AuthorGuma'a, Guma'a Sayed.
Water in agriculture -- Arizona.
Soil moisture -- Measurement.
Soils, Irrigated -- Arizona.
Committee ChairWarrick, Arthur W.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractSpatial variation of in situ available water content was studied along with related parameters over three 16-ha irrigated fields. The fields, two near Marana (Pima County, Arizona) and one near Casa Grande (Pillai County), range in texture from very fine loam to loamy sand. All soil series present are mapped as Entisols or Aridisols. A 50-m grid provided 56 sampling sites in each field. Each site was sampled at 30, 60, 90, 120 and 150 cm. Samples were collected from each field following a heavy irrigation in March 1977. Bulk samples were collected two days and four weeks after the application of approximately 300 mm of water, to determine in situ water content at field capacity and moisture redistribution with time. Related parameters such as particle size distribution and soil water characteristics were also studied. Bulk density and saturated hydraulic conductivity were determined from undisturbed, core samples. The measured parameters showed different patterns of variation within the same field as well as from one field to the other. Spatial variability of saturated hydraulic conductivity was the highest for which coefficient of variability (CV) ranged upward to 108%. Bulk density, on the other hand, showed the lowest coefficient of variability, as low as 5%. The in situ available water content (AWC), estimated by subtracting moisture content at 15 bars from the corresponding in situ FC values, showed a general tendency to increase with depth corresponding to the increase in percent silt plus sand with depth in all three fields. The coefficient of correlation between the two parameters was high (up to 0.70). The mean values of AWC as estimated using 0.1 bar values for field capacity in the laboratory were consistently higher than the in situ values. The values were within 25 - 35% of each other in Fields 1 and 2, while in the sandier soil of Field 3, the AWC was overestimated by an average of 74% in the laboratory. The CV showed an irregular tendency to increase with depth, but was consistently high in the 150 cm layer in all three fields. Values estimated in the laboratory showed lower CV and higher correlations with soil separates than in situ AWC in all three fields. These two observations can be attributed to the elimination of in situ factors such as texture stratification, compaction, and/or amount of water applied. Agricultural soil formed on water transported material at 0.1 bar were highly correlated with sand (r = -0.8) and the 15 bar values were better correlated with clay (r = 0.5). Also, the coefficient of variability increased consistently with decreases in moisture content. The analysis of variance showed the three fields to be heterogeneous. The variation for within and between the 5 depth groups was significant. A two-way interaction between depths and subareas within each field accounted for 44, 45 and 38% of the total variability in Fields 1, 2, and 3 respectively. Cumulative frequency distribution plots, full normal plots, Kolmogorov-Smirnov tests of goodness-of-fit, tests of skewness and tests of kurtosis were conducted to test the null hypothesis of normal distribution for each parameter. The full normal plots, being sensitive to deviations from normality, rejected the null hypothesis in all cases with few exceptions. They showed the data tends to be skewed to the right and/or kurtic. The alternative frequency distribution of the parameters indicated the data to be asymmetric, short tailed with the exception of percent sand which was symmetric, short tailed for all three fields. A power transformation is suggested as a possibility for transforming the data to get near normal distribution.
Degree NamePh. D.
Degree ProgramSoils, Water and Engineering