Forecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks

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 pre...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Gregor Skok, Doruntina Hoxha, Žiga Zaplotnik
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	machine learning neural network prediction maximum temperature minimum temperature radiosonde measurements Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
Acceso en línea:	https://doaj.org/article/53a914e579b74611b9ede873e1180743
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:53a914e579b74611b9ede873e1180743
record_format	dspace
spelling	oai:doaj.org-article:53a914e579b74611b9ede873e11807432021-11-25T16:39:21ZForecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks10.3390/app1122108522076-3417https://doaj.org/article/53a914e579b74611b9ede873e11807432021-11-01T00:00:00Zhttps://www.mdpi.com/2076-3417/11/22/10852https://doaj.org/toc/2076-3417This 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.Gregor SkokDoruntina HoxhaŽiga ZaplotnikMDPI AGarticlemachine learningneural networkpredictionmaximum temperatureminimum temperatureradiosonde measurementsTechnologyTEngineering (General). Civil engineering (General)TA1-2040Biology (General)QH301-705.5PhysicsQC1-999ChemistryQD1-999ENApplied Sciences, Vol 11, Iss 10852, p 10852 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	machine learning neural network prediction maximum temperature minimum temperature radiosonde measurements Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
spellingShingle	machine learning neural network prediction maximum temperature minimum temperature radiosonde measurements Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999 Gregor Skok Doruntina Hoxha Žiga Zaplotnik Forecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks
description	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.
format	article
author	Gregor Skok Doruntina Hoxha Žiga Zaplotnik
author_facet	Gregor Skok Doruntina Hoxha Žiga Zaplotnik
author_sort	Gregor Skok
title	Forecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks
title_short	Forecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks
title_full	Forecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks
title_fullStr	Forecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks
title_full_unstemmed	Forecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks
title_sort	forecasting the daily maximal and minimal temperatures from radiosonde measurements using neural networks
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/53a914e579b74611b9ede873e1180743
work_keys_str_mv	AT gregorskok forecastingthedailymaximalandminimaltemperaturesfromradiosondemeasurementsusingneuralnetworks AT doruntinahoxha forecastingthedailymaximalandminimaltemperaturesfromradiosondemeasurementsusingneuralnetworks AT zigazaplotnik forecastingthedailymaximalandminimaltemperaturesfromradiosondemeasurementsusingneuralnetworks
_version_	1718413074737659904

Forecasting the Daily Maximal and Minimal Temperatures from Radiosonde Measurements Using Neural Networks

Ejemplares similares