Land cover changes and their causes in Libokemkem District of South Gonder, Ethiopia

Author’s Accepted Manuscript Land use/Land cover changes and their causes in Libokemkem District of South Gonder, Ethiopia Fikirte Demissie, Kumelachew Yeshitila, Mengistie Kindu, Thomas Schneider www.elsevier.com/locate/rsase

PII: DOI: Reference:

S2352-9385(17)30076-9 https://doi.org/10.1016/j.rsase.2017.10.001 RSASE92

To appear in: Remote Sensing Applications: Society and Environment Received date: 6 April 2017 Revised date: 26 September 2017 Accepted date: 3 October 2017 Cite this article as: Fikirte Demissie, Kumelachew Yeshitila, Mengistie Kindu and Thomas Schneider, Land use/Land cover changes and their causes in Libokemkem District of South Gonder, Ethiopia, Remote Sensing Applications: Society and Environment, https://doi.org/10.1016/j.rsase.2017.10.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Land use/Land cover changes and their causes in Libokemkem District of South Gonder, Ethiopia Fikirte Demissiea,b,*, Kumelachew Yeshitilaa, Mengistie Kinduc, Thomas Schneiderc a

Ethiopian Institute of Architecture, Building Construction and City Development (EiABC), Addis

Ababa University, P.O. Box 518, Addis Ababa, Ethiopia b

Department of Geography and Environmental Studies, Bahir Dar University, P.O.Box 79, Bahir Dar

Ethiopia c

Institute of Forest Management, Department of Ecology and Ecosystem Management, Center of Life

and Food Sciences Weihenstephan, Technische Universität München, Hans-Carl-von-Carlowitz-Platz 2, D-85354 Freising, Germany [email protected] [email protected] [email protected] [email protected] [email protected] [email protected](T.Schneider) *Corresponding author: P.O.Box 1871 Code 1110 Addis Ababa, Ethiopia

Abstract This study analyzed land use/land cover (LULC) changes and their causes for the last 42 years from 1973 to 2015 covering an areas of about 265km2 in South Gonder Zone of Ethiopia, where area closures and/or forest conservations have been practiced since the late 1970s. Five sets of Landsat images (1973, 1985, 1995, 2003 and 2015) were the main input data from which five LULC maps were produced by employing pixel-based supervised image classification and changes of four periods were analyzed in GIS. Key informants interview and focus group discussion methods were used to 1

identify the causes linked to the changes. The change results showed continuous decline of forest lands throughout the first (1973-1985), second (1985-1995), third (1995-2003) and fourth (2003-2015) periods by 37.6%, 43.2%, 26.9% and 31.5% respectively. Grasslands also decreased in all the four periods by 4.6%, 13.6%, 19.5% and 11.4% respectively. On the other hand, agricultural lands increased by 28.2% in the first, 23% in the second, 20.7% in third and 11.7% in the fourth periods. All in all, about 60.1% of the land experienced changes in LULC in 42 years. The identified causes were population pressure and associated demands (e.g increasing wood extraction for fuel), regime changes, interventions and programs implemented in the area. We conclude that previously introduced strategies could not stop the decline of vegetation (e.g. forests and grasslands) in the study area. We suggest sustainable land use planning and management, proper implementation of different strategies together with local communities and further studies to explore alternative strategies. Keywords: Area closure; community; Ethiopia; GIS; image; Landsat.

1. Introduction Land use /land cover (LULC) changes are the result of a complex interplay between socioeconomic, institutional and environmental factors (Turner et al. 1994; Turner et al. 1995; Tegene, 2002; Lesschen et al. 2005; Lambin and Geist, 2006; Falcucci et al. 2007; Li et al. 2009). They have important impacts on the functioning of socio-economic and environmental systems with important tradeoffs for sustainability, food security, biodiversity and the vulnerability of people and ecosystems to global change impacts (Lesschen et al. 2005; Falcucci et al. 2007). LULC changes are investigated through the application of remote sensing techniques, which otherwise are not possible with conventional ground methods, especially for capturing data for inaccessible areas (Kindu et al. 2013). Remote sensing has been used successfully in 2

