Space-time variability of rainfall and hydrological trends in the Alto São Francisco River basin (original) (raw)

Variability of rainfall in the semi-arid region of Brazil

The Brazilian semi-arid region presents a highly variable rainfall regime in space and time. Although the mechanisms of rainfall have been well described and numerically modelled, reliable forecasts for more than six months in advance cannot be produced yet, as well as projections of climate change in the long term. This paper contributes to the understanding of the variability examining a more than 100-year time series of a rainfall gauge in this region, using three techniques: wavelet analysis, Mann-Kendall and Sen tests. The techniques allow the description of the patterns of variability of rainfall and suggest that there is not still a clear evidence of climate change.

Variations of Rainfall Rhythm in Alto Pardo Watershed, Brazil: Analysis of Two Specific Years, a Wet and a Dry One, and Their Relation with the River Flow

Climate

This research aims to understand the variability and rhythm of rainfall for two specific standard-years, and their relation with the river flow of the Alto Pardo watershed, located in southeastern Brazil, and thus identify atmospheric systems that can cause extreme events, and which may be reflected in heavy rainfall, floods, or drought episodes. Therefore, the research chose to investigate the years 1983 and 1984, rainy and dry standard-years respectively in the study area, where rainfall was described and spatialized through the geostatistical method of kriging at the monthly level and the rhythmic analysis technique was applied in order to identify what weather types are usual and extreme in the area. The results indicate that a high involvement of the frontal system in the year 1983 was responsible for the episodes of greater rainfall and peak water flow, especially in stationary front episodes. The year 1984 presented low rainfall in summer, a meteorological drought during the ...

Temporal distribution of rainfall in the Far West region of Santa Catarina, Brazil

Revista Engenharia na Agricultura - Reveng, 2021

In the definition of design rainfall, one must determine the temporal distribution of rainfall. In Brazil there are few studies on the temporal distribution of heavy rainfall. This work aimed to characterize the temporal distribution of intense rainfall for the Far West region of Santa Catarina. Data from four rainfall stations were used. The rainfall was individualized and classified into four types according to the quartile with the highest intensity. With the total of 3212 rainfall events it was observed that the most frequent rains are of type I (37.6%) followed by types II (32.3%). The time variation curves of the four rainfall stations show differences of less than 5% in relation to the regional average. No significant seasonal differences were observed, however significant differences were found with respect to rainfall duration. The values of the temporal distribution with a probability of 50% were necessary for the rains of the four quartiles, as well as for the duration ra...

Observing the Existence of Low-Frequency Variability in Monthly Rainfall Data at Southeastern Brazil using R Package Tools – Neural Networks and Wavelet

2020

ABSTRACT. This study aimed to analyze 70 years historical series in the Brazilian Southeastern region, using monthly rainfall data. Statistical modeling techniques such as cross-wavelet spectra and artificial neural networks (ANN), from the R statistical package, were used to perform the analyses. Two different types of neural networks were employed: the multi-layer perceptron (MLP) and extreme learning machine (ELM). From the cited time series, the analysis shows the existence of a decadal and multi-decadal signal with cycles of 5, 11, and 22 years in the monthly rainfall in Brazilian Southeastern region, observing the existence of low-frequency variability. This shows a significant degree of modulation and association for the precipitation with solar activity. The neural networks were also used as forecasting tools, with a better performance for MLP-NN – smaller root mean square error. However, the MLP-NN presented a greater confidence interval than ELM-NN. Keywords: Monthly Rainf...