and water body. The land use and land cover dynamics is discussed in the subsequent sections.

Farm land In the study area agriculture occupy the largest share of land cover class (57.1, 61.9 and 67.5 % in 1984, 1999 and 2015 respectively). This implies farm land has been extensively increased at the expense of grazing, barren and forest land. This is due to increased demand because of population growth, additional farm land required to full fill food demand. As a result of extensive expansion of farm land (18.2 %), negatively contributed for the decrease of grazing and barren land by 50.6 and 60.7 % respectively in the last 3 decades (Figs. 3a, b, c, 4, 5, 6, 7; Table 3). Due to ever increasing of cultivated land farmers were exert pressure on forest, bushes/shrubs, grazing and barren land and resulted for further accelerate erosion and degradation. Likewise, similar study elsewhere, alarming rate of population growth resulted for the change of land cover class through time (Turner 2009). Similarly, Shiferaw (2011), limited access of off-farm employment opportunity has made the farmers involved clearing of forest and further expanding cultivation into steep slope, grazing land, barren land and other types of land cover class.

Settlement Settlement area in the study watershed has shown increased persistently in the time periods. The total area of the watershed covered by settlement has increased by 24.9 % from 1984 to 1999, 28 % from 1999 to 2015 (Figs. 3a, b, c, 4, 5, 6, 7; Table 3). Other than other types of land use systems, the expansion of both rural and urban settlement took the largest share by converting other land use types for instance forest, bushes, and shrubs, barren and grazing land. In the stated time periods (1984–2015) there has been 59.7 % increase of settlement land within the last 31 years. Due to continuous increase of population number they need additional land for settlement area, which costs 520.5 hectare of other land use type.

Grazing land One of the most dominant land cover of the watershed was grazing land it holds 12.4 % of the total land cove types (2633.2 ha) in 1984; however, in 1999 the coverage has been decreased by 2.4 % (from 12.4 to 10 %). Likewise, in the 2015 the share of grazing land was shrunk to 6.1 % which was double of the decrease in the 1999. Overall in the first periods from 1984 to 1999 the shrinkage of grazing land was reduced by 510.2 ha (19.4 %); similarly, in the second periods from 1999 to 2015 it has been decreased by 823 ha (38.7 %) (Figs. 3a, b, c, 4, 5, 6, 7; Table 3). The analysis of land use cover change for 31 years between 1984 and 2015 of this study indicated that 1333.2 ha (50.6 %) of grazing land has lost. The decrease of gazing land possibly the result of growing demand of more arable land for agricultural cultivation and growing demand for newly formed household for settlement in the watershed. Thus conversion grazing land into farm land and settlement is the common phenomenon practiced in the watershed considered in the present study.

Barren land Barren land within the stated years has shown continuously decreasing trend from 13.1 % in 1984 to 10.1 %, 5.2 % 1999 and 2015 respectively. With alarmingly and intensively declining trends in the first periods from 1984 to 1999 by 23.1 % and further in the second periods from 1999 to 2015 declined by 48.9 %. For the last 31 years, about 60.7 % of barren land was changed into other type of land cover classes. This is due to availability of fixed plot of farm land in collaboration with alarming rate of population growth negatively contributed for the decline of barren land.

Forest Land Another most dominant land use cover class of the study catchment was forest resource which ranged densely vegetated trees (natural forest), plantations, shrubs and bushes. The area covered by such forest could be ever green and mixed forest land. From the total area of the catchment in 1984 the share of forest coverage has 12.8 %, in contrast the coverage slightly decreased into 12.4 % in the year 1999. As we all know the decrease in forest coverage corresponds to increase population number and extensive expansion of agricultural land. However, in 2015 it has regenerated by 1.2 %.