mapping a range of LULC and their changes at a variety of spatial and temporal scales (Zeleke and Hurni, 2001; Temesgen et al. 2013; Desalegn et al. 2014). Although the LULC changes vary greatly across the world, their ultimate outcome is generally the same, i.e., the use of natural resources to meet the increasing demands caused by human population growth, often at the expense of degrading environmental conditions (Costa et al. 2003; MEA, 2005; Foley et al. 2005; Lambin and Meyfroidt, 2011; Wondie et al. 2011; Meshesha et al. 2014). This resulted in loss of biodiversity (Klenner et al. 2009), land degradation (Meshesha et al. 2014; Sewnet, 2015) and forest fragmentations (Ligdi et al. 2010; Rands et al. 2010). To reverse the adverse impacts of LULC changes, international and local policy initiatives are, thus, pushing for conservation (Lindenmayer et al. 2006; Rands et al. 2010; PorterBolland et al. 2012) and restoration as means to protect biodiversity and food security (Jacobs et al. 2015). Several authors (Naughton-Treves et al. 2005; Bray et al. 2008; Porter-Bolland et al. 2012) also stressed the need for different forest conservation strategies across the tropics that integrate local economic development, protected area and community managed areas. Garcia-Frapolli et al. (2007) reported that identifying the causes of LULC changes is an important step towards successful biodiversity preservation. Therefore, accurate information on land use and land cover changes and the causes behind them is essential for designing a sound environmental planning and management strategy. Over the last decades, a number of publications reported the occurrences of LULC changes, their causes and the resulting land degradation in different parts of Ethiopia (Bishaw, 2001; Taddese, 2001; Zeleke and Hurni, 2001; Birhane et al. 2006; Bongers et al. 2006; Alelign et al. 2007; Tadele and Förch, 2007; Kindu et al. 2013; Temesgen et al. 2013; Meshesha et al. 2014; Desalegn et al. 2014; Yesuf et al. 2015). 3

To address the ongoing problem of land degradation in Ethiopia, efforts have been made at various levels. For example, between mid-1970s and early 1980s, there were massive conservation programs with the main components of afforestation, on-farm and hillside terracing, area closures and gully control (Rahmato, 2001; Nedessa et al. 2005). Similar natural resources conservation and management efforts have been made since the late 1970s by governmental and non-governmental organization for the land degradation problems in South Gondar, Libokemkem District, of Ethiopia. However, the effect of such efforts on the changes of the LULCs in relation to the conservation measures is not systematically studied so far. Therefore, in this study, we analyzed and evaluated land use/land cover changes from 1973 to 2015 within Libokemkem District, where area closures and/or forest conservations have been practiced since the late 1970s on the one hand and explored the causes behind the changes on the other. 2. Materials and Methods 2.1 Study Area The study was conducted in Libokemkem District, which is part of the Lake Tana watershed, located in the South Gondar Administrative Zone of the Amhara National Regional State at about 645 km Northwest of Addis Ababa, the capital of Ethiopia. The district lies within 110 58' 1.5''and 120 22' 6.7'' latitude, and 370 33' 25.4'' and 370 58' 16.5'' longitude (Figure 1). The District comprises a total of 32 Kebeles (the smallest administrative unit of Ethiopia) and an area of 1082km2 (CSA, 2007). According to the 2007 census, the population of the District was 198,435 (CSA, 2007). The study area is generally characterized by sub-humid highlands. The mean annual temperature of the region ranges from 18ºC to 25ºC. This area receives a unimodal rainfall of approximately 900mm to 1400mm per year, the majority of which falls between June and August. Volcanic rocks mainly basalt characterize the geology of the study 4

area. Light dark, grey, whitish, reddish or brown are the common rocks. The soils are mostly shallow and sandy and characterized by low organic matter due to land degradation as a result of erosion and continuous cultivation (Zegeye et al. 2011). Agriculture, including crop farming and animal husbandry, is the basis of livelihoods of the people of the study area and it is characterized by rain fed, oxen driven, small scale subsistence oriented and labor intensive activities. Average land holding in the District is about one hectare/head. In the area various types of crops such as teff (Eragrotis teff), beans (Phaseolus vulgaris L.), wheat (Triticum aestivum L.), and barley (Hordeum vulgare L) are grown.

Figure 1. Map of the Libokemkem District within Amhara Region of Ethiopia (Source: CSA, 2007). 5

2.2 Data collection A total of 8 Kebeles covering an area of 265 km2, where area closures and/or forest conservations have been practiced were selected for this study. The Kebeles were selected due presence of to area closures since 1970s, nearness to the Lake Tana, implementation of bylaws and Participatory Forest Management (PFM). Major types of spatial data used to study the land use land cover dynamics of the study area were Landsat MSS, TM, ETM+, and Landsat 8OLI of the year 1973, 1985, 1995, 2003 and 2015, respectively (Table 1). The images were obtained from the United States Geological Survey (USGS) Earth Resources Observation Systems (EROS) data center under the Landsat Archive (http://earthexplorer.usgs.gov/). The years were selected by taking into consideration events such as years of conservation measure practiced (i.e. area closure), government change, quality and availability of Landsat images, and the implementation of land redistribution.The images were acquired in the same season to avoid the effect of seasonal variations (Kindu et al. 2013). The images were geometrically and radio metrically corrected. Pre-processing of the images like sub-setting and layer stacking task were performed before the commencement of the actual classification. Moreover, all the acquired satellite images were enhanced using histogram equalization to improve image quality. Other data were also collected to the analyses, including GPS points of major LULC classes, historical aerial photos, Kebele boundary, location of towns and area closures. All the datasets were projected to the Universal Transverse Mercator (UTM) map projection system zone 37N and datum of World Geodetic System 84 (WGS84), ensuring consistency between datasets during analysis. Table 1. Summary of spatial datasets used in this study Year

Sensor

Resolution

Acquisition date

Path/row

1973

Landsat L1-5MSS

60m

February 01, 1973

182/52

6

1985

Landsat L4-5 TM

30m

February 26, 1985

169/52

1995

Landsat L4-5TM

30m

February 06, 1995

169/52

2003

Landsat ETM+

30m

February 04, 2003

169/52

2015

Landsat 8OLI/TIRS

30m

February 13, 2015

169/52

Seven LULC classes were identified considering the local specific conditions and following the approaches of Zeleke and Hurni (2000) and Kindu et al. (2013) (Table 2).

