Oasis classification

Objectives: Develop a typology of oases in northern Oman

Contact Person: Prof. Dr. Andreas Bürkert, Institute of Crop Science, University of Kassel, Germany

Email: tropcrops@uni-kassel.de

Scientific staff: Dr. Eike Luedeling and Dr. Stefan Siebert

Study area: Northern Oman

Duration: 2005-2007      

Improving the SRTM model

Many environmental studies on the landscape level require topographic information, which is often derived from digital elevation models (DEMs). For the Oman Mountains, such DEMs are scarce, and their quality is relatively low. The best model available to date is the DEM derived from the Shuttle Radar Topography Mission (SRTM), which provides a spatial resolution of 90 m with a fairly high vertical accuracy. The method used to create this model, however, was impaired by rough topography, where no topographic information could be obtained. Since one of the defining characteristics of many mountain oases in Oman is their location in rough terrain, oases often lie in data voids, which cover a substantial part of the Oman Mountains. As a preparatory step for landscape ecological studies, these voids first had to be filled.

This project aimed at filling the data voids based on supplemental topographic information derived from digitized Russian military maps. For each data void within the Oman Mountains, points around the edge of the void were extracted, and a Triangular Irregular Network was constructed through these points. A corresponding surface was created for a DEM derived from the Russian maps. After correcting the latter surface for the elevation difference between the two TIN surfaces, the elevation models could seamlessly be merged (Figure 1). The resulting elevation model covered all of the Oman Mountains (Figure 2), and verification by a large number of GPS locations throughout the study area indicated that the accuracy was suitable for large scale ecological studies.

Typology of oases in Oman

Northern Oman contains a large number of oases, but the exact number of such settlements is unknown. Research on these oases is also hindered by the lack of a classification system for these oases. This knowledge gap was addressed by deriving a typology of oases in Oman based on a digital elevation model, Landsat images and geological survey data.
Oases were detected by calculating a vegetation index from multispectral Landsat satellite image and, applying correction procedures for season, altitude, natural vegetation and vegetation patch size (Figure 3).This procedure resulted in the detection of 2663 oases, including 2428 oases with cultivated areas above 0.4 ha.

These oases were then classified by examining their topographic, hydrologic and geological settings. This process indicated that there were six major types of oases. The most common type was the Plain Oasis, which obtained its irrigation water using modern pumping technologies (Figure 4). The second most common type was the Foothill Oasis, located in topographic settings, where drainage water was naturally channeled and accessible to oasis farmers by tapping underground wadi flows. Much less frequent were Mountain Oases and Kawr Oases, which relied for their water supply on natural springs emerging from limestone formations. Drainage Oases were similar to Foothill Oases, but had substantially larger orchard areas due to their location along one of the major wadis in northern Oman. Urban Oases, finally, did not lie in hydrologically meaningful settings. These were mostly parks or sports facilities, which relied on off-site water (Figure 4).



Luedeling, E., Siebert, S. & Buerkert. A. 2007. Filling the voids in the SRTM elevation model – A TIN-based delta surface approach. ISPRS Journal of Photogrammetry & Remote Sensing 62, 283--294.



Luedeling, E. & Buerkert, A. 2007. Classification of oases in northern Oman. Pride. Al Roya Press and Publishing House, Muscat, Oman, pp. 84-87. [Arabic version pp. 86-92].



Luedeling, E. &Buerkert, A. 2008. Typology of oases in northern Oman based on Landsat and SRTM imagery and geological survey data. Remote Sensing of Environment 112, 1181-1195.