Scholarly article on topic 'Land-use/land-cover dynamics in Sego Irrigation Farm, southern Ethiopia: A comparison of temporal soil salinization using geospatial tools'

Land-use/land-cover dynamics in Sego Irrigation Farm, southern Ethiopia: A comparison of temporal soil salinization using geospatial tools Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Keywords
{"Barren lands" / "Environmental impacts" / "Geographic information system" / Irrigation / "Land-use patterns" / "Remote sensing"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Shegena Zewdu, K.V. Suryabhagavan, M. Balakrishnan

Abstract Land-use patterns are changing fast in most of the tropical nations in relation to the human population growth and related impacts. Majority of the rural population in Ethiopia depends on agriculture, and hence the land-use changes during the past couple of decades in rural Ethiopia are mostly linked to agricultural developments. The present study deals with the status and trends of land-use and land-cover dynamics in Sego Irrigation Farm in southern Ethiopia. Geospatial tools were used to assess changes in land-use/land-cover patterns in the study area during 1984–2010. Patch dynamics was assessed to understand the degree of fragmentation and changes along the terrain topography. Detailed analyses have revealed that the extent of cultivated land, which was 38.1% in 1984 has increased to 60.7% by 2010, with an average change of 58ha per year. The extent of land, which was intensively and sparsely cultivated in 1984 and 1995, was converted to barren and fallow land due to irrigation-related salinization problems. The water body/swamp, which was 55ha in 1984 has significantly decreased to 2ha by 2010. Land-use changes have been attributed to factors such as population pressure and environmental changes as more land area was put under irrigated cultivation, leading to salinization and lowering productivity of the soils in the area. Findings of the present study have implications for other rural areas in Ethiopia and elsewhere in the tropical regions, where irrigated agriculture is practiced.

Academic research paper on topic "Land-use/land-cover dynamics in Sego Irrigation Farm, southern Ethiopia: A comparison of temporal soil salinization using geospatial tools"

Journal of the Saudi Society of Agricultural Sciences (2014) xxx, xxx-xxx

King Saud University Journal of the Saudi Society of Agricultural Sciences

www.ksu.edu.sa www.sciencedirect.com

n irljJII pq__lflU n IifjoIilB nipp^ll

SAUDI SOCIEIY FOB AGRICULTURAL SCIENCES

FULL LENGTH ARTICLE

Land-use/land-cover dynamics in Sego Irrigation Farm, southern Ethiopia: A comparison of temporal soil salinization using geospatial tools

Shegena Zewdu a, K.V. Suryabhagavan a,% M. Balakrishnan b

a School of Earth Sciences, Addis Ababa University, P.O. Box No. 1176, Addis Ababa, Ethiopia b Department of Zoological Sciences, Addis Ababa University, P.O. Box No. 1176, Addis Ababa, Ethiopia

Received 21 November 2013; revised 24 March 2014; accepted 29 March 2014

KEYWORDS

Barren lands; Environmental impacts; Geographic information system; Irrigation; Land-use patterns; Remote sensing

Abstract Land-use patterns are changing fast in most of the tropical nations in relation to the human population growth and related impacts. Majority of the rural population in Ethiopia depends on agriculture, and hence the land-use changes during the past couple of decades in rural Ethiopia are mostly linked to agricultural developments. The present study deals with the status and trends of land-use and land-cover dynamics in Sego Irrigation Farm in southern Ethiopia. Geospatial tools were used to assess changes in land-use/land-cover patterns in the study area during 1984-2010. Patch dynamics was assessed to understand the degree of fragmentation and changes along the terrain topography. Detailed analyses have revealed that the extent of cultivated land, which was 38.1% in 1984 has increased to 60.7% by 2010, with an average change of 58 ha per year. The extent of land, which was intensively and sparsely cultivated in 1984 and 1995, was converted to barren and fallow land due to irrigation-related salinization problems. The water body/ swamp, which was 55 ha in 1984 has significantly decreased to 2 ha by 2010. Land-use changes have been attributed to factors such as population pressure and environmental changes as more land area was put under irrigated cultivation, leading to salinization and lowering productivity of the soils in the area. Findings of the present study have implications for other rural areas in Ethiopia and elsewhere in the tropical regions, where irrigated agriculture is practiced.

