DETECTING DESERTIFICATION PROCESSES USING TM AND ETM+ DATA, NORTH OF ISFAHAN, IRAN

doi:10.3773/j.issn.1005-264x.2008.02.009

Abstract

Abstract:

Aims Desertification results in ecological and biological diminution of the earth, and can happen naturally or cause by anthropogenic activities. This process especially affects arid and semi-arid regions, such as the Isfahan region, where the spread of desertification is reaching critical proportions. The aim of this study is to use remotely sensed data to review the trend of desertification in the northern of Isfahan, Iran.
Methods Multi-temporal images were employed to evaluate the trend of desertification, specifically the TM and ETM⁺ data of September, 1990 and September, 2001. Geometric and radiometric corrections were applied to each image prior to image processing and supervised classification, and vegetation indices were applied to produce a land use map of each image in nine classes. The land use classifications in the two map images were compared and changes between land use classes were detected over the 11 year period using a fuzzy and post-classification technique.
Important findings The maps and their comparison with false color composite images showed the differences efficiently. With the fuzzy and post-classification method the land use changes were sited on the map. Fuzzy confirmed 53% changed area and 47% unchanged areas in the study region. The results verify the desertification expansion in the study areas. Because of poor land management, agricultural lands converted to desert and abandoned areas, and some marginal pasture lands had to be changed to agricultural land which are desertification spreading according to United Nations Conference on Desertification (UNCOD). Also farmland and pastures have been converted to urban and industrial areas, and the rangelands have been spoiled due to opencast mine excavations. With the mine margins eroding as well as their debris accumulating on the pasture lands, desertification has become worse. Three areas of less-elevated mountains have remained unchanged. This study confirmed that the anthropogenic activities accelerated the desertification process and severely endangered the remaining areas.

Key words: desertification, remote sensing, Isfahan, Iran, fuzzy logic, post classification

Khajeddin SJ, Akbari M, Karimzadeh HR, Eghbal MK. DETECTING DESERTIFICATION PROCESSES USING TM AND ETM⁺ DATA, NORTH OF ISFAHAN, IRAN[J]. Chin J Plant Ecol, 2008, 32(2): 328-335.

Add to citation manager EndNote|Ris|BibTeX

URL: https://www.plant-ecology.com/EN/10.3773/j.issn.1005-264x.2008.02.009

https://www.plant-ecology.com/EN/Y2008/V32/I2/328

Figures/Tables 7

References 18

[1]	Alrababah MA, Alhamid MN (2006). Land use/cover classification of arid and semi-arid Mediterranean landscapes using Landsat ETM. International Journal of Remote Sensing, 27, 2703-2718.
[2]	Babaev AM, Kharin NG (1999). The monitoring and forecast of desertification processes. In: Babaev AM ed. Desert Problems and Desertification in Central Asia. Springer-Verlag, Berlin, 59-75.
[3]	Chen H, Goodenough DG, Dyk A, Mcdonald S, Han T (2004). Change detection with ALI and Landsat satellite data. In: Smits P, Bruzzone L eds. Analysis of Multi-Temporal Remote Sensing Images. World Scientific, London, 89-97.
[4]	Dhir RP (1993). Remote sensing in monitoring drought and desertification. In: Sen AK, Kar A eds. Desertification and Its Control in the Thar, Sahara and Sahel Region. Scientific Publishers, Jodhpur, India, 107-118.
[5]	Ghassemian-Yazdei MH, Ghyasvand G (1999). Change Detection in TM Images Using Principle Component Analysis and Fuzzy Logic. Proceeding of Surveying, 78 Conference, National Cartographic Center, Tehran, Iran, 1-15. (in Persian)
[6]	Goetz SJ, Varlyguin D, Smith AJ, Wright RK, Prince SD, Mazzacato ME, Tringe J, Jantz C, Melchoir B (2004). Application of multi-temporal Landsat data to map and monitor land cover and land use change in the Chesapeake Bay watershed. In: Smits PC, Bruzzone L eds. Analysis of Multi-Temporal Remote Sensing Images. World Scientific, London, 223-232.
[7]	Granger A (1984). Desertification, How People Make Deserts, How People Can Stop and Why They Don't. International Institute for Environmental and Development, London, 94.
[8]	Harasheh H, Tateishi R (2000). Desertification mapping of west Asia—a GIS and remote sensing application. http://www.gis.development.net/aars/acrs/.
[9]	Kral K, Pavlis J (2006). The first detailed land-cover map of Scotra Island by Landsat/ETM⁺ data. International Journal of Remote Sensing, 27, 3239-3250.
[10]	Lillesand TM, Kiefer RW (2000). Remote Sensing and Image Interpretation. John Wiley, New York, 470-615.
[11]	Probeck M, Ludwig R, Mauser W (2004). Spectral un-mixing of sub-scale land cover from multi-temporal NOAA-AVHRR imagery using a combined GIS and fuzzy logic approach. In: Smits PC, Bruzzone L eds. Analysis of Multi-Temporal Remote Sensing Images. World Scientific, London, 80-88.
[12]	Rama Krishna YS (1993). Climatological analysis of drought situation in Western Rajasthan. In: Sen AK, Kar A eds. Desertification and Its Control in the Thar, Sahara and Sahel Region. Scientific Publishers, Jodhpur, India, 187-203.
[13]	Rappenhoner D (1989). Resource Conservation and Desertification Control in the Near East. Deutsche Stifung fur Internationale Entwicklung, Feldafing, Germany, 40-157.
[14]	Sen AK (1993). Methodology on the study of monitoring of desertification in India—A Review of some case studies. In: Sen AK, Kar A eds. Desertification and Its Control in the Thar, Sahara and Sahel Region. Scientific Publishers, Jodhpur, India, 187-203.
[15]	Sen AK, Sharma KD (1995). Causative Agents Indicators of Monitoring and Desertification in Asia and Pacific Region. Scientific Publishers, Jodhpur, India, 41-58.
[16]	Williams LO (2002). An End To Global Warming. Pergamon Elsevier Science, Amsterdam, New York.
[17]	Zehtabian GR, Tabatabaei MR (2000). Investigation of Desertification change in Khozestan province by using image processing (IP) and geographic information system (GIS). Biaban, 4, 57-67. (in Persian with English abstract)
[18]	Zhang J, Foody GM (1998). A fuzzy classification of sub-urban land covers from remotely sensed imagery. International Journal of Remote Sensing, 19, 2721-2738.