The decrease of forest in the first periods between 1984 and 1999 around by 72.8 ha (0.3 % lost), even though too small, given the increase in expansion of agricultural land and increase in population number, it is contradicted from the expectations. However, in the second periods between 1999 and 2015 the share of forest coverage has increased by 4.7 % because the shrinkage of natural forest coverage attributable to regenerated and increased household and community level tree planting. Therefore, over the 31- year time period in between 1984 and 2015 the share has increased by 6.5 %. On the basis of local resident response obtained from interview and group discussion, various major reasons have been positively contributed for the increase of share of forest coverage such afforestation, private and community level tree plantation of sesbania susban, tree Lucerne and eucalyptus trees.

It is obviously true that land surface with little vegetative cover exposed to erosion, degradation runoff and insignificant water retention. The increased in runoff caused for intensive sheet erosion and further extend to rills and formation of gullies. The masses of top soil removed from highland to lowland areas, and cause for intense problems for the downstream siltation of ponds and reservoirs, water borne diseases, water pollution, and sediment deposition on fertile farm lands. In many part of the country such problems have already manifested.

Water bodies Water bodies include ponds, springs, streams, and rivers. In the study area water bodies covered only 0.56 %, in 1984 and decreased to 0.5 % but increased to 1.1 % in 2015 (Figs. 3a, b, c, 4, 5, 6, 7; Table 3). The possible explanation of this fluctuation water bodies in the first periods (1984–1999 decreased by 0.3 %) is due to decrease rainfall pattern in the area. Likewise, the water harvesting habit of local farmers was low. But in the second periods (1999–2015) the availability of ponds and springs was persistently increased by 101.6 %. In the study periods for the last 31 years it has increased by 101 %.

Rate of land use and land cover change dynamics

The rate of changes of farm land, grazing land, forest land, water body barren land and settlement area cover for the study watershed have already been presented in Table 2. This result indicated that though resource is fixed, there was various rate of change in different land cover types. However, the rate of change of different land cover types has slightly variables among them.

The analysis indicated that between 1984 and 1999, farm land and settlement area has increased with the rate of 67.5 ha/year and 14.5 ha/year respectively caused for the outflow of grazing land, barren land and forest land; in the same periods grazing land, barren land, forest land and water body was decreased by 33.3 ha/year, 43.1 ha/year, 5 ha/year, and 0.03 ha/year respectively. Likewise, between 1999 and 2015 expansion of farm land and settlement persistently increased with a rate of 75.5 and 18.9 ha/annum. Unlike in the first periods, unexpectedly with increasing rate of settlement and farm land between 1999 and 2015 forest land and water body was increased by 15.6 and 7.1 ha/year. In the second periods of study years between 1999 and 2015 the share of forest coverage surprisingly increased which was attributable to household and community level afforestation and reforestation practice, the government of Ethiopia in collaboration with donor organization and mobilized the community to attention for indigenous and multipurpose trees; beside this households were planting eucalyptus trees around homestead and farm land which is significantly contributed for the present extent of forest coverage of the study watershed. However, grazing and barren land rapidly decreased by with the rate of 51.4 and 65.8 ha/year in between 1999 and 2015. Over the last 31-year expansion farm field, settlement area, forest land and water body increased with a rate of 71.6, 16.8, 5.7, and 3.7 ha/year contrary to this the share of grazing land and barren land shrunk with a rate of 43 and 54.8 ha/year respectively (Table 4).

Table 4 Rate of change in LULC of Beressa Watershed
Full size table

The result of this finding particularly barren land different from a finding conducted by (Shiferaw 2011) indicated that in his study it has been increased by 256 ha/year between 1972 and 1985. Different study elsewhere made by Gashaw et al. (2014), and Dessie and Kleman (2007), indicated that the size of farm land and settlement area have been intensively expanded at the expense of forest cover, bare land and brazing land without significant conservation measure.

