Water balance and soil erosion in the Brazilian Cerrado

Author(s)

Paulo Tarso Sanches de Oliveira

• Document type: Doctoral Thesis
• Press: São Carlos.
• Institution: Universidade de São Paulo (USP). Escola de Engenharia de São Carlos (EESC/SBD)
• Defense date: 2014-12-12

Examining board members

Edson Cezar Wendland; Silvio Crestana; Sergio Koide; Luisa Fernanda Ribeiro Reis; Humberto Ribeiro da Rocha

• Advisor: Edson Cezar Wendland

Abstract

Deforestation of the Brazilian savanna (Cerrado) region has caused major changes in hydrological processes. These changes in water balance and soil erosion are still poorly understood, but are important for making land management decisions in this region. Therefore, it is necessary to understand the magnitudes of hydrological processes and soil erosion changes on local, regional and continental scales, and the consequences that are generated. The main objective of the study presented in this doctoral thesis was to better understand the mechanism of hydrological processes and soil erosion in the Cerrado. To achieve that, I worked with different scales (hillslope, watershed and continental) and using data from experimental field, laboratory, and remote sensing. The literature review reveals that the annual rainfall erosivity in Brazil ranges from 1672 to 22,452 MJ mm ha-1 h-1 yr-1. The smallest values are found in the northeastern region, and the largest in the north and the southeastern region. I found that the canopy interception may range from 4 to 20% of gross precipitation and stemflow around 1% of gross precipitation in the cerrado. The average runoff coefficient was less than 1% in the plots under cerrado and that the deforestation has the potential to increase up to 20 fold the runoff coefficient value. The results indicate that the Curve Number method was not suitable to estimate runoff under undisturbed Cerrado, bare soil (hydrologic soil group A), pasture, and millet. Therefore, in these cases the curve number is inappropriate and the runoff is more aptly modeled by the equation Q = CP, where C is the runoff coefficient. The water balance from the remote sensing data across the Brazilian Cerrado indicates that the main source of uncertainty in the estimated runoff arises from errors in the TRMM precipitation data. The water storage change computed as a residual of the water budget equation using remote sensing data (TRMM and MOD16) and measured discharge data shows a significant correlation with terrestrial water storage change obtained from the GRACE data. The results show that the GRACE data may provide a satisfactory representation of water storage change for large areas in the Cerrado. The average annual soil loss in the plots under bare soil and cerrado were 15.25 t ha-1 yr-1 and 0.17 t ha-1 yr-1, respectively. The Universal Soil Loss Equation cover and management factor (C-factor) for the plots under native cerrado vegetation was 0.013. The results showed that the surface runoff, soil erosion and C-factor for the undisturbed Cerrado changes between seasons. The greatest C-factor values were found in the summer and fall. The results found in this doctoral thesis provide benchmark values of the water balance components and soil erosion in the Brazilian Cerrado that will be useful to evaluate past and future land cover and land use changes for this region. In addition, I conclude that the remote sensing data are useful to evaluate the water balance components over Cerrado regions, identify dry periods, and assess changes in water balance due to land cover and land use change. (AU)

• FAPESP's process: 10/18788-5 - Water balance and soil erosion in native forest of the Cerrado biome
• Grantee: Paulo Tarso Sanches de Oliveira
• Support Opportunities: Scholarships in Brazil - Doctorate