© 2014 King Saud University and Saudi Society of Agricultural Sciences. Production and hosting by

Elsevier B.V. All rights reserved.

* Corresponding author. Tel.: +251 911998588. E-mail address: drsuryabhagavan@gmail.com (K.V. Suryabhagavan). Peer review under responsibility of King Saud University.

^jjfl I

Elsevier I Production and hosting by Elsevier

1658-077X © 2014 King Saud University and Saudi Society of Agricultural Sciences. Production and hosting by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jssas.2014.03.003

1. Introduction

Like many other developing countries across the globe, significant land-cover changes have occurred in Ethiopia since the last century. These changes were primarily due to anthropogenic activities, in connection with the population increase, which forced people to clear forest for cultivation and other activities (Gebresamuel et al., 2010). In the present study area, irrigation has been in practice since over four decades. Significant land-cover and land-use changes occurred in the area in response to the increase of soil salinity from time to time affecting agricultural productivity leading to land-use changes. Accurate information on land-cover changes and the forces behind are essential for designing sound environmental programmes and management activities. Land-cover analysis provides the baseline data required for proper understanding on the land-use patterns in the past and its impacts. It also aids to understand the ratio of the past land-cover changes, and the physical and socio-economic factors behind (Mengistu and Salami, 2007).

Studies on land-cover changes also yield valuable information for analysis of effects of climate change and environmental impacts of human activities. Such information is of great use for resource managers to help in resolving conflicts between human use of natural resources and sustainability of natural habitats and ecosystems (Belay, 2002). A change in the land-cover of an area can negatively affect the potential characteristics of the area, and may ultimately lead to degradation and loss of productivity.

Soil salinization is one of the major impacts of land-use/ land-cover (LULC) changes in arid and semi-arid regions, wherever irrigation is practiced. Soil degradation due to salinity and sodicity is increasing at an alarming rate of endangering the environment and agricultural ecosystems (Reza et al., 2008). It is a severe environmental hazard that impacts the growth of many crops. Even though salt affected areas are extended through all continents, statistical data on the extent of salt affected areas vary. However, general estimates show the extent of such areas as close to one billion hectares, which represents about 7% of the extent of the terrestrial area of earth. Salt affected areas are primarily located in the irrigated areas of the old alluvial plains and zones of low rainfall in shallow water table depth and in hot and dry moisture regions (Mandal and Sharma, 2005). More than 120 countries are directly affected by the problem of soil salinity (Al-Khaier, 2003).

Land degradation, which is a product of complex interactions of many of the physical and biological variables, reduces the potential capability of soil to produce goods and services. Semi-arid regions are under high pressure to supply the required food for their rapidly increasing populations. Consequent changes in the land-use patterns due to agricultural intensification, together with the harsh climatic conditions including global climate change, have accelerated land degradation processes, with yield reduction in many parts of the arid world. Therefore, the need for detecting the occurrence of land degradation while assessing its severity at any given time becomes vital. Hence, the present study was aimed to detect temporal and spatial dynamics of LULC and soil salinity, which are major driving forces of land degradation.

2. The study area and methods

The present study area is situated in the Southern Nations Nationalities and Peoples Regional State (SNNPRS) of Ethiopia in Arba Minch Zuria Woreda (Gamo Gofa Zone), at about 24 km in the south of Arba Minch town, extended on both sides of Arba Minch - Konso asphalt road. Geographically, the study area is bounded by longitude 37°25'9"-37°30'50" E and latitude 5o48'12''-5o54'9'' N with an area of 6747 ha (Fig. 1). This area is a flood plain land bounded by Lake Chamo in the eastern and southern directions, while its northern and western directions are bounded by mountains. The area is drained by Sile and Sego Rivers, which drains into Lake Chamo.