Table 2. Description of land use and land cover classes No.

Land use types

Description

1

Agricultural lands

Area of land ploughed/prepared for growing various crops. This category includes areas currently under crop, fallow and land under preparation

2

Wetlands

Area partly covered by water and marshlands

3

Degraded lands

Land surface without vegetation cover, or with rocky land and quarries

4

Settlements

Urban and rural housing area

5

Bush/shrub lands

Land covered by small trees, bushes/shrubs

6

Grasslands

Land covered by grass and small shrubs dominated by grass

7

Forest lands

Area dominated by high forests including plantations

To further understand the process behind the LULC dynamics, major causes of the changes were explored using 61 key informant interviews and focus group discussions, including from local communities, government officials and NGOs working in the study site (e.g. GIZ and SOS Sahle). The key informants from each kebele were selected considering age (elder community member of age 50 and above), continually living in the area for more than 30 years, willingness to participate in the interview and considered by local people knowledgeable about the study area. Officials and experts from Libokemekem District 7

Agricultural Development office Amhara Region Bureau of Agriculture and Bureau of Environmental Protection, and Land Administration were interviewed as key informants because of their adequate knowledge about the issues under investigation. The key informant interview with each individual took about two hours. Moreover, the study included eight Focus Group Discussions (FGDs), each having six members from both sexes. Each focus group discussion lasted three hours. In addition, secondary data sources, such as technical reports, published works and public statistics were utilized to document the scale of the major causes. 2.3. Data analyses A pixel based supervised image classification with maximum likelihood classification algorithm was used to classify the land-use/cover types of each reference years (Lillesand and Kiefer, 2000). Ground truth points collected from different sources were used during classification and accuracy assessments of each reference years. A minimum of about 30 random ground points per class were used for efficient accuracy assessment (Congalton and Green, 2009). The sources for ground sample points were field visits, Google Earth image, historical black and white aerial photos, and raw images. Raw images through visual interpretation in combination with historical black and white aerial photos were used to collect sample points for earlier reference years, .i.e. 1973, 1985, 1995 and 2003. Similarly Kindu et al. (2013) followed the same approach to study LULC change of Munessa-shashemene Landscape of the Ethiopian highlands. Field visit and Google Earth datasets were used to collect samples for the 2015 reference year. The accuracy assessment points were independent from those used as training samples. Accuracy assessment was carried out by creating error matrix. The matrix compares information obtained by reference sites to that provided by classified image for a number of sample areas. Accordingly, overall accuracy, producer’s and 8

user’s accuracies, and Kappa statistic were calculated from the error matrix for each reference years. Then, the LULC changes between four periods (i.e., 1973-1985, 1985-1995, 1995-2003 and 2003-2015) were quantified. Change analysis was conducted using post-classification image comparison technique, which was used in order to minimize possible effects of atmospheric variations and sensor differences with spatial resolution (Lu et al. 2004). Images of different reference years were first independently classified, and, afterwards, change detection processes were performed. The percentage of land use/land cover change detection was made using the following formula (Kindu et al. 2013): (

)

Hence, positive values suggest an increase whereas negative values imply a decrease in extent of LULC. In addition, change detection matrix of ‘from-to’ was derived to show LULC class conversion transitions during the 42-year period by overlaying the 1973 and 2015 classified maps. Image classification, accuracy assessments and change analysis were undertaken in ArcGIS10.3 software. The datasets collected about the causes of LULC changes from FGD and key informants were analyzed qualitatively and triangulated with secondary data and the result of LULC. 3. Results and discussion 3.1. Land use/land cover classification The areal extent of seven LULC types and their spatial distribution, for each time step, are presented in Table 3 and Figure 2, respectively. In 1973, 45.3% of the land was covered with grasslands followed by agricultural lands (26.2%), forest lands (17.9%), and bush/shrub lands (8%). Wetlands, degraded lands, and settlements accounted the smallest proportion of the 9

study area. Though the extent varied among land use classes, the order of proportion occupied by the LULC types in the study area remained the same in 1985.However, in 1995, agricultural lands were dominant (41.3%), followed by grasslands (37.4%), bush/shrub lands (12.2%), and forest lands (6.3%).The order of distribution continued to be the same in 2003. In 2015, more than half of the study area occupied by agricultural lands, followed by grasslands (26.6%), and bush/shrub lands (7.1%) were the dominant land use types of the study area. The remaining part occupied by others LULC types. Table 3. Summary of the classified land use/land cover from 1973 to 2015 1973

