Fabrício Daniel dos Santos Silva

Modelagem hidrológica determinística e estocástica aplicada à região hidrográfica do Xingu- Pará

Hydrologic modeling is an important tool for the planning and management of water resources use in river basins. In this work, a two-parameter monthly deterministic hydrologic model and the stochastic model, ARIMA, were applied to simulate the monthly runoff of the Xingu river basin in the State of Para. The main objective of this work is to simulate the monthly runoff using the two models and to compare their results. The applied hydrological deterministic model has a simple structure and presented good results, but seems to be very sensitive to extreme precipitation events. The stochastic model ARIMA was able to capture the dynamic of the temporal series, presenting very satisfactory results for the simulation of the monthly runoff at the basin stations. Both models should be applied with caution during the rainy season, when extreme precipitation events and consequently peaks of runoff occur.

Risco climático para o cultivo do algodoeiro na região nordeste do Brasil

Weekly mean values of solar radiation, rainfall, evapotranspiration and air temperature were used for establishing the lower climatic risk to the cotton crop. For irrigated fields, the potential crop yield was evaluated by the cumulative growth index (CGI), while for rain-fed conditions the climatic risk was determined by the water potential deficit (WPD). Under irrigated and rain-fed conditions, the lower climatic risk occurs when cotton crop is sowing from 6 to 19 August and from the 44th (West region of Bahia state) to the 19th week of the year. From the 1st week of the year, the most suitable date for cotton sowing start in the Central region of Piaui state, reaching maximum covering area of Piaui, Rio Grande do Norte, Paraiba, Pernambuco states and North of Bahia state in the period between the 4th and the 8th week. In the South and West regions of Bahia state, the lower climatic risk is reached by cotton crop when sowing in the 10th and 46th weeks. In the northeast Coast, otherwise the cotton crop sowing is more suitable in the period from the 14th to 17th week, covering most of the coastal shore table lands from Rio Grande do Norte to Bahia states and almost the total areas of the Alagoas and Sergipe states.

Um modelo estocástico combinado de previsão sazonal para a precipitação no Brasil

This article discusses a combined model to perform climate forecast in a seasonal scale. In it, forecasts of specific stochastic models are aggregated to obtain the best forecasts in time. Stochastic models are used in the auto regressive integrated moving average, exponential smoothing and the analysis of forecasts by canonical correlation. The quality control of the forecast is based on the residual analysis and the evaluation of the percentage of reduction of the unexplained variance of the combined model with respect to the individual ones. Examples of application of those concepts to models developed at the Brazilian National Institute of Meteorology (INMET) show good results and illustrate that the forecast of the combined model exceeds in most cases each component model, when compared to observed data.

Precipitação na Cidade de Salvador: Variabilidade Temporal e Classificação em Quantis

Este trabalho tem como objetivo contribuir com o estudo da precipitacao na cidade de Salvador–BA. Os totais mensais de precipitacao obtidos pelo Instituto Nacional de Meteorologia no periodo de 51 anos (1961-2011) fundamentaram esta pesquisa. A variabilidade pluviometrica foi analisada com enfase na determinacao das classes da precipitacao nas escalas anual, quadrimestral e mensal. Foram utilizados os testes de significância nao parametricos de Mann-Kendall e as estimativas de inclinacao pelo metodo da Curvatura de Sen. Na classificacao pluviometrica foi utilizado o Metodo dos Percentis (Quantis). Para a analise anual, os resultados mostraram que 1964 (1961) foi um ano classificado como “muito chuvoso” (“muito seco”). Na analise do quadrimestre mais chuvoso AMJJ (mais seco ASON), o ano de 1971 (1980) foi o mais extremo “muito chuvoso” (“muito seco”). Na analise mensal, o evento que obteve maior volume pluviometrico ocorreu em abril de 1984 com registros de 889,8 mm. A analise da tendencia mostrou uma reducao no volume de chuva na escala anual, entretanto, sem significância estatistica. Grande parte dos eventos analisados ocorreram com condicoes oceânicas no Pacifico equatorial e/ou Atlântico Sul Tropical favoraveis (desfavoraveis) ao aumento (diminuicao) do volume de chuva na area em estudo, tendo por base uma analise qualitativa.

