SPIE Startup Challenge 2015 Founding Partner - JENOPTIK Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:

SPIE Photonics West 2017 | Register Today

SPIE Defense + Commercial Sensing 2017 | Call for Papers

Get Down (loaded) - SPIE Journals OPEN ACCESS


Print PageEmail PageView PDF

Remote Sensing

Wetland change detection in the Nile swamps of southern Sudan

Monitoring changes in the Sudd wetland is complicated by remoteness and inaccessibility due to civil war.
15 November 2011, SPIE Newsroom. DOI: 10.1117/2.1201111.003899

The Sudd wetland in the southern Sudan's Nile swamps has valuable socio-ecological functions that fulfill human needs, but they are threatened by overgrazing and loss of vegetation cover due to flooding during the wet season. Remoteness and inaccessibility due to the civil war, which resumed in the southern Sudan in 1983 and ended in 2005, are the major factors affecting field studies on the Sudd wetland. However, these obstacles have protected the Sudd from development so far. Although satellite imagery has been used commonly since the 1980s to improve change detection in wetlands, there are no available studies using remotely sensed data and the post-classification comparison change-detection technique in the Sudd wetland (see Figure 1).1

We monitored the wetland cover changes during the wet season from 1986–2006. Due to the Sudd's immense size, we extracted a sub-scene area of interest in order to minimize data processing, time, and data storage space. We took two main steps using Erdas Imagine geospatial data-authoring software for quantifying wetland cover changes. The first was to classify separately the Landsat imagery in 1986, 1999, and 2006 using maximum-likelihood supervised classifications. We produced thematic maps of five land-cover types—water, vegetation, urban, wetland-vegetation, and wetland-no vegetation. The second step was to compare the images using the post-classification comparison change-detection technique, pixel-by-pixel, to produce change-detection maps of the wetland-vegetation and wetland-no vegetation areas from 1986–1999 and 1999–2006.

Figure 1. A map of the Sudd wetland (Source: Dumont, 2009).

This study demonstrates a decrease of 3.69% from the 1986 to 1999 change detection map in the wetland-vegetation areas, indicating that 19577.61 hectares of those areas were transformed into any of the remaining four land-cover classes. The decline in the wetland vegetation areas by the end of November can be attributed mainly to a significant increase of 28449.72 hectares in the vegetation areas from 1986–1999. Similarly, the 1986–1999 change-detection map showed a decrease of 2.68% in the wetland-no vegetation areas, which can be mainly attributed to a significant increase in the wetland-vegetation and vegetation areas by 8376.93 hectares and 4246.11 hectares, respectively. According to the 1999–2006 change-detection map, the wetland vegetation swamp increased a dramatic 14.95% due to a significant decrease of 46311.84 hectares in the vegetation areas in October of the rainy season. The wetland-vegetation expansion associated with vegetation changes can be attributed to increased precipitation and flooding in the Sudd wetland.2 By contrast, the wetland-no vegetation area decreased by 5.18% in the same period due to the conversion toward wetland-vegetation and urban areas, which increased significantly by 15169.14 hectares and 15007.14 hectares, respectively.

The significant increase of the urban area is from high evaporation throughout the entire Sudd plain due to soil moisture (wetland-no vegetation) saturation during the rainy season's peak (August-October), which in turn leads to an increase of the soil moisture dry-down in October.3

Figure 2. (1) Land cover class influencing trend change in wetland-vegetation. (2) and (3) Land cover classes influencing trend changes in wetland-no vegetation.

In summary, there is a trend of increases and decreases in the wetland system (wetland-vegetation and wetland-no vegetation) during the wet season in the Sudd wetland. Figure 2 shows the most significant land cover classes affecting the wetland system's trend changes. These results showed conformity with the ecology and regime of the Sudd swamp, which is partially covered with wetland-vegetation. The extent is greatly influenced by flooding and precipitation. Further research could map the Sudd's wetland vegetation communities using Landsat images supplemented by field checks.

Ghada Soliman, Hoda Soussa
Ain Shams University
Cairo, Egypt

Ghada Soliman holds a bachelor's degree in computer engineering from the faculty of engineering at Ain Shams and a master's in computer engineering from the Arab Academy for Science and Technology. She is currently a PhD student in the Deptartment of Environmental Engineering Science in the Institute of Environmental Studies and Research.

Hoda Soussa is a professor in the Water Resources Engineering Irrigation and Hydraulics Dept., Faculty of Engineering.

1. H. J. Dumont ed., The Nile: Origin, Environments, Limnology and Human Use (Monographiae Biologicae), pp. 193, Springer Science + Business Media B.V., Dordrecht, 2009.
2. A. El-Moghraby, F. A. Sinada, K. M. Riak, Information Sheet on Ramsar Wetlands (RIS), Ramsar Information Sheet, Khartoum, Sudan, 2006.
3. Y. A. Mohamed, W. G. M. Bastiaanssen, H. H. G. Savenije, Spatial variability of evaporation and moisture storage in the swamps of the upper Nile studied by remote sensing techniques, J. Hydrology 289, pp. 145-164, 2004.