1985

1995

2003

2015

Land use class Ha

%

Ha

%

Ha

%

Ha

%

Ha

%

6,960.8

26.2

8,924.7

33.6

Wetlands

436.4

1.6

222.0

0.8

279.2

1.1

227.6

0.9

175.5

0.7

Degraded lands

32.1

0.1

87.4

0.3

125.9

0.5

239.3

0.9

615.8

2.3

Settlements

230.6

0.9

255.4

1.0

328.4

1.2

886.0

3.3

1,169.1

4.4

Bush/Shrub lands

2,113.3

8.0

2,633.0

9.9

3,256.4

12.2

2,756.0

10.4

1,894.8

7.1

Grasslands

12,050.0

45.3

11,494.5

43.2

9,933.3

37.4

7,993.0

30.1

7,083.0

26.6

Forest land

4,757.7

17.9

2,969.3

11.2

1,687.3

6.3

1,233.7

4.6

845.3

3.2

Agricultural 10,975.2 41.3 13,250.9 49.8 14,804.1

55.7

lands

10

Figure 2. Land use/land cover map of the study area from 1973- 2015. 3.2. Accuracy assessments Table 4 presents summary of the LULC classification accuracy for each of the considered years. The overall accuracies were between 85% and 89% with the Kappa statistic of 0.81 and 0.86, respectively. Landis and Koch (1977) grouped Kappa coefficient into three: a value greater than 0.80 representing strong agreement; a value between 0.40 and 0.80 representing moderate agreement; and a value below 0.40 representing poor agreement. Thus, the Kappa results of this study showed a strong agreement for each of the five classified images and the overall accuracies were within the acceptable range for further LULC change analysis (Kindu et al. 2013).

11

Table 4. Summary of accuracy assessments of LULC classification from 1973 to 2015 1973 Class Name

1985

1995

2003

2015

UA(%) PA(%) UA(%) PA(%) UA(%) PA(%) UA(%) PA(%) UA(%) PA(%)

Agricultural lands

85.2

83.3

77.3

93.4

78.6

92.0

90.7

92.4

79.8

92.0

Wetlands

97.4

92.5

96.7

96.7

90.0

90.0

96.7

78.4

100.0

96.7

Degraded lands

82.6

63.3

96.2

83.3

92.0

76.7

75.9

91.7

80.0

88.9

Settlements

78.8

86.7

96.2

62.5

96.8

71.4

90.4

94.0

95.7

88.0

lands

84.6

80.0

83.3

81.8

88.0

80.0

95.8

76.7

92.3

78.7

Grasslands

77.3

83.6

88.9

88.0

86.0

88.9

77.1

87.1

84.1

76.3

Forest lands

92.5

95.4

98.5

100.0

97.0

98.5

100.0

98.2

100.0

100.0

Bush/Shrub

Overall accuracy

85%

88%

87%

89%

88%

Kappa statistic

0.81

0.86

0.85

0.86

0.85

3.3. Land use/land cover changes The land use/land cover changes were categorized into four periods 1973-1985 (first period), 1985-1995 (second period), 1995-2003 (third period) and 2003-2015 (fourth period) (Table 5). The change results showed continuous decline in forest lands and grasslands throughout the study periods though the percentage changes (rate of decline) fluctuate between periods. The forest lands were decreased by 37.6%, 43.2%, 26.9% and 31.5% in the first, second, third and fourth periods, respectively. The highest rate of decline of forest lands was during the second period, which might be associated with governmental changes, as it was observed in other studies (Eshetu, 2014; Kindu et al. 2015). Grasslands also declined by 4.6% (first period), 13.6% (second period), 19.5% (third period) and 11.4% (fourth period). On the other 12

hand, agricultural lands, degraded lands, and settlement showed an increasing trend in all the four periods. Specifically, agricultural lands increased by 28.21% in the first, 22.9%in the second, 20.7% in third and 11.7% in the fourth periods. The incremental percentage changes of agricultural lands reduced through time between periods. Zeleke and Hurni (2001) reported similar trends as a result of absence of suitable land for agriculture. Similarly, degraded lands increased by 172.5%, 44.1%, 90.1% and 157.3% in all the four study period, respectively. However, bush/shrub lands and wetlands showed both decreasing and increasing trends during the four study periods. Table 5. Land use and land cover changesin different periods of time Class Name

