Flood & Other Disaster forecasting using Predictive Modelling and Artificial intelligence.
Download as PPTX, PDF0 likes172 views
The document discusses the use of predictive modeling and artificial intelligence for forecasting floods and other natural disasters, highlighting the significant human and economic impacts of these events, particularly in India. It emphasizes the challenges of flood prediction due to complex natural factors and presents potential solutions involving various machine learning techniques for short-term flood prediction. The focus is on Bihar, India's most flood-prone state, identifying its unique challenges and advocating for advanced technological approaches to improve forecasting accuracy.
Flood & Other Disaster forecasting using Predictive Modelling and Artificial intelligence.
- 1. AKSHIT PRIYESH Flood & other disaster forecasting using predictive modelling and artificial intelligence.
- 2. 2 Every year natural disasters kill around 90,000 people and affect close to 160 million people worldwide.
- 3. FiveCountriesMostFrequentlyHitbyNatural Disasters. China United States Philippines Indonesia India
- 4. Data Story of Natural disasters in India post-Independence Disasters Events count Total deaths Total affected Total damage (million USD) Drought 13 1,500,320 1,391,841,000 5,441 Earthquake 29 51,915 285,656,623 5,297 Epidemic 63 20,874 421,473 - Extreme Temperature 59 17,600 - 544 Floods 283 70,343 861,462,744 58,332 Landslides/Avalanche 51 5,083 3,848,421 54 Storm 166 56,991 106,839,232 21,416 Total 664 1,723,126 2,650,069,743 91,086
- 5. In the last 17 years, India has faced more than 300 natural disasters which include drought, earthquake, epidemics, extreme temperature, floods, landslides and storms. These disasters have resulted in 76,031 deaths in this millennium.
- 6. Extreme Temperature Earthquake Floods Drought Epidemic 6 Someof thedisasterthatcanbepredictedbeforeoccurrence.
- 7. Over 2.3 Billion people are affected due to floods in last 20 years and causing countless death. More than 92 million cattle are lost every year, seven million hectares of land is affected, and damage is over 5 trillions dollars when taken globally in last 5 years. Flooding occurs when an extreme volume of water is carried by rivers, creeks and many other geographical features into areas where the water cannot be drained adequately.
- 9. Its due to the lack of accurate flood prediction system which can predict the situation accurately.
- 10. Floods are complicated natural events. It depends on several parameters, so it is very difficult to model analytically.
- 11. Some Factors Causing Flood • Characteristics of the catchment • Rainfall • Drainage • Opening of Barrage • Antecedent Conditions.
- 12. 12 To Overcome this challenge I have tried building a Flood Prediction System using Predictive modelling.
- 13. However, I have divided the idea of predictive modelling into small fragments to make it more effective.
- 14. We have considered most flooded state of India that is Bihar, but this can be used widely for most of the low lying geographical regions.
- 15. Bihar is India’s most flood-prone State, with 76 percent of the population, in the north Bihar living under the recurring threat of flood devastation. About 68800 sq Km out of total geographical area of 94163 sq Km comprising 73.06 percent is flood affected. Why Bihar?
- 16. From 1979 to Present day more than 8,873 Humans & 27,573 animals have lost their life due to flood. According to some historical data, 16.5% of the total flood affected area in India is located in Bihar while 22.1% of the flood affected population in India lives in Bihar.
- 18. • About 65% of catchments area of these rivers falls in Nepal/Tibet and only 35% of catchments area lies in Bihar. • The plains of Bihar, adjoining Nepal, are drained by a number of rivers that have their catchments in the steep and geologically nascent Himalayas. Kosi, Gandak,Burhi Gandak, Bagmati, Kamla Balan, Mahananda and Adhwara Group of rivers originates in Nepal, carry high discharge and very high sediment load and drops it down in the plains of Bihar. Complexity of Challenge
- 20. • TECHNOLOGY SUGGESTIONS • Watson Studio democratizes machine learning and deep learning to accelerate infusion of AI in to drive innovation. • An Intelligent Hydro- informatics Integration Platform for Regional Flood Inundation Warning Systems. • Three- Parameter Muskingum Model Coupled with an Improved Bat Algorithm. • Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation.
- 21. SOLUTION APPROACH I have studied several Solutions for Short-Term Flood Prediction Using Single ML Methods that is currently been used in different geographical regions across the world. Modeling Technique Reference Flood Resource Variable Prediction Type Region ANN vs. statistical Streamflow and flash food Hourly USA NN vs. traditional Water and surge level Hourly Japan ANN vs. statistical Flood Real-time UK ANN vs. statistical Extreme flow Hourly Greece FFANN vs. ANN Water level Hourly India ANN vs. T–S Flood Hourly India ANN vs. AR Stage level and streamflow Hourly Brazil
- 22. Comparative analysis of single ML models for the prediction of short-term floods. Modelin g Techniqu e Complexi ty of Algorith m Ease of Use Speed Accuracy Input Dataset ANN High Low Fair Fair Historical BPANN Fairly high Low Fairly high Fairly high Historical MLP Fairly high Fair High Fairly high Historical ELM Fair Fairly high Fairly high Fair Historical CART Fair Fair Fair Fairly high Historical SVM Fairly high Low Low Fair Historical ANFIS Fair Fairly high Fair Fairly high Historical
- 23. • 1. District wise rainfall monthly Data from 1901- 2019 • 2.Hydrometer Reading. • 3.River basin discharge. • 4.Rainfall forecast in Nepal. • 5. River basin Catchment Area. • 6.Death Records Human 1979- 2019. • 7.Death Records Animals 1979- 2019. 23 DataSets
- 36. References: 1.Daily Flood Bulletin. 2.Kosi Flood Bulletin 3.FMIS Report 2019. 4.Daily Flood Map.