Please login first
METEOROLOGICAL DROUGHT PREDICTION FOR THE NORTHWEST REGION OF BANGLADESH USING ARTIFICIAL NEURAL NETWORK (ANN)
1 , 2 , * 3
1  Masters, Department of Disaster Science and Climate Resilience, University of Dhaka
2  Professor, Disaster Science and Climate Resilience, University of Dhaka
3  Adjunct Faculty, Disaster Science and Climate Resilience, University of Dhaka
Academic Editor: Charlotte Gardini

Abstract:

Numerous studies have been conducted to minimize the adversity yet continuous monitoring is required for the Northwest region of Bangladesh. The study introduces a method for forecasting meteorological droughts in the Northwest region of Bangladesh using daily precipitation, and temperature data from 1952 to 2020. The Standard Precipitation Index (SPI) and Standard Precipitation-Evapotranspiration Index (SPEI) parameters were created, and an Artificial Neural Network (ANN) model was used to predict droughts over 3-, 6-, 9-, and 12-month lead time. The findings of the study showed that short lead time prediction was better compared to long lead time predictions. The study also found that SPEI-based predictions were better than SPI for the six stations of the study area. Using the ANN model to predict drought using more parameters, the community of that location can be more resilient.

Keywords: SPI, SPEI, Multi-layer perceptron, ANN, Levenberg-Marquardt Algorithm
Comments on this paper
Currently there are no comments available.



 
 
Top