Application of frequency ratio and logistic regression models for assessing physical wetland vulnerability in Punarbhaba river basin of Indo-Bangladesh

ABSTRACT

Growingly scarce ecologically viable flood plain wetland of the Punarbhaba river basin is further endangered due to flow modification through Komardanga dam. This work intends to discover physical vulnerability of the wetlands in Punarbhaba river basin of Indo-Bangladesh considering seven conditioning parameters, e.g., water presence frequency (WPF) map, flood inundation map, frequency of pixel being non-permanent, agriculture presence frequency (APF) map, fragmentation of wetland, normalized differentiation built up index (NDBI), and wetland changes (WC). Frequency Ratio and Logistic Regression models have been used for deriving the vulnerability of wetland for both pre (1988–1992) and post dam (1993–2016) periods. From computed FR models it is exhibited that out of the total wetland area (194.81 km²), 5.88% and 2.92% area are high and very highly vulnerable in pre-dam period but such vulnerable area is increased to 33.45% and 23.10% in post-dam state (total wetland: 126.11 km²). LR models also state that in pre-dam period, high and very high vulnerable wetland area were 5.02% and 3.82% (total wetland: 194.79 km²) and it is enhanced to 28.94% and 24.49% in post-dam state (total wetland: 126.11 km²). Extensions of agricultural land, squeezing of active flood plain, lowering flood frequency are dominant determinants for growing wetland vulnerability.

KEYWORDS:

• physical wetland vulnerability
• hydrological alteration
• frequency ratio model
• logistic regression model and sensitivity analysis

Acknowledgments

I would like to extended hearty acknowledgement to USGS for providing me Landsat imageries. I am extremely thankful to my seniors (Susanta Mahato, Tirthankar Basu and Subrata Mandal) for helping me to develop the model.