No.	Land use classes	Symbol	TM (1990)^*	ETM⁺ (2001)^*	Change^*
1	Agriculture	Agr	11 488	10 254	-1 234	-10.74
2	Rangeland 2	R2	9 892	6 139	-3 753	-37.94
3	Rangeland 3	R3	43 378	22 966	-20 412	-47.06
4	Degraded	Deg	17 958	40 907	22 949	127.79
5	Urban	Urb	2 035	4 370	2 335	114.74
6	Debris	Deb	317	432	115	36.27
7	Surface erosion	G	1 024	1 024	-	-
8	Highland	H	337	337	-	-
9	Kooh-e Khan	I	621	621	-	-
	Total area		87 050	87 050	-	-

No.	Land use classes	Symbol	TM (1990)^*	ETM⁺ (2001)^*	Change^*
1	Agriculture	Agr	11 488	10 254	-1 234	-10.74
2	Rangeland 2	R2	9 892	6 139	-3 753	-37.94
3	Rangeland 3	R3	43 378	22 966	-20 412	-47.06
4	Degraded	Deg	17 958	40 907	22 949	127.79
5	Urban	Urb	2 035	4 370	2 335	114.74
6	Debris	Deb	317	432	115	36.27
7	Surface erosion	G	1 024	1 024	-	-
8	Highland	H	337	337	-	-
9	Kooh-e Khan	I	621	621	-	-
	Total area		87 050	87 050	-	-

No.	Land use classes	Accuracy	TM	ETM⁺	Error	TM	ETM⁺
1	Agriculture	Producer	30.6	74.9	Omission	R3, U, Deg	R2, R3
		Users	57.9	99.8	Commission	Deg, U	-
2	Rangeland 2	Producer	47.5	98.7	Omission	-	-
		Users	46.2	81.6	Commission	R3	Agr
3	Rangeland 3	Producer	70.6	96.4	Omission	R2, Deg	-
		Users	31.6	93.5	Commission	Deg, U, Agr	-
4	Degraded	Producer	46.2	90.2	Omission	R3, Agr	-
		Users	65.4	94.9	Commission	G, R3	-
5	Urban	Producer	49.1	93.7	Omission	R3	-
		Users	84.2	99.8	Commission	Agr	-
6	Debris	Producer	77	100	Omission	R3	-
		Users	90.7	100	Commission	-	-
7	Surface erosion	Producer	87.3	100	Omission	Deg	-
		Users	100	96.3	Commission	-	-
8	Highland	Producer	100	87.2	Omission	-	R2
		Users	100	97.1	Commission	-	-
9	Kooh-e Khan	Producer	100	94.9	Omission	-	-
		Users	100	97.1	Commission	-	-
		Overall	60.4	91.9
		KHAT	52.6	89.8

No.	Land use classes	Accuracy	TM	ETM⁺	Error	TM	ETM⁺
1	Agriculture	Producer	30.6	74.9	Omission	R3, U, Deg	R2, R3
		Users	57.9	99.8	Commission	Deg, U	-
2	Rangeland 2	Producer	47.5	98.7	Omission	-	-
		Users	46.2	81.6	Commission	R3	Agr
3	Rangeland 3	Producer	70.6	96.4	Omission	R2, Deg	-
		Users	31.6	93.5	Commission	Deg, U, Agr	-
4	Degraded	Producer	46.2	90.2	Omission	R3, Agr	-
		Users	65.4	94.9	Commission	G, R3	-
5	Urban	Producer	49.1	93.7	Omission	R3	-
		Users	84.2	99.8	Commission	Agr	-
6	Debris	Producer	77	100	Omission	R3	-
		Users	90.7	100	Commission	-	-
7	Surface erosion	Producer	87.3	100	Omission	Deg	-
		Users	100	96.3	Commission	-	-
8	Highland	Producer	100	87.2	Omission	-	R2
		Users	100	97.1	Commission	-	-
9	Kooh-e Khan	Producer	100	94.9	Omission	-	-
		Users	100	97.1	Commission	-	-
		Overall	60.4	91.9
		KHAT	52.6	89.8

Type of changes	Fuzzy analysis			Time series analysis
Type of changes	(%)	(hm²)		(%)	(hm²)
Unchanged areas	47%		41 244	45%	38 817
Changed areas	53%		45 806	55%	48 233
Total	100%		87 050	100%	87 050