MACHINE LEARNING APPROACH FOR EVAPORATION DUCT NOWCAST

The Evaporation Duct Height (EDH) and Strength (EDS) are properties of the evaporation duct that affects electromagnetic (EM) signal propagation close to the air-sea interface. Hence, the accuracies of EDH and EDS affect radar and communication propagation, which can be exploited for detection and counter-detection operations. The EDH/EDS can be calculated utilizing meteorological and oceanographical (METOC) data collected onboard naval ships, including air temperature, sea surface temperature, wind direction, wind speed, sea level pressure, and relative humidity. In this work, we explore the utilization of artificial intelligence/machine learning (AI/ML) algorithms to demonstrate the feasibility to nowcast (up to six-hour forecast) EDH/EDS while a naval vessel is underway. The tested AI/ML algorithms include linear regression, decision trees, random forest, and neural networks. Datasets from the 2017 Coupled Air-Sea Processes and Electromagnetic Ducting Research (CASPER-West) project were used to train, test, and verify the predictions from the AI/ML algorithms. Two methods to forecast EDH/EDS are tested—one to forecast EDH/EDS directly, the other to calculate EDH/EDS based on the AI/ML forecast variables as input to NAVSLaM. The results are compared to those directly derived from the CASPER measurements. The effectiveness and limitations of the methods and algorithms are discussed.

Department dc:contributor.department

Meteorology (MR)

Grantor dc:publisher

Monterey, CA; Naval Postgraduate School

Year dc:date.issued

2021

Author dc:creator

• Yanez, Josue F.

Advisors dc:contributor.advisor

• Wang, Qing
• Feldmeier, Joel W.

Statement dc:rights

• This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.

Handle dc:identifier.uri

https://hdl.handle.net/10945/67835

OAI identifier oai:identifier

oai:calhoun.nps.edu:10945/67835