Physical attributes and water infiltration in soil under eucalyptus and pasture
Name: LUNALDA APARECIDA VAZ POLA
Publication date: 25/03/2024
Examining board:
Name |
Role |
|---|---|
| FABIO RIBEIRO PIRES | Examinador Externo |
| FERNANDO FRANÇA DA CUNHA | Examinador Externo |
| RENATO RIBEIRO PASSOS | Examinador Externo |
| ROBERTO AVELINO CECILIO | Presidente |
Summary: Understanding water infiltration into the soil is of great importance for adopting the best soil and water management practices. Therefore, this study aimed to evaluate the physical attributes and infiltration behavior in a soil under eucalyptus compared to its
previous use as pastureland. Infiltration was determined through ring infiltrometer tests at three different positions in the terrain, with three infiltration tests conducted at each position. To characterize the soil, soil density, total porosity, macroporosity,
microporosity, particle size distribution, total organic carbon, and penetration resistance were determined. Instantaneous infiltration equations were adjusted using the Kostiakov, Kostiakov-Lewis, Horton, and Philip models. The values found for
Horton were used to calculate the average and applied to the equations along with field data; the same approach was used with the average along with the average Tie. Exploratory analysis included clustering of points by Euclidean Distance, Principal
Component Analysis, and Pearson correlation. To evaluate the difference between terrain positions, the data underwent ANOVA and Tukey's test at a 5% significance level. The difference between land uses was assessed using the ANOVA F-test at a
5% significance level. Model performance was assessed using the coefficient of determination (R²), Nash-Sutcliffe efficiency coefficient (NSE), square root of the mean squared error (RMSE) and mean absolute error (MAE). There was no statistical
difference between the lower, middle, and upper positions of the areas, indicating homogeneity. Stable infiltration rates were moderate for both areas, with no statistical difference (63.13 and 95.16 mm h-1, for eucalyptus and pasture, respectively). Area
differences were more related to their physical properties, especially macroporosity, microporosity, clay, and sand content. The adjustments made showed good performance in most study points for the Kostiakov, Kostiakov-Lewis, and Horton models, with Horton being the most faithful to infiltration. The Philip model did not fit the field data well and performed the worst, with a mean MAE of 25.48 mm h-1 for the eucalyptus area and 47.98 mm h-1for the pasture area. Adjustments of the Horton equation using the average resulted in an MAE of 15.64 mm h-1for the eucalyptus area and 33.95 mm h-1for the pasture area. Using the average along with the average Tie increased the MAE to 25.54 mm h-1 for eucalyptus and 68.05 mm h-1 for the pasture area, therefore, its application is recommended only for the area under eucalyptus.
Keywords: Infiltration models, Land use change, Oxisol, Ring infiltrometer
