Técnicas de clasificación supervisada para la discriminación entre ecos meteorológicos y no meteorológicos usando informacion de un radar de banda C

Translated title of the contribution: Supervised classification techniques for discrimination between meteorological and non-meteorological echoes using a C-band radar

Sofia Ruiz Suarez, Mariela Sued, Luciano Vidal, Paola Salio, Daniela Rodriguez, Stephen Nesbitt, Yanina Garcia Skabar

Research output: Contribution to journalArticle

Abstract

Data coming from meteorological radars is of the utmost importance for the diagnosis and monitoring of precipitation systems and their possible associated severe phenomena. The echoes caused by objectives that are not meteorological introduce errors in the information. Therefore, it is necessary to detect their presence before using this data. This paper presents four supervised classification techniques based on different models which seek to give an answer to this problem. In addition, as an important part of this work, resampling techniques were implemented on the training set in order to further asses the results. Resampling methods are an indispensable tool in modern statistics. Those techniques provide additional information about the model of interest by repeatedly drawing samples from then data. Based on data from a C-band Dual-PolarizationDoppler weather radar located in Anguil and from a previous expert's manual classification, four supervised classification methods with different degrees of flexibility in their structure were implemented: Lineal Model, Quadratic Model, Logistic Model and Bayes Naive Model. Finally, the results of each of them were assessed and compared. Although difficulties were encountered in classifying boundary zones between classes, the results obtained were adequate, showing the best performance in the least flexible model, the linear one. It is considered necessary to keep working in this line of research in order to include more cases in the analysis and allow a better inference on the results.

Original languageSpanish
Pages (from-to)45-65
Number of pages21
JournalMeteorologica
Volume44
Issue number1
StatePublished - Jan 1 2019

Fingerprint

image classification
radar
logistics
weather
monitoring

Keywords

  • Non-meteorological echoes
  • Resampling methods
  • Supervised classification
  • Weather radar

ASJC Scopus subject areas

  • Atmospheric Science

Cite this

Técnicas de clasificación supervisada para la discriminación entre ecos meteorológicos y no meteorológicos usando informacion de un radar de banda C. / Suarez, Sofia Ruiz; Sued, Mariela; Vidal, Luciano; Salio, Paola; Rodriguez, Daniela; Nesbitt, Stephen; Skabar, Yanina Garcia.

In: Meteorologica, Vol. 44, No. 1, 01.01.2019, p. 45-65.

Research output: Contribution to journalArticle

Suarez, Sofia Ruiz ; Sued, Mariela ; Vidal, Luciano ; Salio, Paola ; Rodriguez, Daniela ; Nesbitt, Stephen ; Skabar, Yanina Garcia. / Técnicas de clasificación supervisada para la discriminación entre ecos meteorológicos y no meteorológicos usando informacion de un radar de banda C. In: Meteorologica. 2019 ; Vol. 44, No. 1. pp. 45-65.
@article{8abfc53e4eac4940b2c236d0fe370ee4,
title = "T{\'e}cnicas de clasificaci{\'o}n supervisada para la discriminaci{\'o}n entre ecos meteorol{\'o}gicos y no meteorol{\'o}gicos usando informacion de un radar de banda C",
abstract = "Data coming from meteorological radars is of the utmost importance for the diagnosis and monitoring of precipitation systems and their possible associated severe phenomena. The echoes caused by objectives that are not meteorological introduce errors in the information. Therefore, it is necessary to detect their presence before using this data. This paper presents four supervised classification techniques based on different models which seek to give an answer to this problem. In addition, as an important part of this work, resampling techniques were implemented on the training set in order to further asses the results. Resampling methods are an indispensable tool in modern statistics. Those techniques provide additional information about the model of interest by repeatedly drawing samples from then data. Based on data from a C-band Dual-PolarizationDoppler weather radar located in Anguil and from a previous expert's manual classification, four supervised classification methods with different degrees of flexibility in their structure were implemented: Lineal Model, Quadratic Model, Logistic Model and Bayes Naive Model. Finally, the results of each of them were assessed and compared. Although difficulties were encountered in classifying boundary zones between classes, the results obtained were adequate, showing the best performance in the least flexible model, the linear one. It is considered necessary to keep working in this line of research in order to include more cases in the analysis and allow a better inference on the results.",
keywords = "Non-meteorological echoes, Resampling methods, Supervised classification, Weather radar",
author = "Suarez, {Sofia Ruiz} and Mariela Sued and Luciano Vidal and Paola Salio and Daniela Rodriguez and Stephen Nesbitt and Skabar, {Yanina Garcia}",
year = "2019",
month = "1",
day = "1",
language = "Spanish",
volume = "44",
pages = "45--65",
journal = "Meteorologica",
issn = "0325-187X",
publisher = "Centro Argentino de Meteorologos",
number = "1",