2.1. Topography and climate

There is an abrupt change of topography where Sego and Sile rivers emerge into a plain, bordering Lake Chamo (at the command area). This is the lowest part of the Great Ethiopian Rift Valley in the area at an elevation of 1100-1250 m asl, composed of recent superficial deposits with no rock exposure, and with very shallow gradients sloping toward Lake Chamo.

The study area is characterized by hot sub-humid lowlands in the eastern and northern parts and warm sub-humid to warm humid lowlands in the west and northwestern parts of the watershed. Its evapotranspiration is almost uniformly distributed throughout the year. Seasonal temperature variation is very small. At Arba Minch, the township close to the study area, the mean monthly temperature is 23.9 0C, varying between 22.7 0C (July) to 25.7 0C (March). Rainfall distribution in the study area is bimodal with a long rainy season from the beginning of March to the end of May with maximum rainfall around the month of April (228 mm), and a short rainy season from mid-August to mid-October. The minimum monthly rainfall is recorded in January (18 mm).

2.2. Materials and methods

Landsat satellite images of 1984, 1995 and 2010 (path and row of 169/056) were used to analyze temporal changes in the study area. The Ethiopian Mapping Agency (EMA) toposheet (0537 A2) of 1:50,000 scale pertaining to the study area was used for geo-referencing of satellite images. Landsat satellite data provided by Global Land Cover Network (GLCN) was radiomet-rically and geometrically processed using ERDAS Imagine 9.2. (Ortho-rectification with Adindin UTM Zone 37 N).

A pre-field interpretation map was prepared and the ground truth of the area was verified by direct observation. The Ground truth information was collected by GPS survey by 85 GCP points for the entire study area. Recent satellite data were used during the field survey and the current land-use practices were noted in the field. The past LCLU information was gathered by informal interviews with the local people and concerned government departments. The GPS points were downloaded and overlaid on the imagery and used for refinement of the pre-field interpreted LULC map.

For LULC mapping, visual interpretation technique was adopted. With the extensive ground truth and detailed information, an interpretation key was developed. The mapping

exercise was carried out first on the datasets of the year 2010. Same vector layer was overlaid upon the 1984 and 1995 datasets, and the polygons were modified wherever changes were found. In the process of classification, training area was given to each of the land-cover class, based on the reflectance signature of different features for the band combinations of false color composite 4-3-2 (FCC). Maximum likelihood classifier technique was used for image classification, and a 3 x 3 pixel moving windows majority filter was employed to smoothen the classification. Supervised land-cover classes were exported to ArcGIS 9.2, and LULC maps were prepared for the years 1984, 1995 and 2010.

Normalized Difference Salinity Index (NDSI) for Landsat image of 1984, 1995 and 2010 was analyzed for temporal and spatial detection of salt affected areas. Soil salinity index for the three Landsat images were executed, following NDSI. The salt affected area was extracted by NDSI (Khan et al., 2005). This index was derived by dividing differences of red and near infra red to their sum. The mathematical formula for NDSI was:

NDSI = [(Band3 - Band4)/(Band3 + Band4)]

The value of NDSI ranges between 1 and -1. Those areas with values 0 and <0 were classified as non-saline and areas with NDSI values >0 were classified as slightly saline to strongly saline, which have a strong correlation with the status of LULC.

3. Results

Results from supervised classification of the study area have revealed that the extent of the areas under intensive cultivation has decreased from 40.9% to about 18.4% during the period 1984-2010 with an average rate of 3.1% per year (Table 1). Most of the previously intensively cultivated areas was changed to sparse cultivation due to salinity developed as a result of irrigation (Table 2). There was a slight decrease in the extent of intensive cultivation during the period 1984-1995, but greater decrease was observed during 1995-2010. Water body along the shore of Lake Chamo, which was about 55 ha in

Table 1 Land-cover class and rate of changes in the study area for the years 1984, 1995 and 2010.

