Detecting spatio-temporal changes in the extent of seasonal and annual flooding in South Asia using multi-resolution satellite data.

This paper presents algorithm for flood inundation mapping to understand seasonal and annual changes in the flood extent and in the context of emergency response. Time-series profiles of Land Surface Water Index (LSWI), Enhanced Vegetation Index (EVI), Normalized Difference Vegetation Index (NDVI) and Normalized Difference Snow Index (NDSI) are obtained from MOD09 8-day composite time-series data (resolution 500m; time period: 2000-2011). The proposed algorithm was applied for MODIS data to produce time-series inundation maps for the ten annual flood season over the period from 2000 to 2011. The flood product has three classes as flood, mixed and long-term water bodies. The MODIS flood products were validated via comparison with ALOS AVINIR / PALSAR and Landsat TM using the flood fraction comparison method. Compared with the ALOS satellite data sets at a grid size of 10km the obtained RMSE range from 5.5 to 15 km2 and the determination coefficients range from 0.72 to 0.97. The spatial characteristics of the estimated early, peak and late and duration of inundation cycle were also determined for the period from 2000 to 2011. There are clear contracts in the distribution of the estimated flood duration of inundation cycles between large-scale floods (2008-2010) and medium and small-scale floods (2002 and 2004). Examples on the analysis of spatial extent and temporal pattern of flood-inundated areas are of prime importance for the mitigation of floods. The generic approach can be used to quantify the damage caused by floods, since floods have been increasing each year resulting in the loss of lives, property and agricultural production.

Bibliographic Details

Main Authors:	Amarnath, Giriraj, Ameer, Mohamed, Aggarwal, Pramod K., Smakhtin, Vladimir U.
Format:	Conference Paper biblioteca
Language:	English
Published:	International Society for Optics and Photonics 2012
Subjects:	flooding, satellites, satellite imagery, time series analysis,
Online Access:	https://hdl.handle.net/10568/34513 http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=1387506
Tags:	Add Tag No Tags, Be the first to tag this record!