Sand Dune Encroachment and Desertification Processes of the Rigboland Sand Sea, Central Iran

Abstract

Early studies on sand dune movement and desertification in Iran have not always been convincingly demonstrated because of problems with the field-based measurements. In some areas where various land uses have been engulfed by aeolian sand dunes, desertification is clear, but in other less settled areas, it may not be so obvious. The objective of this study is to demonstrate encroachments of the Rigboland sand sea, central Iran, in its different directions and variable magnitude rates. Determining the rate and direction of the sand sea movements is critical for specifying which lands should be prioritized and quickly protected. The study has trialed a change detection technique which uses a Cross-Tabulation module to compare two available Landsat_TM images over the Rigboland sand sea. This indicates that within a ten-year span (from 1988 to 1998) more than 200 ha/yr were added to the Rigboland sand sea, from the alluvial fan landforms in the eastern upstream, outer margins of the Rigboland sand sea. Coupled with GIS techniques, this type of analysis of the remote sensing (RS) images provides an effective tool for the monitoring and prognostication of sand dune movement and sand sea change.

Figure 1

The location of the Rigboland sand sea shown by yellow rectangle in centre of Iran (Image from Google Earth Software).

Figure 2

The confines of Rigboland sand sea with brown color in centre of Iran (ENVI software version 5.1).

Figure 3

The Landsat_TM image of the Rigboland sand sea and surrounding areas with geometry and atmospheric corrections plus radiometric calibration (ENVI software version 5.1).

Figure 4

(A) Location (shown as black circles) of 250 ground control points (GCPs) for accuracy assessment of classification of the geomorphological features and land uses in the Rigboland sand sea. (B) 1988 Landsat_TM image classification through maximum likelihood classifier and (C) 1998 Landsat_TM image classification through maximum likelihood classifier. Comparison of the black shaded areas show changes of the Rigboland sand sea over the 10 years via image classification (IDRISI software version Kilimanjaro).

Figure 5

Changes detected between two Landsat_TM images using the cross-tabulation module. Red represents the Rigboland sand sea area without change. The yellow and green show areas added to and diminished from the Rigboland sand sea from other geomorphological features and land uses (i.e. residential areas, industrial parks, agricultural and orchard lands, rangelands) respectively over the period from 1988 to 1998 (Arc GIS software version 10.2).

Figure 6

The location of five weather stations around the Rigboland sand sea. The wind roses show the annual distribution and magnitude of wind direction and speed (Image from Google Earth Software).

Figure 7

Landsat_TM image showing alluvial fan regions (red rectangle) and direction of sand supply (blue arrow) supplying sand to the expanding eastern margins of the Rigboland sand sea. Also shown are petroleum mulch-sprayed sites marginal to Kashan city, designed to prevent sand dunes from encroaching (ENVI software version 5.1).