1973-1985

1985-1995

1995-2003

2003-2015

Ha

%

Ha

%

Ha

%

Ha

%

Agricultural lands

1963.96

28.21

2050.5

22.98

2275.7

20.73

1553.2

11.72

Wetlands

-214.45

-49.14

57.18

25.76

-51.58

-18.48

-52.12

-22.90

Degraded lands

55.32

172.53

38.51

44.07

113.44

90.11

376.44

157.28

Settlements

24.73

10.72

72.99

28.58

557.67

169.83

283.07

31.95

Bush/shrub lands

519.69

24.59

623.42

23.68

-500.36

-15.37

-861.24

-31.25

Grasslands

-555.50

-4.61

-1561.16

-13.58

-1940.38

-19.53

-910.01

-11.39

Forest lands

-1788.34

-37.59

-1282.04

-43.18

-453.61

-26.88

-388.37

-31.48

The LULC change matrix analysis of 42 years from 1973 to 2015 of study area is presented in Table 6. A total of about 15,939.8 ha (60.1%) of the land experienced changes in LULC. For example, of the 12,026.6 ha grasslands in 1973, 4,372.50 ha (36.4%) remained unchanged. Similarly, only 635.7ha (13.4%) of the forest lands remained unchanged out of the 4,747.1 ha in 1973. The remaining 87.6% of the 1973 forestlands were converted to others, mainly, to 13

agricultural lands (about 45.3%). Similar changes in LULC were reported by previous studies in different parts of the country (Zeleke and Hurni, 2001; Tegene, 2002; Barreto, 2006; Kindu et al. 2013; Meshesha et al. 2014; Hailemariam et al. 2016). For example, Zeleke and Hurni, 2001 reported that 99% of the forest covers were converted to agricultural land between 1957 and 1995 in northwestern Ethiopian highlands. Similarly, Kindu et al. (2013) found high conversion of natural forests and woodlands from 1973 to 2012 in Munessa-Shashemene landscape of the Ethiopian highlands. Also, Hailemariam et al. (2016) pointed out loss of forests and again of farmlands at the same magnitude in Bale mountain of Ethiopia during 1985 to 2015. Table 6. Land use/land cover change matrix from 1973 to 2015 in Hectare Class Name

Agricultural

Wetlands Degra

From lands

ded

To Agricultural

Settle

Bush and

Grassland

Forest

Total

ments

Shrub

s

land

2015

lands

lands

5302.55

166.31

5.48

114.09

1263.00

5760.55

2148.37

14760.35

Wetlands

14.47

131.14

0.00

0.00

0.00

25.67

0.02

171.29

Degraded

118.90

8.52

7.78

15.77

45.35

352.96

66.02

615.30

Settlements

248.68

6.20

2.20

25.19

95.97

654.15

132.79

1165.17

Bush/ Shrub

244.95

17.91

0.13

5.65

101.08

698.63

821.96

1890.31

Grasslands

993.50

101.25

16.37

68.76

573.83

4372.50

942.22

7068.43

Forest land

18.33

0.00

0.00

0.44

28.27

162.15

635.68

844.88

6941.39

431.33

31.96

229.90

2107.49

12026.61

4747.05

26515.73

lands

lands

lands

total 1973

14

3.4. Causes of land use/land cover changes LULC changes were caused by a number of factors. Based on the results of key informants and focused group discussions (FGDs), the causes for the LULC changes of the study area include population pressure, cultivated lands expansions, increasing wood collection for fuel, collection of farm implement and construction wood, charcoal production, regime changes, livestock grazing, and interventions (e.g. area closures) and government programs (military camp construction, villagization, land redistribution, and land compensation). These causes are also reported in regional and national documents (e.g. Bureau of Agriculture (BoA), 2012 and 2013; Bureau of Industry and Urban Development (BoIUD), 2012 V and 2012 XI). The population of the study area increased from 53,005 in 1994 (CSA, 1994) to 65,129 in 2007 (CSA, 2007). Based on the data from the District office, the population further increased to 91,690 in 2015, making the population of the study area doubled for the past 20 years. This increased the associated demands on natural resources through expansions of cultivated land, settlements, grazing lands, fuelwood and charcoal production, as it is also reported in earlier studies (e.g. Bishaw, 2001; Zeleke and Hurni, 2001; Tegene, 2002; Barreto, 2006; Kindu et al. 2015; Yesuf et al. 2015). During FGDs, area closures and other forest conservation/ protection activities were described as one of the causes for the changes. BoA (2012) reported limitation to have strong forest protection and development policy and strategy, which resulted in decline of forest resources from time to time. The participants mentioned that to satisfy their demands, unprotected forests resources of the study area were used. It was also understood from the account of the participants that unprotected areas were used without any restriction. Similar trends were also observed in other studies (Wittemyer et al. 2008; Lambin and Meyfroidt, 2011; Barber et al. 2012). For example, Barber et al. (2012) reported reduction of deforestation in protected 15