Land-use/land-cover class Years Change/year (1984-2010)

1984 1995 2010

Area (ha) Area (%) Area (ha) Area (%) Area (ha) Area (%) Change (ha) Change (%)

Intense cultivation 2760 40.9 2511 37.2 1244 18.4 -58 -3.1

Dispersed cultivation 2572 38.1 3451 51.1 4094 60.7 59 1.8

Bare land 1211 17.9 405 6.0 911 13.5 -12 -1.1

Fallow land 148 2.2 379 5.6 495 7.3 13 4.6

Water body/swamp 55 0.8 1 0.0 2 0.0 -2 -13.7

Total 6747 100.0 6747 100.0 6747 100.0 - -

Figure 1 Location map of the study area.

Table 2 Proportion of land-cover converted from each class in the study area during 1984-2010.

LULC Classes Land-cover type of 2010 (ha) Total

Intense cultivation Dispersed cultivation Bare land Fallow land Water body

Land cover type Intense cultivation 660.6 1472.1 375.5 261.2 0.0 2769.3

of 1984 (ha) Dispersed cultivation 446.5 1593.3 358.4 165.5 0.6 2564.3

Bare land 100.8 890.5 168.4 50.6 0.3 1210.7

Fallow land 23.5 96.6 9.4 17.5 0.6 147.6

Water body 17.5 36.4 0.0 1.1 0.0 55.0

Total 1248.8 4088.9 911.7 495.8 1.6 6746.9

333000

Figure 2 Land-use/land-cover maps of 1984, 1995 and 2010.

1984 had disappeared, and swamps emerged in the middle of the farm during 1995-2010. Area under fallow was about 2.2% in 1984, which was increased to 5.6% by 1995 and further increased to 7.3% by 2010 (Fig. 2).

From the analysis of Landsat images, five land classes such as non-saline (including areas of vegetation and crops), slightly saline, moderately saline, strongly saline and water body (Fig. 3) were identified. About 64.9% of the extent of the area was non-saline in 1984, which gradually decreased to 40% by 2010 (Table 3). Out of the 4376 ha under non-saline status in 1984, only 1976 ha remained as non-saline, whereas the rest 2400 ha was changed to slightly saline, moderately saline and strongly saline (Table 4). Large area change was observed in the strongly and moderately saline soils. Moderately saline

area was about 7.6% in 1984, which has increased to 13.6% and 22.0% in 1995 and 2010, respectively. This shows an average rate of 4.1% per year in the 26 year study period. On the other hand, strongly saline area was increased at the rate of 5.5% per year (6 ha/year). A large area change was observed during the period 1995-2010. Map of the salt affected area generated from indices analysis for the three different years has revealed that moderately and strongly saline areas were concentrated in the central and largely in the eastern part of the study area, along the sides of Lake Chamo, where bare lands, fallow lands and sparsely cultivated lands were observed. Water body (part of Lake Chamo) accounting to 53 ha present in the image of 1984 was not seen in the images of 1995 and 2010.

327000 330000 333000

Figure 3 Salinity affected area map of the years 1984, 1995 and 2010.

Table 3 Salinity class derived from NDSI and the rate of change for 1984, 1995 and 2010.

Salinity class Area change Change/year (1984-2010)

1984 1995 2010

(ha) (%) (ha) (%) (ha) (%) (ha) (%)

Non-saline 4376 64.9 3979 59.0 2702 40.0 -64 -1.9

Slightly saline 1755 26.0 1777 26.3 2360 35.0 23 1.1

Moderately saline 514 7.6 916 13.6 1485 22.0 37 4.1

Strongly saline 48 0.7 75 1.1 200 3.0 6 5.5

Water body 53 0.8 0 0.0 0 0.0 -2

Total 6747 100.0 6747 100.0 6747 100.0 - -

Table 4 Proportion of salinity converted from each class to the rest during 1984-2010.

