Articles | Volume 15, issue 4
23 Jul 2008
 | 23 Jul 2008

Quantification of tillage, plant cover, and cumulative rainfall effects on soil surface microrelief by statistical, geostatistical and fractal indices

J. Paz-Ferreiro, I. Bertol, and E. Vidal Vázquez

Abstract. Changes in soil surface microrelief with cumulative rainfall under different tillage systems and crop cover conditions were investigated in southern Brazil. Surface cover was none (fallow) or the crop succession maize followed by oats. Tillage treatments were: 1) conventional tillage on bare soil (BS), 2) conventional tillage (CT), 3) minimum tillage (MT) and 4) no tillage (NT) under maize and oats. Measurements were taken with a manual relief meter on small rectangular grids of 0.234 and 0.156 m2, throughout growing season of maize and oats, respectively. Each data set consisted of 200 point height readings, the size of the smallest cells being 3×5 cm during maize and 2×5 cm during oats growth periods. Random Roughness (RR), Limiting Difference (LD), Limiting Slope (LS) and two fractal parameters, fractal dimension (D) and crossover length (l) were estimated from the measured microtopographic data sets. Indices describing the vertical component of soil roughness such as RR, LD and l generally decreased with cumulative rain in the BS treatment, left fallow, and in the CT and MT treatments under maize and oats canopy. However, these indices were not substantially affected by cumulative rain in the NT treatment, whose surface was protected with previous crop residues. Roughness decay from initial values was larger in the BS treatment than in CT and MT treatments. Moreover, roughness decay generally tended to be faster under maize than under oats. The RR and LD indices decreased quadratically, while the l index decreased exponentially in the tilled, BS, CT and MT treatments. Crossover length was sensitive to differences in soil roughness conditions allowing a description of microrelief decay due to rainfall in the tilled treatments, although better correlations between cumulative rainfall and the most commonly used indices RR and LD were obtained. At the studied scale, parameters l and D have been found to be useful in interpreting the configuration properties of the soil surface microrelief.