forests but much larger deforestation rates in the surroundings of unprotected forests in the Brazilian Amazon. Furthermore, BoIUD (2012 V) pointed out desired objectives of conservation efforts were not achieved in most cases due to inappropriate concepts and approaches such as giving priority to natural resource conservation; little focus on human activities and people priority and needs; and neglecting beneficiaries involvement and contribution in the planning and implementation. Regime changes were also mentioned by the FGD participants and key informants as one of the important factors behind the changes in the study area. They described, during the changes, there were loss of forests within which indigenous trees were removed due selective logging for house construction and charcoal making. Schefflera abyssinica, Combretum molle, and species of Acacia were the woody species that were seriously affected. Those events in the absence of firm political control to enforce and maintain rules for protection of common property such as forestlands have also resulted in massive forestland decline elsewhere in the country (Rehamato, 2001; Tekle, 2001; Taddese, 2001; Pankhurst, 2001; Eshetu, 2014; Kindu et al. 2015). Furthermore, according to Meshesha et al. (2014) Ethiopian land tenures system plays a role in land use change and land degradation because the policy transferred land ownership from individual to the government. Farmers do not own land but have only useright and thus they are reluctant to take natural resources conservation measures. The FGD participants also mentioned that programs, such as military camp construction, villagization program in the 1980’s, land redistribution and allocating forest lands as compensation were also responsible for LULC changes in the study area. Extensive deforestations were also reported in other parts of the country through such programs (e.g. Rahmato, 2001; Taddese, 2001; Eshetu, 2014).

16

Neither forest cover increment nor reductions of degraded lands were found as a result of natural conservation measures undertaken in the study area. Instead, our study endorsed the worrying report of Zegeye et al. (2011) who conducted about diversity and regeneration status of woody species in the two state protected forests of the study area, and indicated that forest resources were under great human pressure and will diminish in the near future unless appropriate and immediate measures are taken. 4. Conclusions The evidence obtained through analyses of multi-temporal satellite images using remote sensing application, key informants interview and focus group discussions improved to understand the LULC dynamics and their causes of changes in South Gonder Zone of Ethiopia, where area closures and/or forest conservations have been practiced since 1970 th. The study has found that the area has under gone extensive LULC alterations since 1973. In all of the study periods considered (1973-1985, 1985-1995, 1995-2003, and 2003-2015); the major changes were from forest lands, grasslands and shrub/bush lands to agricultural lands, degraded lands and settlements. The expansion of agricultural lands was massive and rapid during the whole study period, and the conversions were, mainly, from grasslands, forest lands and wetlands. This implies that the natural resource conservation measures that have been practiced by the governmental and non-governmental organizations in the study area did not bring the desired conservation effects. Population pressure, cultivated lands expansions, increasing fuelwood demand, charcoal production, construction materials, farm implement, regime changes, need for grazing lands for livestock, interventions, such as area closures implemented in the area and other programs were the factors identified to underpin the changes in land use and land cover in the study area. Sustainable land use planning and management, proper implementation of participatory 17

forest conservation measures, provision of alternative livelihood strategies for local communities, awareness creation and environmental education on sustainable natural resources management for local communities may reverse the undesired situation in the study area. To make things effective, further studies are recommended on the socio-economic significance of the resource, alternative strategy of reducing the ongoing pressure to the remaining natural resources, and ways of correction measures for future interventions in the study area. Acknowledgment We gratefully acknowledge the International Foundation for Science (IFS) in Stockholm of Sweden,the German Academic Exchange Service (DAAD), Addis Ababa University and Bahir Dar University for their finical support to the first author. We acknowledge the editor and two anonymous reviewers whose comments and suggestions greatly improved the manuscript. References Alelign A, Teketay D, Yemshaw Y, Edwards S., 2007. Diversity and status of regeneration of woody plants on the peninsula of Zegie, northwestern Ethiopia.TropicalEcology 48:37–49. Barber CP, Cochrane MA, Souza C, Veríssimo A., 2012. Dynamic performance assessment of protected areas. Biological Conservation 149:6–14. Barreto P., 2006. Human pressure on the Brazilian Amazon forests: Janice Wiles. Washington, D.C.: World Resources Institute (WRI report). ISBN: 1-56973-605-7. Birhane E, Teketay D, Pia B., 2006. Actual and potential contribution of exclosures to enhance biodiversity of woody species in the drylands of Eastern Tigray. Journal of the Drylands1:134–147.

18

Bishaw B., 2001. Deforestation and Land Degradation on the Ethiopian Highlands: A Strategy for Physical Recovery. International Conference on African Development Archives.Paper 2. Bongers F, Wassie A, Sterck FJ, Bekele T, Teketay D., 2006. Ecological restoration and church forests in northern Ethiopia. Journal of the Drylands 1:35–44. Bray DB, Duran E, Ramos VH, Mas J-F, Velazquez A, McNab RB, Barry D, Radachowsky J., 2008. Tropical Deforestation, Community Forests, and Protected Areas in the Maya Forest. Ecology andSociety 13:56. Bureau of Agriculture (BoA)., 2012. GIS Based Forest Resource Assessment, Quantification and Mapping in Amhara Region, Bahir Dar, Ethiopia (Report). Bureau of Agriculture (BoA)., 2013. Baseline report on land use and land administration. Community-based integrated natural resource management project in Lake Tana Subbasin. Amhara National Regional State Bureau of Agriculture, Bahir Dar, Ethiopia 2:89 Bureau of Industry and Urban Development (BoIUD)., 2012 V. Thematic report (Volume V): land use land cover and assessment of potentials and constraint of resources uses (Resources survey). Potential assessment, identification of opportunities and designing strategic plan for sustainable development of tourism and transport in Lake Tana and its environs. Amhara National Regional State Bureau of Industry and Urban Development, Bahir Dar, Ethiopia. Bureau of Industry and Urban Development (BoIUD)., 2012 XI. Technical report (Volume XI): Proposed land use plan of tourism and transport in Lake Tana and its environs. Potential assessment, identification of opportunities and designing strategic plan for sustainable development of tourism and transport in Lake Tana and its environs. Amhara National Regional State Bureau of Industry and Urban Development, Bahir Dar, Ethiopia.

