Open AccessUsing ANN to Evaluate the Climate Data that High Affect on Calculate Daily Potential Evapotranspiration with Modified-Penman Method in The Tropical Regions
Author(s) -
D. A. D. Nusantara,
Feriza Nadiar
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1569/4/042028
Subject(s) - evapotranspiration , wind speed , environmental science , relative humidity , pan evaporation , climate change , climatology , meteorology , humidity , artificial neural network , climate model , atmospheric sciences , evaporation , geography , computer science , machine learning , ecology , geology , biology
In Indonesia, as part of tropical regions, calculating the amount of daily potential evapotranspiration (PET) becomes significant to ascertain the water balance. Applied Artificial Neural Network (ANN) as data-driven modeling in the period of Industrial Revolution 4.0 to develop a better world. Using ANN simplifies the process of modeling the various input climate data to count up the amount of daily PET. There six climate data that will be input in the Modified Penman Method, which are average-temperature, wind velocity, rainfall, relative-humidity, evaporation, and the time of solar radiation. Therefore, for further modeling using fewer input data than typically used in this method, so the high affect climatology data must be known. The methodology is to compare the best model, which has input six of the climate data with the model developed by just five of the climate data. The results of this research are the sequence of climate data that influence the daily PET model from the highest to the smallest: wind velocity, relative humidity, the time of solar radiation, temperature, rainfall, evaporation.