Changes in the relationship between maximum annual stage level and peak discharge in the Acre river

Código

I-EBHE0163

Autores

Luiz Felipe Pereira de Brito, MARCUS SUASSUNA SANTOS, Alondra Beatriz Alvarez Perez, Pedro Luiz Borges Chaffe

Tema

WG 1.10: Hydrologic Design - Solutions & Communication

Resumo

The increase in the magnitude, frequency, and extent of floods in the Amazon highlights the need for a better understanding of flood changes in the region. Typically, these flood changes are attributed to climatic or land cover drivers. In this study, we analyzed the variability of the river conveyance capacity, to investigate whether the non-stationarity of floods is related to changes in the relationship between stage level and discharge. We analyzed the historical series of annual peak stage levels and discharges (1968-2024), the cross-sections, and the hydraulic geometry relationships of the Acre river basin in Brazil (area of ~23,500 km²). This basin has been facing high susceptibility to erosion and flooding. To evaluate whether the changes in annual peak stage levels and discharges are significantly different, the magnitude of trends in the series of annual peaks was estimated using the linear regression model. To assess the statistical significance of these trends, the non-parametric Mann-Kendall test was used. Our results show that the three largest floods occurred in the last ten years. However, while the increasing trend in discharge was significant, there was no significant trend in stage levels. Moreover, ranking of largest flood and stage level are not exactly coincident. For instance, the second-largest flood discharge occurred in 1997 while the second-largest stage level in 2023. We hypothesize that the increase in discharge is compensated by the widening of the river's cross-section, indicating an increase in the river conveyance capacity; that is, changes in flood frequency are being compensated by the change in width. As impact assessment, we calculate the changes in flood frequency using the non-stationary GEV distribution with the location parameter varying over time. The probability of a flood similar to that of 2014 (3490 m³/s, the largest one on record) has doubled in the past 50 years. There is a need to jointly consider the trends in both discharge and river conveyance capacity to accurately quantify changes in flood risks. This understanding is crucial for predicting future changes and developing better adaptation strategies.