}

TY - JOUR

T1 - Técnicas de clasificación supervisada para la discriminación entre ecos meteorológicos y no meteorológicos usando informacion de un radar de banda C

AU - Suarez, Sofia Ruiz

AU - Sued, Mariela

AU - Vidal, Luciano

AU - Salio, Paola

AU - Rodriguez, Daniela

AU - Nesbitt, Stephen

AU - Skabar, Yanina Garcia

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Data coming from meteorological radars is of the utmost importance for the diagnosis and monitoring of precipitation systems and their possible associated severe phenomena. The echoes caused by objectives that are not meteorological introduce errors in the information. Therefore, it is necessary to detect their presence before using this data. This paper presents four supervised classification techniques based on different models which seek to give an answer to this problem. In addition, as an important part of this work, resampling techniques were implemented on the training set in order to further asses the results. Resampling methods are an indispensable tool in modern statistics. Those techniques provide additional information about the model of interest by repeatedly drawing samples from then data. Based on data from a C-band Dual-PolarizationDoppler weather radar located in Anguil and from a previous expert's manual classification, four supervised classification methods with different degrees of flexibility in their structure were implemented: Lineal Model, Quadratic Model, Logistic Model and Bayes Naive Model. Finally, the results of each of them were assessed and compared. Although difficulties were encountered in classifying boundary zones between classes, the results obtained were adequate, showing the best performance in the least flexible model, the linear one. It is considered necessary to keep working in this line of research in order to include more cases in the analysis and allow a better inference on the results.

AB - Data coming from meteorological radars is of the utmost importance for the diagnosis and monitoring of precipitation systems and their possible associated severe phenomena. The echoes caused by objectives that are not meteorological introduce errors in the information. Therefore, it is necessary to detect their presence before using this data. This paper presents four supervised classification techniques based on different models which seek to give an answer to this problem. In addition, as an important part of this work, resampling techniques were implemented on the training set in order to further asses the results. Resampling methods are an indispensable tool in modern statistics. Those techniques provide additional information about the model of interest by repeatedly drawing samples from then data. Based on data from a C-band Dual-PolarizationDoppler weather radar located in Anguil and from a previous expert's manual classification, four supervised classification methods with different degrees of flexibility in their structure were implemented: Lineal Model, Quadratic Model, Logistic Model and Bayes Naive Model. Finally, the results of each of them were assessed and compared. Although difficulties were encountered in classifying boundary zones between classes, the results obtained were adequate, showing the best performance in the least flexible model, the linear one. It is considered necessary to keep working in this line of research in order to include more cases in the analysis and allow a better inference on the results.

KW - Non-meteorological echoes

KW - Resampling methods

KW - Supervised classification

KW - Weather radar

UR - http://www.scopus.com/inward/record.url?scp=85072763892&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072763892&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:85072763892

VL - 44

SP - 45

EP - 65

JO - Meteorologica

JF - Meteorologica

SN - 0325-187X

IS - 1

ER -