19

Central Statistical Agency of Ethiopia (CSA), 1994. Population and Housing Census of Ethiopia: Results at a Country Level. Addis Ababa, Ethiopia: Office of population and housing census commission. Central Statistical Agency of Ethiopia (CSA), 2007. Population and Housing Census of Ethiopia: Statistical Report for Amhara Region. Addis Ababa, Ethiopia: Central Statistical Agency of Ethiopia. Central Statistical Agency of Ethiopia (CSA), 2007. Population and Housing Census of Atlas of Ethiopia: Addis Ababa, Ethiopia: Central Statistical Agency of Ethiopia. Congalton RG., Green K., 2009. Assessing the Accuracy of Remotely Sensed Data: Principles and Practices, 2nd ed.; CRC Press/Taylor & Francis: Boca Raton, FL, USA. Costa MH, Botta A, Cardille JA., 2003. Effects of large-scale changes in land cover on the discharge of the Tocantins River, Southeastern Amazonia. Journal of Hydrology 283:206– 217. Desalegn T, Cruz F, Kindu M, Turrión MB, Gonzalo J., 2014.Land-use/land-cover (LULC) change and socioeconomic conditions of local community in the central highlands of Ethiopia. International Journal of Sustainable Development & World Ecology 21:406–413. Eshetu A., 2014. Forest resource management systems in Ethiopia: Historical perspective. International Journal of Biodiversity and Conservation. 6:121–131. Falcucci A, Maiorano L, Boitani L., 2007. Changes in land-use/land-cover patterns in Italy and their implications for biodiversity conservation. LandscapeEcology22:617–631. Foley JA, Defries R, Asner GP, Barford C, Bonan G, Carpenter SR, Chapin FS, Coe MT, Daily GC, Gibbs HK., 2005. Global consequences of land use. Science 309:570–574. ENG. García-Frapolli E, Ayala-Orozco B, Bonilla-Moheno M, Espadas-Manrique C, RamosFernández, G., 2007.Biodiversity conservation, traditional agriculture and ecotourism. 20

Land cover/land use change projections for a natural protected area in the northeastern Yucatan Peninsula, Mexico. Landscape and Urban Planning 83: 137–153. Hailemariam SN, Soromessa T, Teketay D., 2016. Land Use and Land Cover Change in the Bale Mountain Eco-Region of Ethiopia during 1985 to 2015. Land 5:41 Jacobs DF, Oliet JA, Aronson J, Bolte A, Bullock JM, Donoso PJ, Landhäusser SM, Madsen P, Peng S, Rey-Benayas JM, Weber JC., 2015. Restoring forests: What constitutes success in the twenty-first century? New Forests 46:601–614. Kindu M, Schneider T, Teketay D, Knoke T., 2013. Land Use/Land Cover Change Analysis Using Object-Based Classification Approach in Munessa-Shashemene Landscape of the Ethiopian Highlands. Remote Sensing. 5:2411–2435. 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.Environmental monitoring and Assessment187:452. Klenner W, Arsenault A, Brockerhoff EG, Vyse A., 2009. Biodiversity in forest ecosystems and landscapes: A conference to discuss future directions in biodiversity management for sustainable forestry. Forest Ecology and Management258:S1-S4. Lambin EF, Geist H., 2006. Land-Use and Land-Cover Change: Local Processes and Global Impacts. Printed in Germany: Springer-Verlag Berlin Heidelberg. 236 p. ISBN: 1619-2435. Lambin EF, Meyfroidt P., 2011. Global land use change, economic globalization, and the looming land scarcity. Proceedings National Academy of Science,U S A. 108:3465–3472. ENG. Landis J, Koch G., 1977. The measurement of observer agreement for categorical data. Biometrics 33:159–174 Lesschen JP, Verburg PH, Staal SJ., 2005. Land-Use and Land-Cover Change (LUCC) Project. IV. International Human Dimensions Programme on Global Environmental 21

