TRENDS IN STREAMFLOW OF HYDROGRAPHIC BASINS IN CENTRAL-SOUTH ESPÍRITO SANTO
Name: JULIANA OLIVEIRA RODRIGUES
Publication date: 26/02/2026
Examining board:
Name |
Role |
|---|---|
| FRANCIELLE RODRIGUES DE OLIVEIRA | Examinador Externo |
| JÉFERSON LUIZ FERRARI | Examinador Externo |
| ROBERTO AVELINO CECILIO | Presidente |
Summary: The analysis of streamflow temporal dynamics is essential for understanding alterations in hydrological regimes associated with climate variability and land use and land cover changes. This study evaluated trends in annual minimum 7-day average streamflow (Q7), annual mean streamflow (Qmean), and maximum streamflow (Qmax), in ten gauging stations located in five river basins in the southern and central regions of Espírito Santo State, Brazil (Itabapoana, Itapemirim, Jucu, Novo, and Santa Maria da Vitória), over the period 1985–2023. Initially, Ljung–Box and Run tests were applied to assess serial autocorrelation and randomness in the time series. Trend detection was performed using the Mann–Kendall (MK) and Modified Mann–Kendall (MMK) tests, with magnitude estimated through Sen’s slope estimator, while abrupt change points were identified using the Pettitt test. The relationship between streamflow metrics, hydroclimatic variables (annual precipitation and reference evapotranspiration), and land use and land cover was investigated using Spearman’s rank correlation coefficient, considering time lags of 0, 1, 3, and 5 years. Results indicated no statistically significant trends for Qmean in any of the analyzed areas, whereas minimum flows (Q7) showed predominantly decreasing trends and maximum flows (Qmáx) exhibited increasing trends at part of the stations. Detected change points were mainly concentrated between 2013 and 2015, coinciding with severe drought events in southeastern Brazil. Correlation analysis revealed greater sensitivity of minimum and mean flows to atmospheric water balance, while maximum flows showed more localized associations with land use variables. Overall, the findings suggest that hydrological changes in the studied basins are primarily expressed through flow extremes rather than annual averages, highlighting the importance of integrating trend, change-point, and correlation analyses to better understand the evolution of regional hydrological regimes.
Keywords: hydrological trends; Mann–Kendall test; Sen’s slope estimator; Pettitt test; Streamflow.