Salinity class Salinity in 2010 (ha)

Non-saline Slightly saline Moderately saline Strongly saline Total

Salinity in 1984 (ha) Non-saline 1976.1 1395.3 893.7 110.5 4375.5

Slightly saline 556.7 695.8 439.7 63.9 1756.2

Moderately saline 136.7 225.7 133.0 20.2 515.6

Strongly saline 12.9 18.4 13.3 1.8 46.4

Water body 25.7 21.8 5.0 0.5 53.1

Total 2708.2 2357.1 1484.7 196.9 6746.9

4. Discussion

Unsustainable agricultural practices along with many other physical, socio-economic and political factors have been the driving forces of land degradation. Land-use changes especially cultivation in deforested and in unsuitable lands may rapidly diminish the soil quality, as ecologically sensitive components of the habitats are not able to buffer the adverse effects. As a result, severe deterioration of the soil quality may result, leading to a permanent degradation of land productivity, and land degradation increases agricultural costs to maintain soil fertility (Abera and Belachew, 2011; Mojiri et al., 2011). Modern geospatial tools such as remote sensing and GIS are capable of analyzing changes in LULC patterns and its impacts on regional and global scales (Mengistu and Salami, 2007). The present case study has analyzed such changes impacted by the ever growing human population during the past over three decades in one of the least developed nations, where agricultural economy is in the process of fast advancement.

There is a correlation between LULC changes and soil salinization from long term irrigation practices. Land-use and land-cover changes are particularly related to the increase of human population and intensive agriculture (Verburg and Chen, 2000). Kalra et al. (2010) studied salt affected soils of Kotri and Taswaria villages on the basis of IRS LISS III FCC images appeared in bright white to light gray, and smooth texture with white mottles. In case studies from Iran and Egypt, the inclusion of thermal band 6 into selected (best) Landsat TM band combinations improved the separation of saline soils from gypsiferous and coarse textured desert soils, and substantially increased both the overall and the class accuracies (Goossens et al., 1999).

Land-cover modification and conversion are driven by interactions in space and time between biophysical and human dimensions (Skole et al., 1994; Turner et al., 1994, 1995; Misrak et al., 2012). Salinity increase in the soils has made LULC pattern changes in the area, and as a result of increased salinity, the land has become unproductive for farming. Areas, which were under intensive cultivation earlier were abandoned in the recent past and developed as barren, fallow or sparsely cultivated lands due to salinity increase. Most of the non-saline areas in the three images correspond to the intensively cultivated land, whereas the moderately and strongly saline areas correspond to the sparsely cultivated, barren or fallow lands (see, Figs. 2 and 3).

The false color composite was visually interpreted for intensive cultivation, sparse cultivation, crop affected by moderate salinity and crop affected by severe salinity with the help of image elements like tone, texture, pattern and association. Salinity affects any morphological, physiological and biochemical processes, including plant growth and nutrient uptake. During the survey, it was observed that the wheat crop looked burned, and the plant growth was stunted. Water and salt stresses are of particular significance to irrigated crops (Fowden et al., 1993; Willenborg et al., 2004). Areas of salt affected soils were abandoned due to its low productivity as most crops have no tolerance to grow in highly saline soils.

The levels of soil salinization in the study area are increasing from year to year, as the area is under irrigated agriculture for the last three to four decades (Ali et al., 2008). There has been major changes in LULC patterns and soil salinization in the study area. In the future, multi-temporal satellite images should be used for continuous monitoring of farming areas and salinity dynamics in such areas. Integrated analysis of spatial and non-spatial data parameters using geospatial tools must be made and used for decision making.

Acknowledgements

We are thankful to the School of Earth Sciences, Addis Ababa University, for facilities and funds. Generation Integrated Rural Development Consultant (GIRDC) deserves special thanks for their support provided during this study. Sego Irrigation Farm workers are thanked for providing helpful information during the field study.

References

Abera, Y., Belachew, T., 2011. Effects of land use on soil organic carbon and nitrogen in soils of Bale, South Eastern Ethiopia. J. Trop. Subtrop. Agroecosyt. 14, 229-235. Al-Khaier, F., 2003. Soil Salinity Detection Using Satellite Remote Sensing. (M.Sc. thesis). The International Institute for Geo-information Science and Earth Observation, Enschede, Netherlands.