Change (IHDP) V. International Geosphere-Biosphere Programme (IGBP. International Livestock Research Institute, Kenya Li X-Y, Ma Y-J, Xu H-Y, Wang J-H, Zhang D-S., 2009. Impact of land use and land cover change on environmental degradation in lake Qinghai watershed, northeast Qinghai-Tibet Plateau. Land Degradation and Development 20:69–83. LigdiE, Kahloun ME, Meire P., 2010. Ecohydrological status of Lake Tana: a shallow highland lake in the Blue Nile (Abbay) basin in Ethiopia: Review. Ecohydrology & Hydrobiology 10:109–122. Lillesand TM., Kiefer RW., 2000.Remote Sensing and Image Interpretation, 4th ed.; John Wiley & Sons: New York, NY, USA. Lindenmayer DB, Franklin JF, Fischer J., 2006. General management principles and a checklist of strategies to guide forest biodiversity conservation. Biological Conservation 131:433–445. Lu D,Mausel P, Brondízio E, Moran E., 2004. Change detection techniques. International Journal of Remote Sensing 25, 2365–2401. Meshesha DT, Tsunekawa A, Tsubo M, Ali SA, Haregeweyn N., 2014. Land-use change and its socio-environmental impact in Eastern Ethiopia’s highland. Regional Environmental Change 14:757–768. Millennium Ecosystem Assessment (MEA), 2005. Ecosystems and Human Well-being: Biodiversity Synthesis. Island Press, Washington, DC.100 p. Naughton-Treves L, Holland MB, Brandon K., 2005.The role of protected areas in conserving biodiversity and sustaining local livelihoods.Annual Review of Environment and Resources 30:219–252.

22

Nedessa B., Jawad Ali, Ingrid Nyborg., 2005. Exploring Ecological and Socio-Economic Issues for the Improvement of Area Enclosure Management. Norway: Drylands Coordination Group. 63 p. Available from: http://www.drylands-group.org/. Pankhurst A., editor., 2001. Natural resource management in Ethiopia: Proceedings of the workshop organised by Forum for Social Studies in collaboration with the University of Sussex, Addis Ababa. Forum for Social Studies; 9 February 2001; Addis Ababa University of Sussex, translator.200 p. Porter-Bolland L, Ellis EA, Guariguata MR, Ruiz-Mallén I, Negrete-Yankelevich S, ReyesGarcía V., 2012. Community managed forests and forest protected areas: An assessment of their conservation effectiveness across the tropics. Forest Ecology and Management 268:6–17. Rahmato D, editor., 2001. Environmental change and state policy in Ethiopia: lessons from past experience. Forum for Social Studies. Addis Ababa: Addis Ababa University. 121 p. (Series No. 2). Rands MRW, Adams WM, Bennun L, Butchart SHM, Clements A, Coomes D, Entwistle A, Hodge I, Kapos V, Scharlemann JPW., 2010. Biodiversity conservation: challenges beyond 2010. Science 329:1298–1303. Sewnet A., 2015. Land Use/Cover Change at Infraz Watershed, Northwestren Ethiopia. Journal of Landscape Ecology 8:69-83 Taddese G., 2001. Land Degradation: A Challenge to Ethiopia. Environ Manage. 27:815–824. Tadele K, Förch G., 2007. Impact of Land Use/Cover Change on Streamflow: The Case of Hare River Watershed, Ethiopia. Catchment and Lake Research. 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. 23

Tekle K., 2001. Natural regeneration of degraded hillslopes in Southern Wello, Ethiopia: A study based on permanent plots. Applied Geography21:275–300. 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). Journal of Arid Environments91: 53–60. Turner BL II, Meyer WB, Skole DL., 1994. Global Land-Use/Land-Cover Change: Towards an Integrated Study. Ambio 23:91–95. Turner BL II, David S, Steven S, Giinther F., 1995. Land-Use and Land-Cover Change: Science/Research Plan. Stockholm and Geneva: 68 p. Wittemyer G, Elsen P, Bean WT, Burton ACO, BrasharesJS., 2008. Accelerated human population growth at protected area edges. Science 321:123–126. Wondie M, Schneider W, Melesse AM, Teketay D., 2011. Spatial and Temporal Land Cover Changes in the Simen Mountains National Park, a World Heritage Site in Northwestern Ethiopia. Remote Sensing 3:752–766. Yesuf HM, Assen M, Melesse AM, Alamirew T., 2015. Detecting land use/land cover changes in the Lake Hayq (Ethiopia) drainage basin, 1957-2007. Lake and Reservoirs: Research and Management20:1–18. Zegeye H, Teketay D, Kelbessa E., 2011. Diversity and regeneration status of woody species in Tara Gedam and Abebaye forests, northwestern Ethiopia. Journal of Forestry Research 22:315–328. Zeleke G, Hurni H., 2001. Implications of Land Use and Land Cover Dynamics for Mountain Resource Degradation in the Northwestern Ethiopian Highlands.Mountain Research and Development 21:184–191.

24

Highlights 

Analyzing land use/land cover change provided insights into effect of area closure and other conservation measures



Agricultural land is the dominant types



Forest cover declined rapidly



Population increments and associated demand are the causes behind the changes

25