Cause of land use and land cover change dynamics

Even though the extent time periods of event occurrence are variables, various human and natural factors are the main cause for land use land cover dynamics (Meyer and Turner 1994). Though the effect of population growth rate on land cover dynamics is controversial, elsewhere in many literatures rapid rate of population growth rate one of the root cause for the change of land cover dynamics. According to (Barbier and Burgess 1996), instead of negatively affecting the study concluded that rapid population growth has positive role in availability of resource. On the contrary particularly in the highlands of Ethiopia in which population pressure is intense resulted for resource erosion and degradation (Grepperud 1996). Likewise, elsewhere in many part of Ethiopian high lands, pressure associated with populations has argued negative implication on forest land, grazing land, barren land, riparian vegetation and farm land (Tekle and Hedlund 2000). It is true to Beressa watershed where rapidly growing of population brought shortage of land, removal of forest cover and soil erosion and land degradation. However, local community were motivated with respect to afforestation practice consequently slow increase in forest coverage in the Beressa watershed. In addition, shortage of land resource forced them to cultivate gazing land, barren land and steep slopes land. Therefore, resource become more vulnerable for further erosion and degradation consequently shifted to other land use land cover class.

Implication of land use and land cover change dynamics

The change of land use and land cover class may not necessarily result in land degradation and soil erosion. However, if the change of land use land cover class is rapidly expanding into farm land, grazing land and barren land, fertile soil is more susceptible to massive erosion and degradation, particularly the land surface without dense forest (Tegene 2002; Maitima et al. 2009).

According to the classified image of change was detected in different land use class between 1984 and 2015 indicated that grazing land, barren land and forest land cover class were transferred into farm land and settlement area. This indicated that how land use and land cover class change over time accelerate erosion and degradation and associated consequence. The shift of other land use land cover class into farm land towards steep slope and barren land is the root cause for intensive erosion, massive degradation, siltation, water borne disseise, and flood.

Likewise, Tegene (2002) explained in his study rapid expansion of agricultural land into steeper slope has aggravated for erosion and degradation in Ethiopia. Similarly, Amede et al. (2001) illustrated destruction of vegetative cover because of expansion of farming practice into steeper slopes particularly in the highland of Ethiopia in which intensive farming practice undertaken without appropriate conservation practice is resulted for depletion of fertile soil. Expansions of farm land at the expense of grazing land resulted for insufficient availability of fodder for livestock and adversely affect the productivity livestock similarly, absence of animals for land preparation and transportation service. Likewise, due to the outflow of gazing land and forced farmers reducing livestock number consequently reduced availability of manure for soil fertility amendment and hence reduced in crop production.

According to Abbas et al. (2010), the change of land use land cover class significantly aggravated the surface runoff, soil erosion, land degradation, sedimentation, siltation, drought, migration, desertification loss of biodiversity, decrease in productivity and famine. In summary, the study has indicated the cause and consequence of land use land cover change with over all environment of the study watershed presented in Fig. 8.

Fig. 8
figure 8

Adopted from Maitima et al. (2009)

Conceptual linkage of cause and consequence of land use land cover change of the watershed.

Full size image

Conclusion

In the last 31-years land use and land cover dynamics have undergone considerable change in the Beressa watershed. The land use and land cover dynamics observations showed the expansion of farm land and settlement land leading. On the contrary the grazing land and barren land declined. Contrary to the finding other studies elsewhere, Minale and Rao (2012a, b), Nurelegn and Amare 2014; Asres et al. (2016), even though farming and settlement was rapidly expanded, the trend of forest cover was increased. Increasing forest cover was possible due to the incentive provided by the government for community and household level indigenous tree plantations. The support of government and the efforts of communities are admirable because it is yielding a positive impact in preserving the ecology and the economic wellbeing of the community. The land use and land cover dynamics beyond onsite implications it has offsite environmental implication because massive soil erosion and land degradation has no limited boundaries.

The increasing forest cover is also affecting offsite environmental condition because massive soil erosion and land degradation has reduced. This could be possible due to proper integrated approach of upstream and downstream community under national and international support for environmental and ecological protection. Therefore, site specific community based awareness created about the appropriate use of available resources as well as the conservation and rehabilitation of environment proved to be very effective.