Application of machine learning to an early warning system for very short-term heavy rainfall

Moon, Seung‐Hyun; Kim, Yong-Hyuk; Lee, Yong Hee; Moon, Byung-Ro

doi:10.1016/j.jhydrol.2018.11.060

Cited by 73 publications

(37 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In future studies, it is anticipated that QC and correction using machine learning will have further improved performance by understanding relationships with other data through methods such as a dimensional reduction technique [35][36][37]. Furthermore, this study may become the basis of leading to practical studies such as the valuation of collected data and prediction of sensor malfunction.…”

Section: Resultsmentioning

confidence: 99%

Spatiotemporal Approaches for Quality Control and Error Correction of Atmospheric Data through Machine Learning

Kim

Park²,

Choi³

et al. 2020

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

We propose three quality control (QC) techniques using machine learning that depend on the type of input data used for training. These include QC based on time series of a single weather element, QC based on time series in conjunction with other weather elements, and QC using spatiotemporal characteristics. We performed machine learning-based QC on each weather element of atmospheric data, such as temperature, acquired from seven types of IoT sensors and applied machine learning algorithms, such as support vector regression, on data with errors to make meaningful estimates from them. By using the root mean squared error (RMSE), we evaluated the performance of the proposed techniques. As a result, the QC done in conjunction with other weather elements had 0.14% lower RMSE on average than QC conducted with only a single weather element. In the case of QC with spatiotemporal characteristic considerations, the QC done via training with AWS data showed performance with 17% lower RMSE than QC done with only raw data.

show abstract

Section: Resultsmentioning

confidence: 99%

Spatiotemporal Approaches for Quality Control and Error Correction of Atmospheric Data through Machine Learning

Kim

Park²,

Choi³

et al. 2020

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

show abstract

“…A los datos bibliográficos de las 1.154 publicaciones que resultaron del proceso de búsqueda se aplicaron los siguientes indicadores bibliométricos. (Moon et al, 2019;Moreno et al, 2018;Santacreu et al, 2015;Šaur, 2017). Como las ya mencionadas, hay muchas más aplicaciones en la cuales se han usado estas herramientas tecnológicas.…”

Section: Aplicación De Indicadores Bibliométricosunclassified

Análisis de la producción de publicaciones científicas en inteligencia artificial aplicada a la formulación de Políticas Públicas

2020

View full text Add to dashboard Cite

El objetivo de este artículo es analizar las publicaciones científicas especializadas en el uso de herramientas de inteligencia artificial en el proceso de toma de decisiones durante la formulación de políticas públicas. Como herramienta metodológica se creó una ecuación de búsqueda para ubicar las publicaciones concernientes, la cual fue probada y perfeccionada varias veces para mejorar los resultados encontrados. Esta ecuación fue aplicada en la base de datos de Scopus, con lo cual se obtuvieron 1154 publicaciones, a las que se aplicaron indicadores bibliométricos. En los resultados obtenidos se encontró que principalmente esta área ha tenido un gran crecimiento en la última década; cuyos países con mayor producción son: Estados Unidos, República Popular de China y el Reino Unido. Al comparar producción científica con las entidades patrocinantes se pudo concluir la importancia del apoyo gubernamental para desarrollo científico de un país.

show abstract

“…However, the purpose of an early warning system (EWS), flood early warning system (FEWS), or an ensemble prediction system (EPS) is to allow warning signals prior to extreme events mainly for very short-term heavy rainfall. Recent developments in machine learning techniques have heightened the need of using the mentioned hydrometeorological data for an effective EWS for very short-term (heavy rain advisory within the next 3-9 h) [16]. Nevertheless, we must accept that short-term quantitative precipitation forecasting (SQPF) is critical for flashflood warning, navigation safety, and many other applications.…”

Section: Introductionmentioning

confidence: 99%

“…13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 6770 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 118 121 124 127 130 133 136…”

unclassified

Algorithm to Predict the Rainfall Starting Point as a Function of Atmospheric Pressure, Humidity, and Dewpoint

Gutiérrez-López

Cruz-Paz

Mandujano

2019

Climate

View full text Add to dashboard Cite

Forecasting extreme precipitations is one of the main priorities of hydrology in Latin America and the Caribbean (LAC). Flood damage in urban areas increases every year, and is mainly caused by convective precipitations and hurricanes. In addition, hydrometeorological monitoring is limited in most countries in this region. Therefore, one of the primary challenges in the LAC region the development of a good rainfall forecasting model that can be used in an early warning system (EWS) or a flood early warning system (FEWS). The aim of this study was to provide an effective forecast of short-term rainfall using a set of climatic variables, based on the Clausius–Clapeyron relationship and taking into account that atmospheric water vapor is one of the variables that determine most meteorological phenomena, particularly regarding precipitation. As a consequence, a simple precipitation forecast model was proposed from data monitored at every minute, such as humidity, surface temperature, atmospheric pressure, and dewpoint. With access to a historical database of 1237 storms, the proposed model allows use of the right combination of these variables to make an accurate forecast of the time of storm onset. The results indicate that the proposed methodology was capable of predicting precipitation onset as a function of the atmospheric pressure, humidity, and dewpoint. The synoptic forecast model was implemented as a hydroinformatics tool in the Extreme Precipitation Monitoring Network of the city of Queretaro, Mexico (RedCIAQ). The improved forecasts provided by the proposed methodology are expected to be useful to support disaster warning systems all over Mexico, mainly during hurricanes and flashfloods.

show abstract

Application of machine learning to an early warning system for very short-term heavy rainfall

Cited by 73 publications

References 47 publications

Spatiotemporal Approaches for Quality Control and Error Correction of Atmospheric Data through Machine Learning

Spatiotemporal Approaches for Quality Control and Error Correction of Atmospheric Data through Machine Learning

Análisis de la producción de publicaciones científicas en inteligencia artificial aplicada a la formulación de Políticas Públicas

Algorithm to Predict the Rainfall Starting Point as a Function of Atmospheric Pressure, Humidity, and Dewpoint

Contact Info

Product

Resources

About