Banca de QUALIFICAÇÃO: DANIELE TORRES RODRIGUES

Uma banca de QUALIFICAÇÃO de DOUTORADO foi cadastrada pelo programa.
DISCENTE : DANIELE TORRES RODRIGUES
DATA : 23/02/2017
HORA: 14:00
LOCAL: Auditório do CCET
TÍTULO:
ANALYSIS OF EXTREMES OF PRECIPITATION IN THE NORTHEAST BRAZIL

PALAVRAS-CHAVES:

satellite TRMM, theory of extreme values, period of return.


PÁGINAS: 80
GRANDE ÁREA: Ciências Exatas e da Terra
ÁREA: Geociências
SUBÁREA: Meteorologia
ESPECIALIDADE: Climatologia
RESUMO:

Extreme weather events are becoming more and more frequent in Brazil. The consequences have been catastrophic climate changes, causing floods, drought, forest fires, landslides, river overflows, waves of cold or heat, among others. These environmental changes have major implications in society, especially in health, agriculture and water resources. Given this scenario, knowing the frequency with which extreme values occur is of great importance to society. The Theory of Extreme Values has helped in the prediction of these values, making it possible to predict the probability of occurrence of major catastrophes. According to the Intergovernmental Panel on Climate Change (IPCC), precipitation in Northeast Brazil (NEB) presents high spatial and temporal variability, which influences the occurrence of extreme events, making NEB vulnerable to climate change. Therefore, the objective of this work is to study the behavior and predict extreme values of precipitation in different areas of NEB. In addition to physically classifying the clouds that generate these extremes. For this purpose, two data bases will be used, the first coming from meteorological stations, from January 1986 to December 2015, and the second coming from the Tropical Rainfall Measuring Mission (TRMM) satellite, from January 1998 until December 2014.To decompose NEB into homogeneous areas according to daily precipitation, the cluster analysis technique will be used in both databases. However, before applying such a technique, the missing data contained in the database from stations will be populated by the multiple imputation method. Through the Generalized Extreme Value (GEV) and Generalized Pareto Distribution (GPD), it will be possible to estimate the return period that precipitation extremes will occur for each NEB area. Preliminary results indicate that the multiple imputations of data using the Bootstrap EM algorithm method can be a tool that corroborates the reconstruction of historical series of climatic data. And that in Natal, one of the capitals of the NEB, it is expected to occur at least once every 10 years daily precipitations of 107.79mm in April, 124.91mm in May, 158.03 in June and 131.94mm in July.

Extreme weather events are becoming more and more frequent in Brazil. The consequences have been catastrophic climate changes, causing floods, drought, forest fires, landslides, river overflows, waves of cold or heat, among others. These environmental changes have major implications in society, especially in health, agriculture and water resources. Given this scenario, knowing the frequency with which extreme values occur is of great importance to society. The Theory of Extreme Values has helped in the prediction of these values, making it possible to predict the probability of occurrence of major catastrophes. According to the Intergovernmental Panel on Climate Change (IPCC), precipitation in Northeast Brazil (NEB) presents high spatial and temporal variability, which influences the occurrence of extreme events, making NEB vulnerable to climate change. Therefore, the objective of this work is to study the behavior and predict extreme values of precipitation in different areas of NEB. In addition to physically classifying the clouds that generate these extremes. For this purpose, two data bases will be used, the first coming from meteorological stations, from January 1986 to December 2015, and the second coming from the Tropical Rainfall Measuring Mission (TRMM) satellite, from January 1998 until December 2014.To decompose NEB into homogeneous areas according to daily precipitation, the cluster analysis technique will be used in both databases. However, before applying such a technique, the missing data contained in the database from stations will be populated by the multiple imputation method. Through the Generalized Extreme Value (GEV) and Generalized Pareto Distribution (GPD), it will be possible to estimate the return period that precipitation extremes will occur for each NEB area. Preliminary results indicate that the multiple imputations of data using the Bootstrap EM algorithm method can be a tool that corroborates the reconstruction of historical series of climatic data. And that in Natal, one of the capitals of the NEB, it is expected to occur at least once every 10 years daily precipitations of 107.79mm in April, 124.91mm in May, 158.03 in June and 131.94mm in July.

MEMBROS DA BANCA:
Presidente - 1164414 - WEBER ANDRADE GONCALVES
Interno - 1752417 - CLAUDIO MOISES SANTOS E SILVA
Interno - 1914304 - KELLEN CARLA LIMA
Interno - 350691 - MARIA HELENA CONSTANTINO SPYRIDES
Interno - 320597 - PAULO SERGIO LUCIO
Notícia cadastrada em: 10/02/2017 13:03
SIGAA | Superintendência de Tecnologia da Informação - (84) 3342 2210 | Copyright © 2006-2024 - UFRN - sigaa04-producao.info.ufrn.br.sigaa04-producao