CONVECÇÃO LINEARMENTE ORGANIZADA NA ÁREA DE PETROLINA, SEMI-ÁRIDO DO NORDESTE DO BRASIL: ASPECTOS EM MESO E GRANDE ESCALA

A study on linearly organized convective systems observed by a C-band meteorological radar in the semi-arid area of Northeast Brazil is presented. The convective activity during three selected days of 1985 March (27 to 29) was analyzed emphasizing the role played by local and large-scale factors on the development of rain systems. On the large-scale scenario, the covered radar area was influenced by an austral upper-level air trough on March 27th and by an upper-level cyclonic vortex on the 29th. Near-surface moisture convergence favored convective activity on days 27 and 29, while near-surface moisture divergence inhibited convective activity on day 28. On the mesoscale, it was found that diurnal heating was an important factor influencing the formation of the convective cells, while their location was determined mainly by the orography. In general, the radar images show the linearly-organized convective systems over elevated terrain and intense convective cells embedded by a stratiform rain area. The results indicate that large-scale convergence of the moisture flux and radiative heating are determinant factors on the evolution and development of the echoes in the area of study.

IDENTIFICATION AND CLASSIFICATION SPATIOTEMPORAL OF RAINY OR DRY CRITICAL EVENTS IN PERNAMBUCO STATE

The objective of this study was to analyze the dry and wet events historically occurred in the Pernambuco state and diagnose their intensities and durations. The identification and classification of areas susceptible to critical events has become a frequent need in the current context of climate change, which has caused disasters in the world. The diagnosis based on meteorological, agricultural and hydrological impacts can be measured with the application of the methodology of the Standardized Precipitation Index (SPI) for different time scales. The SPI was able to identify critical events in the state of Pernambuco and highlight the main through their intensities and durations. Scales were analyzed from quarterly to biennial which indicated the beginning, trend, duration and intensity of events. The results identified years with rainfall events in the moderate and severe categories, but also dry events in moderate to extreme categories. It was observed that the values of correlations between SPI scales are better as they approach in time. The cluster analysis of the SPIs showed well-defined areas, up to four groups, at all scales of time, this indicates that these groups have similar behavior during critical events.

Padrão Mensal de Anomalias de Precipitação: Uma Análise Estatística de Eventos Hidrológicos Extremos nas Sub-Bacias do Rio São Francisco (Standard Monthly Precipitation Anomalies: A Statistical Analysis of Extreme Hydrological Events in Sub-Basins from...

Neste trabalho foi investigada a influencia da variabilidade climatica na ocorrencia de enchentes no Medio e Submedio da bacia hidrografica do Sao Francisco (BHSF). A cheia registrada em 1985 foi selecionada para analise por representar um evento hidrologico extremo com caracteristicas ambientais pouco documentadas. A utilizacao do indice RAI (“Rainfall Anomaly Index”) na escala mensal como metodo de analise permitiu detectar a intensidade, duracao e extensao das areas atingidas pelo fenomeno. Uma analise estatistica complementar atraves de Analise Fatorial por Componentes Principais (ACP), aplicada ao conjunto de valores mensais do RAI, permitiu identificar vinculos entre configuracoes espaciais de anomalias de precipitacao na BHSF e padroes atmosfericos que influenciaram a intensidade e distribuicao das chuvas. Os resultados mostraram que a atuacao conjunta de sistemas meteorologicos transientes de diferentes escalas e responsavel pela organizacao da conveccao, precipitacao intensa e formacao de cheias com inundacoes no Submedio da BHSF. A B S T R A C T In this work the influence of the climatic variability upon the occurrence of floods in the Sao Francisco River Basin (SFRB)is investigated. The major flood of 1985 was selected for analysis since it represents an extreme hydrological event whose environmental characteristics are poorly documented. The Rainfall Anomaly Index (RAI) obtained on a monthly scale allowed detection of the intensity, duration and extent of theareas affectedby the phenomenon. A complementary statistical analysisby means of the factorial analysis by principal components (PCA) applied to the monthly RAI values allowed to identify links between spatial configurations of the rainfall anomalies in the SFRB and atmospheric patterns that influenced rainfall intensity and distribution. The results showed that the joint action of transient meteorological systems of various scales is responsible for convection organization, heavy precipitation and formation of floods with inundations in the Lower Basin of the SFRB. Key-Words: Hydrographic Basin of the Sao Francisco River, inundations, RAI index, factorial analysis by principal components.

