Características físico-hídricas dos solos da estação experimental da embrapa-algodão: variabilidade espacial.

Abstract

The study aimed to characterize physic and hydrically the Embrapa Experimental Station soils located at the Irrigated Area at Sao Goncalo-PB, Brazil and analyze the spatial variability of the characteristics. For this, 40 sites were selected and on these soil samples at depth intervals of 0-30, 30-60 and 60-90cm, were collected. The soil properties analyzed were the texture, global density, particle density, total porosity, field capacity, permanent willing point, available water for the plants, soil water retention curve, infiltration capacity and hydraulic conductivity. The sampling was systematic obeying a linear criterion, with prc-fixcd distances among points. To evaluate the spatial variability of the data, statistical methods, based on measures of position and dispersion, were used. In general the soil presents a sandy loam texture and therefore high values of global density. The particle density values are those normally found on sandy loam soils.. Both, the global density and the total porosity suffer the effect of the soil compaction due to high traffic of agricultural implements. The water contents at the field capacity and at the wilting point are lows to moderates, founding the higher values at the soil surface. Each soil depth has a different behavior with respect to water retention, due to the different physical properties. The soil has a low infiltration capacity, with an initial infiltration capacity of 10.5 cmh'1 decreasing to a constant value of 1,21 cmh"1 , denominated basic infiltration.. The Kostiakov Model was more adequate to estimate the water infiltration, than the I lorton Model. The saturated hydraulic conductivity values found show a soil with a moderately low drainage. The results show a great vertical and horizontal spatial variability of the analyzed properties. With exception of the particle density, the properties showed different degrees of variation with accentuated standard errors, amplitudes and variation coefficients. Of the studied properties, the hydraulic conductivity was the one that showed the higher variability Also it was the only property whose frequency distribution did not have a normal distribution. Due to the great vertical variability, sampling requirements increased with depth.