Forecasting the daily maximal and minimal temperatures from radiosonde measurements using neural networks

Skok, Gregor (author), Hoxha, Doruntina (author), Zaplotnik, Žiga (author)

This study investigates the potential of direct prediction of daily extremes of temperature at 2 m from a vertical profile measurement using neural networks (NNs). The analysis is based on 3800 daily profiles measured in the period 2004–2019. Various setups of dense sequential NNs are trained to predict the daily extremes at different lead times ranging from 0 to 500 days into the future. The short- to medium-range forecasts rely mainly on the profile data from the lowest layer—mostly on the temperature in the lowest 1 km. For the long-range forecasts (e.g., 100 days), the NN relies on the data from the whole troposphere. The error increases with forecast lead time, but at the same time, it exhibits periodic behavior for long lead times. The NN forecast beats the persistence forecast but becomes worse than the climatological forecast on day two or three. The forecast slightly improves when the previous-day measurements of temperature extremes are added as a predictor. The best forecast is obtained when the climatological value is added as well, with the biggest improvement in the long-term range where the error is constrained to the climatological forecast error.

• Language: English
• Keywords: machine learning, neural networks, weather forecasting, air temperature, climatology, radiosonde measurements, prediction, maximum temperature, minimum temperature, explainable AI
• Work type: Article
• Typology: 1.01 - Original Scientific Article
• Organization: FMF - Faculty of Mathematics and Physics
• Publication status: Published
• Publication version: Version of Record
• Year: 2021
• Number of pages: 17 str.
• Numbering: Vol. 11, iss. 22, art. 10852
• PID: 20.500.12556/RUL-136369
• UDC: 551.509
• ISSN on article: 2076-3417
• DOI: 10.3390/app112210852
• COBISS.SI-ID: 85351683
• Publication date in RUL: 26.04.2022

Record is a part of a journal

• Title: Applied sciences
• Shortened title: Appl. sci.
• Publisher: MDPI
• ISSN: 2076-3417
• COBISS.SI-ID: 522979353

Licences

• License: CC BY 4.0, Creative Commons Attribution 4.0 International
• Link: http://creativecommons.org/licenses/by/4.0/
• Description: This is the standard Creative Commons license that gives others maximum freedom to do what they want with the work as long as they credit the author.
• Licensing start date: 17.11.2021

Secondary language

• Language: Slovenian
• Keywords: strojno učenje, nevronske mreže, napovedovanje vremena, temperatura zraka, klimatologija, meritve z radiosondo

Projects

• Funder: ARRS - Slovenian Research Agency
• Project number: P1-0188
• Name: Astrofizika in fizika atmosfere

• Funder: ARRS - Slovenian Research Agency
• Project number: J1-9431
• Name: Prispevek Rossbyjevih in inercijsko-težnostnih valov k vertikalni hitrosti in pretoku gibalne količine v ozračju