Imputação Multivariada de Dados Diários de Precipitação e Análise de Índices de Extremos Climáticos (Imputation Multivariate of Precipitation Daily Data and Analysis of Climate Extremes Index)

Neste trabalho utilizou-se a tecnica MICE (do ingles “ Multivariate Imputation by Chained Equations ”) para imputacao de dados diarios de precipitacao em seis series com falhas do Estado da Paraiba: Areia, Campina Grande, Monteiro, Joao Pessoa, Patos e Sao Goncalo, entre periodo de 1979 e 2010, usando como fonte de informacao para o preenchimento dados de pontos de grade proximos a estacao meteorologica selecionada. Com a finalidade de validacao do metodo, foram geradas falhas em dados observados para determinados anos - principio da validacao cruzada. A metodologia apresentou resultados promissores. Foram obtidos altos valores de correlacoes tanto entre os dados diarios observados e imputados, assim como quando estes foram acumulados mensalmente. A partir dos dados imputados, utilizou-se o software R-ClimDex com o objetivo de avaliar possiveis tendencias nos indices climaticos relacionados a precipitacao. As analises das series sem falhas mostraram que das cidades estudadas, Areia foi a unica que apresentou tendencia de reducao no numero de dias com chuvas acima de 1mm, enquanto as demais cidades apresentaram tendencia de aumento. Palavras - chave: dados imputados, MICE, R-ClimDex, tendencias climaticas. Imputation Multivariate of Precipitation Daily Data and Analysis of Climate Extremes Index ABSTRACT This work used the MICE ( Multivariate Imputation by Chained Equations ) technique for daily rainfall data imputation in series with gaps , for six series of Paraiba State: Areia , Campina Grande, Monteiro , Joao Pessoa , Patos and Sao Goncalo, between 1979 and 2010, using as a source of information data grid points near the selected station (nearest neighbors) . In order to validate the method one generates missing values in observed data for certain years– the cross-validation principle . The methodology presented promising results . One obtains high values ​​for both correlations between daily observed and imputed data , as well as when they are monthly accumulates. Based on the imputed data, used the software R-ClimDex with the objective to evaluate possible trends in climate indices related to precipitation. The analysis of the series without flaws of the cities studied showed that Areia was the only one that showed decrease trend in the number of days with precipitation above1 mm , while the other cities showed increase trend. Keywords: imputed data, MICE, R-ClimDex, climatic trends.

AVALIAÇÃO DAS ALTERAÇÕES NO REGIME HIDROLÓGICO DA BACIA DO SÃO FRANCISCO ATRAVÉS DO ÍNDICE RAI (RAINFALL ANOMALY INDEX)

1. Introdução O monitoramento de eventos meteorológicos extremos é de grande relevância para projetos de abastecimento de água, geração de energia elétrica e atividades agrícolas. Informações sobre a intensidade, duração e severidade de sistemas atmosféricos, possibilita medidas preventivas em curto prazo. Neste trabalho duas questões são abordadas: a variabilidade do regime de chuvas na BHSF e a distribuição espacial das anomalias de chuvas no Alto, parte do Médio e Submédio sob a influência do fenômeno El Nino Oscilação Sul (ENOS). O evento de 1983 foi escolhido para análise.