Fábio Alex Beling’s research while affiliated with Federal University of Santa Maria and other places

The Ibicuí basin, located in the south of Brazil, is close to 50,000 km2 in drainage area. The basin has big problems with water deficits, attributed to the indiscriminate use of water to irrigate rice. The objective of this study is a statistical analysis of water flow data in the Ibicuí basin to verify if there are significant trends in water availability related to the withdrawal of water for rice crop irrigation. We used data from 11 fluviometric stations for 1970 to 2011, corresponding to the period of major growth in rice cultivation. Records of daily flow data were normalized, then for each month, the flow at durations between 50% to 99% were calculated. Trends in these series were evaluated using the Mann-Kendall test. The results showed that there are trends of increasing water flow for 8 of the 11 stations, and in 6 of those 8 stations the increasing trend was statistically significant. Just 3 stations had negative trends and these were in sub-basins with higher percentage area in rice. Analyzing the trends for several flow durations, it was observed that there was a reduction of the trends with duration. Also, in a river with sequential stations, the significance of trends as reflected by the Mann-Kendall Zs decreased with irrigated area. We conclude that for the Ibicuí Basin analysis of trends in the flow data does not clearly reflect the effect of water withdrawals for irrigation of rice.

Simulating the hydrologic response of a watershed for different scenarios is an important tool for assessing the rational use of the land and natural resources, especially in environments where urbanization is not ever an organized procedure. This study used the Kineros2 event oriented hydrological model to simulate the runoff response of a 4.9 km² urban basin located in the Atlantic Forest biome in Southern Brazil, with 35% of the area being impermeable. The goal of the study was to estimate the characteristic parameters of soils and land cover for the watershed to enable the evaluation of basin response for different land uses. To achieve this objective, the responses of ten measured rainfall-runoff events were used to calibrate five parameters of the model. Two of these events were then used to simulate several scenarios. Using 100% forest land cover as reference, a scenario of 100% pasture land use increases runoff volume by 20% and peak flow by 50%. For the current land use (35% impermeable), the runoff volume is 78% higher and the peak 145% higher than the reference. For a scenario with 57% impermeable area, the runoff volume increases in average 124% and the peak 231%. For the most urbanized condition, with 78% impermeable area, the runoff volume increases in average 214% and the peak flow rate 470%.

This study aims at monitoring the behaviour of the rainfall, runoff, drainage, soil water storage, and evapotranspiration variables involved in the water balance measured by lysimeter data. The evaluation of the water balance considered different time scales, where the components were monitored daily and in 10-day accumulated period intervals. The results demonstrated that in wet periods the soil water content was greater at a depth of 10 cm, whereas in the dry periods a greater concentration was observed at 70 cm depth. At the depth of 30 cm, the lowest values of soil water content were observed for both wet and dry periods. The results, obtained through the use of tensiometers and time domain reflectometry installed internally and externally to the lysimeter, were very close, which was more noticeable during the periods of lower water loss by the soil. The water balance, calculated from the lysimeter data, demonstrated that 70% of the total rainfall was lost by the process of evapotranspiration. The drainage accounted for 27·5% of the precipitated water, highlighting the fact that this component should not be disregarded in the water balance calculation. Copyright © 2011 John Wiley & Sons, Ltd.