Ali R.A., Sanaeinejad, S.H., Parisa, M.H., Marjan, G., Atefeh, K., 2008. Evaluation of vegetation cover and soil indices for saline land classification in Neyshabour Region using ETM+ Landsat. International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences, Mashhad. Iran.

Belay, T., 2002. Land-cover/land-use changes in the Derekolli Catchment of the South Welo Zone of Amhara Region, Ethiopia. East. Afr. Soc. Sci. Res. Rev. 18, 1-20. Fowden, L., Mansfield, T., Stoddart, J., 1993. Plant Adaptation to

Environmental Stress. Chapman & Hall, London, U.K. Gebresamuel, G., Singh, B.R., Dick, O., 2010. Land-use changes and their impacts on soil degradation and surface runoff of two catchments of Northern Ethiopia. Acta Agric. Scand, Section B Plant Soil Sci. 60, 211-226. Goossens, R., Alavi, P.S.K., De Dapper, M., Kissyar, O., 1999. The use of thermal band of Landsat TM for the study of soil salinity in Iran (Ardakan area) and Egypt (Ismailia Province). In: Proceedings International Conference on Geoinformatics for Natural Resource Assessment, Monitoring and Management. Indian Institute of Remote Sensing, Dehradun, India. Kalra, N.K., Singh, L., Kachhwah, R., Joshi, D.C., 2010. Remote sensing and GIS in identification of soil constraints for sustainable

development in Bhilwara District, Rajasthan. J. Indian Soc. Remote Sens. 38, 279-290.

Khan, M.N., Rastoskuev, V.V., Sato, Y., Shiozawa, S., 2005. Assessment of hydrosaline land degradation by using a simple approach of remote sensing indicators. Agric. Water Manage. 77, 96-109.

Mandal, A.K., Sharma, R.C., 2005. Computerization database on salt affected soils of Haryana State. J. Indian Soc. Remote Sens. 33, 447-455.

Mengistu, D., Salami, A., 2007. Application of remote sensing and GIS in land use/land cover mapping and change detection in a part of south western Nigeria. Afr. J. Environ. Sci. Technol. 1, 99-109.

Misrak, A., Suryabhagavan, K.V., Balakrishnan, M., 2012. Assessment of cover change in the Harenna habitats in Bale Mountains, Ethiopia using GIS and remote sensing. Int. J. Ecol. Environ. Sci. 38, 39-45.

Mojiri, A., Emami, N., Ghafari, N., 2011. Effects of land use changes on soil quality attributes (a review). Res. J. Agri. Biol. Sci. 7, 1-3.

Reza, A., Sanaeinejad, S. H., Parisa, M.H., Marjan, G., Atefeh, K., 2008. Evaluation of vegetation cover and soil indices for saline land classification in Neyshabour Region using ETM+ Landsat. International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences, Mashhad, Iran.

Skole, D.L., Chomentowski, W.H., Salas, W.A., Nobre, A.D., 1994. Physical and human dimensions of deforestation in Amazonia. Bioscience 44, 314-322.

Turner II, B.L., William, B.M., Ross, R.H., Skole, D.L., 1994. Global land-use and global land-cover change: towards an integrated study. Ambio 23, 91-95.

Turner, B.L., Skole, D., Sanderson, S., Fischer, G., Fresco, L., Leemans, R., 1995. Land-Use and Land-cover change Science/ Research Plan. Joint publication of the International Geosphere-Biosphere Programme (Report No. 35) and the Human Dimensions of Global Environmental Change Programme (Report No. 7). Royal Swedish Academy of Sciences. Stockholm, Sweden.

Verburg, P.H., Chen, Y.Q., 2000. Multiscale characterization of land-use patterns in China. Ecosystems 3, 369-385.

Willenborg, C.J., Gulden, R.H., Johnson, E.N., Shirtliffe, S.J., 2004. Germination characteristics of polymer-coated canola (Brassica napus L.) seeds subjected to moisture stress at different temperatures. Agron J. 96, 786-791.