A Deep Learning and Hidden Hierarchical Markov Model based analysis of Polar Ice Depletions and effects on Movements of Vulnerable Polar Animals

Frenzied depletion of polar icebergs has been witnessed over the last five decades with little to no efforts to mitigate its effects. This resulted in the endangerment of thousands of polar species and imbalanced the polar ecosystem. Researchers over the last two decades exhaustively studied the decay of polar ice, movements of icebergs, and effects in global climates using several deep learning methods. Multiple Hidden Markov Model based approaches also discussed the global migration of species due to unprecedented climate changes. In this paper, a dedicated hybrid of a Convolutional Neural Network and a Long Short-Term Memory Model was developed to study the deterioration of four primary parameters of polar ice followed by the prediction of continued depletion through the next two decades. Without considering triggers to slow down the decay, a Hidden Hierarchical Markov Model was developed to study the movements of three vulnerable polar animals. The study of three basic features of movements of each of them in the first layer of the hierarchy helped to predict their adaptations to new destinations, which in turn depicted their earnest efforts of survival by migrating towards warmer climates at lower latitudes.