Abstract

Flood is the most common natural disaster of Sri Lanka. Sri Lanka is suffering from extensive flood events more frequently than previously due to climatic changes. Due to rapid urbanization, there are limited land resources to live in Kalutara district which is not prone to flood events. This research aims to model the flood susceptibility in Kalutara district Sri Lanka which is having frequent flood events using Logistic Regression Machine Learning model. Rainfall, Elevation, Slope, Aspect, Stream Power Index (SPI), Topographic Wetness Index (TWI), Distance to Rivers, River Water Level were selected as flood contribution factors. Flood extent was extracted using SAR images captured from Sentinel 1 using Google Earth Engine (GEE). Elevation data was downloaded from ALOS PALSAR dataset. Logistic Regression model fitting was done using python language with some open-source libraries. Fitted Logistic Regression model was able to achieve 0.89 Area Under the Curve (AUC) value in the Receiver Operating Characteristic (ROC) curve. The results show that flood susceptibility is very high around the major water features which are Kalu Ganga and Kuda Ganga. Among the other factors, River Water Level is having the highest contribution to the flood susceptibility in the area. Rainfall does not make  much contribution to the flood susceptibility over the area.

Keywords: Flood, flood susceptibility, Google Earth engine, logistic regression, machine learning, SAR, sentinel

🔒 This is a subscription article

Full text is available to subscribers and institutional members. Please choose an option below to access it.

1. Thomas V, Albert JRG, Hepburn C. Contributors to the frequency of intense climate disasters in Asia-Pacific countries. Clim Change. 2014;126(3-4):381–98.
2. Costache R, Arabameri A, Elkharachy I, Ghorbanzadeh O, Pham QB. Detection of areas prone to flood risk using state-of-the-art machine learning models. Geomatics, Nat Hazards Risk. 2021;12(5):1488–507.
3. Şen Z. Flood modeling, prediction, and mitigation. Cham: Springer; 2018.
4. Mandych AF. Classification of floods. Proc 2011 Int Symp ELMAR. 2011:5–10.
5. Burt TP, Weerasinghe KDN. Rainfall distributions in Sri Lanka in time and space: An analysis based on daily rainfall data. Climate. 2014;2(3):242–63.
6. Alahacoon N, Edirisinghe M. Spatial variability of rainfall trends in Sri Lanka from 1989 to 2019 as an indication of climate change. ISPRS Int J Geo-Inf. 2021;10(84):1–15.
7. Weerasinghe KM, Gehrels H, Arambepola N, Vajja HP, Herath J, Atapattu K. Qualitative flood risk assessment for the Western Province of Sri Lanka. Procedia Eng. 2018;212:503–10.
8. Munawar HS, Hammad AWA, Waller ST. Remote sensing methods for flood prediction: A review. Sensors. 2022;22(3):960.
9. Planinšič P, Gleich D. SAR-images categorization and applications. Proc 2017 Int Symp ELMAR. 2017:5–10.
10. Gorelick N, Hancher M, Dixon M, Ilyushchenko S, Thau D, Moore R. Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sens Environ. 2017;202:18–27.
11. Chowdhuri I, Pal SC, Chakrabortty R. Flood susceptibility mapping by ensemble evidential belief function and binomial logistic regression model on river basin of eastern India. Adv Space Res. 2020;65(6):1466–89.
12. Disaster Management Center (DMC). Floods and landslides: Integrated risk assessment. 2017.

Abstract

Flood is the most common natural disaster of Sri Lanka. Sri Lanka is suffering from extensive flood events more frequently than previously due to climatic changes. Due to rapid urbanization, there are limited land resources to live in Kalutara district which is not prone to flood events. This research aims to model the flood susceptibility in Kalutara district Sri Lanka which is having frequent flood events using Logistic Regression Machine Learning model. Rainfall, Elevation, Slope, Aspect, Stream Power Index (SPI), Topographic Wetness Index (TWI), Distance to Rivers, River Water Level were selected as flood contribution factors. Flood extent was extracted using SAR images captured from Sentinel 1 using Google Earth Engine (GEE). Elevation data was downloaded from ALOS PALSAR dataset. Logistic Regression model fitting was done using python language with some open-source libraries. Fitted Logistic Regression model was able to achieve 0.89 Area Under the Curve (AUC) value in the Receiver Operating Characteristic (ROC) curve. The results show that flood susceptibility is very high around the major water features which are Kalu Ganga and Kuda Ganga. Among the other factors, River Water Level is having the highest contribution to the flood susceptibility in the area. Rainfall does not make  much contribution to the flood susceptibility over the area.

Keywords: Flood, flood susceptibility, Google Earth engine, logistic regression, machine learning, SAR, sentinel

🔒 This is a subscription article

Full text is available to subscribers and institutional members. Please choose an option below to access it.

1. Thomas V, Albert JRG, Hepburn C. Contributors to the frequency of intense climate disasters in Asia-Pacific countries. Clim Change. 2014;126(3-4):381–98.
2. Costache R, Arabameri A, Elkharachy I, Ghorbanzadeh O, Pham QB. Detection of areas prone to flood risk using state-of-the-art machine learning models. Geomatics, Nat Hazards Risk. 2021;12(5):1488–507.
3. Şen Z. Flood modeling, prediction, and mitigation. Cham: Springer; 2018.
4. Mandych AF. Classification of floods. Proc 2011 Int Symp ELMAR. 2011:5–10.
5. Burt TP, Weerasinghe KDN. Rainfall distributions in Sri Lanka in time and space: An analysis based on daily rainfall data. Climate. 2014;2(3):242–63.
6. Alahacoon N, Edirisinghe M. Spatial variability of rainfall trends in Sri Lanka from 1989 to 2019 as an indication of climate change. ISPRS Int J Geo-Inf. 2021;10(84):1–15.
7. Weerasinghe KM, Gehrels H, Arambepola N, Vajja HP, Herath J, Atapattu K. Qualitative flood risk assessment for the Western Province of Sri Lanka. Procedia Eng. 2018;212:503–10.
8. Munawar HS, Hammad AWA, Waller ST. Remote sensing methods for flood prediction: A review. Sensors. 2022;22(3):960.
9. Planinšič P, Gleich D. SAR-images categorization and applications. Proc 2017 Int Symp ELMAR. 2017:5–10.
10. Gorelick N, Hancher M, Dixon M, Ilyushchenko S, Thau D, Moore R. Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sens Environ. 2017;202:18–27.
11. Chowdhuri I, Pal SC, Chakrabortty R. Flood susceptibility mapping by ensemble evidential belief function and binomial logistic regression model on river basin of eastern India. Adv Space Res. 2020;65(6):1466–89.
12. Disaster Management Center (DMC). Floods and landslides: Integrated risk assessment. 2017.