Monthly flooded area classification using low resolution SAR imagery in the Sudd wetland from 2007 to 2011


Wilusz D. C., Zaitchik B. F., Anderson M. C., Hain C. R., YILMAZ M. T., Mladenova I. E.

REMOTE SENSING OF ENVIRONMENT, cilt.194, ss.205-218, 2017 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 194
  • Basım Tarihi: 2017
  • Doi Numarası: 10.1016/j.rse.2017.03.005
  • Dergi Adı: REMOTE SENSING OF ENVIRONMENT
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.205-218
  • Anahtar Kelimeler: Synthetic aperture radar, Wetland classification, Flood detection, Sudd wetland, Nile River basin, ENVISAT ASAR, ALEXI, SYNTHETIC-APERTURE RADAR, REMOTE-SENSING DATA, FORESTED WETLANDS, SOUTHERN SUDAN, SOIL-MOISTURE, VARIABILITY, HYDROLOGY, FLOODPLAIN, VEGETATION, PATTERNS
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

The annual flood cycle of the Sudd wetland in South Sudan plays an important role in the Nile River Basin water balance. The wetland, however, is extensive and sparsely instrumented, which has inhibited credible understanding of regional flooding across space and time. Here we explore the potential to apply low resolution C-band ENVISAT Advanced SAR imagery for remote estimation of Sudd flooded area. Over a five year study period (2007-2011) the time-averaged flooded Sudd area was found to be 18,033 km(2) with an average annual high of 29,702 km(2) in late September and a low of 10,128 km(2) in early May. Annual peak flood area ranges considerably from 19,259 km(2) in 2009 to 36,649 km(2) in 2007, but we found no systematic trend over the five year study period. Derived flood frequency maps identify areas of open water and permanent flooding (12% of total area), seasonal flooding (29%), and intermittent flooding (48%). To evaluate the certainty of our results, we consider their consistency with (1) prior studies, (2) evapotranspiration estimates from the Atmosphere-Land Exchange Inverse (ALEXI) surface energy balance algorithm, (3) watershed storage anomaly estimates from GRACE, (4) supervised classification of open water area using Landsat, and (5) a rough measure of water availability (antecedent precipitation). The analyses show reasonable temporal and spatial consistency with available lines of evidence. We conclude that low resolution C-band SAR imagery shows promise for study of Sudd wetland flood dynamics. (C) 2017 Elsevier Inc. All rights reserved.