Author(s)

JUSSARA SOCORRO CURY MACIEL, LUNA GRIPP SIMÕES ALVES, BRUNO GABRIEL DOS SANTOS CORRÊA, IRAÚNA MAICONÃ RODRIGUES DE CARVALHO, MARCO ANTÔNIO OLIVEIRA

Abstract

The Amazon region has rivers of great contribution to the biodiversity balance and regional landscape. Among them, the stand out are the Negro River – the most extensive, the Amazon River – with the largest volume, the Purus River – a fish supplier, and the Madeira River – the most modified. Flooding in the Amazon is often widespread and sometimes severe for those who live very close to rivers, yet urban areas are often more socially impacted than rural areas, as flooding is part of the riverside population’s daily life. Flood forecasting models are important for composing extreme event alerts as well as for knowledge of decision-makers, public agency representatives, and affected communities. This paper aims to present what is being done in Manaus about the Negro River floods, with the development of a flood alert dissemination project, which uses a linear regression statistical model, based on the historical series over 100 years. The follow-up to the one-year flood forecast begins the previous year, more precisely when the ebb ends and the rainy season begins. Several factors contribute to the flood event, but the main one is the rainy season and how the basin will behave during the six months of rising waters. Manaus station receives contributions from two major rivers: Negro and Solimões, which form the Amazon River downstream. The position of the city, allied to the difference in depth, speed, and flow of the rivers, added to the rainfall period in the basin reveal the flooding characteristic in this locality. Considering the last 15 years, the near-real flood forecast interval in 87% of cases reveals that the project has fulfilled the purpose of alerting the population and public representatives about the flood of Negro River.

Keywords

extreme events, flood forecast, Amazon rivers