Inter-Research >  > Prepress Abstract

CR prepress abstract   -  DOI: https://doi.org/10.3354/cr01612

Evaluation of the potential of a deep learning-GIS hybrid approach in precipitation modeling based on spatio-temporal variables in coastal zone of Turkey

Halit Apaydin, Mohammad Taghi Sattari*

*Corresponding author:

ABSTRACT: It is clearly known that precipitation is an indispensable need for fauna and flora. Studies have shown that location and temporal factors have an effect on precipitation. Accurate prediction of precipitation is very important for water resources management and artificial intelligence methods are frequently used for this purpose. In this study, deep learning and geographic information system (GIS) hybrid approach based on spatio-temporal variables were applied in order to model the amount of precipitation on Turkey's coastline. The latitude, longitude, altitude, distance to the sea and aspect information of the meteorological stations were utilized as spatial variables. The change of monthly precipitation is taken into account as a temporal variable. Artificial intelligence methods such as Gaussian Process Regression, Support Vector Regression, Broyden – Fletcher – Goldfarb – Shanno artificial neural network, M5, Random Forest and Long short-term memory (LSTM) were used. According to the results of the study, in which different input variable alternatives were also evaluated, LSTM was the most successful method with a value of 0.93 R. It has been revealed that the amount of precipitation can be estimated and the distribution map can be drawn by using spatio-temporal data, deep learning and GIS hybrid method at the points where the measurement is not performed.