الفهرس 
ADDRESSOnly 14 pages are availabe for public view
 |  | from | 32 |  |  |
| | | from | 32 | | |
Abstract
Spatial variability of soil physical properties in newly reclaimed area. Abdel Aziz Mohamed Talaat Mostafa Fahmy. Ph.D. Thesis, University of Ain Shams, Faculty of Agriculture, Department of Soil Science, 1998.
The current work was conducted to evaluate the spatial variability in some basic soil physical properties. Such properties included mechanical fractions, bulk density, infitlration rate, sorptivity and hydraulic conductivity.
The study was carried out in the area of Burg EI-Arab Agriculture
Experiment Station, Desert Research Center (DRC). The soils of this area is mainly sandy clay loam and highly calcareous (45-65% CaC03).
Three transects were laid out. The first was on W-E direction. and
extended 185 m, the second was on N-S direction for 220 m and the third was on NW-SE direction for 350m. The number of sampling sites were
38, 45 and 51 for the 1st, 2nd and 3rd transects, respectively. At each site soil samples were collected at 0-25 and 25-60 em depth.
Variabilities of soil physical properties were determined using two
statistical methods; namely, Classical statistics and Geostatistics.
1. Soil physical properties:
The conventional statistics indicated that gravel and, in few cases, sand displayed skewed frequency distribution which fitted the lognormal model better than the normal one. Therefore, the optimum number of samples required for determing the mean gravel content within I 0% of the true mean varied from 85 to 128 in the topsoil and from Ill to 167 in the subsoil. Meanwhile, the number of samples required for the determination of sand ranged from 50 to 79 and from 92 to 136 for the topsoil and subsoil, respectively.
On the other hand, data of silt and clay fractions, except few cases,
displayed approximately symmetric normal frequency distribution. Consequently, the optimum number of samples required to determine their mean values was relatively low and ranged between 30 and 52 for silt and from 36 to 60 for clay. The data also revealed that the coefficient of variations (CV) for mechanical separates ranged from I 0 to about 40%,
indicating intermediate variability.Calcium carbonate: Calcium carbonate contents in the topsoil and subsoil
fitted the normal frequency distribution. The CV ranged between 10.02 and
12.92%, indicating intermediate variability.
Bulk density: The obtained results indicated that bulk density values in top and subsoil fit the normal frequency distribution. CV values ranged between 4 and I 0%, indicating very low variability.
Infiltration rate:The values of infiltration rate along the three transects fit
the Gaussian or normal frequency distribution. TheCV ranged between
50-61 %, indicating intermediate variations.
Sorptivity: Sorptivity data revealed that the CV values ranged between 44 and 52%, indicating intermediate variability and fitted the lognormal frequency distribution.
Hydraulic conductivity: The data fitted the lognormal frequency
distribution and the CV values ranged between 35-47%, i.e. intermediate variability.
2. Geostatistics of soil physical properties:
Particle size fractions: The obtained results indicated that, most of the mechanical separates along the different geographical directions exhibited spatial correlation with distance.
Calcium carbonate: Calcium carbonate values indicated the presence of
highly significant spatial dependence.
Bulk density: Bulk density data revealed the presence of highly significant
spatial dependence.
Infiltration rate: Infiltration data delineated significant spatial independence.
Sorptivity: Sorptivity data declared that its distribution was random (pure nugget effect).
Hydraulic conductivity: Hydraulic conductivity data showed no
significant correlation (spatial dependent) in the topsoil. While the reverse
was true for the subsoil layer.
3. Soil chemical properties:
Soil chemical properties including salinity, cation exchange capacity and
pH have been discussed.