الفهرس 
CONTENTESOnly 14 pages are availabe for public view
 |  | from | 123 |  |  |
| | | from | 123 | | |
Abstract
The areas under investigation are located at El-Kharga and El-Dakhla oases, New Valley Governorate, Egypt. The study areas included four sites, two sites in El-Kharga oasis (El-Monira and Bolaq) and the other two sites in El-Dakhla oasis (Zakhera and Mut).
The major goal of this study was evaluating and mapping the spatial variability of the soil available N, P and K using geostatistical technique. So, the objectives of this study were analyzing some physical and chemical properties of soil,determine and assessment the availability of soil N, P and K contents,evaluate the spatial variability of soil N, P and K by using geostatistical analysis andproduce maps for the spatial variation of soil N, P and K contents and some soil physico-chemical properties using ArcGIS.
One hundred and thirty–seven soil samples (50, 25, 30 and 32 samples) were collected from sites A (El-Monira), B (Bolaq),C (Zakhera) and D (Mut), respectively to represent the study areas. The samples were collected from the surface layer (0-25 cm) of the soil. They were taken using the systematically sampling grid within a distance between two consequent samples of 200m in sites A, B and C and 50m in site D.
The soil available N, P and K and some physical and chemical analyses of the soil samples (soil texture, SP, pH, ECe, OM and CaCO3) were determined according standard methods.
Data analyzedby using SAS software version 11 to get the statistical analysis including minimum, maximum, range, mean, standard deviation, coefficient of variation, skewness and kurtosis, which are generally accepted as indicators of the central tendency. Geostatistical analyses and the distribution maps of the soilavailableN, P and K contents and some physico-chemical properties were produced by ArcGIS (10.2.2).
In ArcGIS geostatistical analyst, the histogram and normal QQPlots tools were used to see what transformations were needed to make the data more normally distributed. Histogram and normal QQPlots analysis were applied for soil available N, P and K of the investigated soil samples to check its data to see if it has normal distribution or not.
Logarithmic transformation had been used for N, P and K content data to normalize too highly skewed and outlier data sets because kriging methods work best if the data is approximately normally distributed.Ordinary Kriging (OK) method was used in the present study as interpolation method because it is simple and has high accuracy for prediction in comparison to other kriging methods.
The obtained results indicatedthat soil texture of these soil samples is coarse and varies between sand, sandy loam, loamy sand, to loam. The results showed that 85.4% of soil samples were coarse-texture, while the soil fine-texture represented 14.6% of the total soil samples.
The soil saturation percentage(SP) values of the studied soil samples ranged from 18.4% to 59.3%.The highest values of saturation percentage were found for site (C), while the lowest one was recorded for site (D).
The soil reaction (pH) was moderate to strong alkaline as indicatedby pH values from 7.4 to 9.1.Also, the soil samples are non to highly saline asECe values ranged between 1.11 dSm-1 and 38.7dSm-1.
The calcium carbonate content (CaCO3)varied from 0.9 to 79.9%.The highest values of CaCO3 were found for site (A), while the lowest one was found for site (C).The organic matter content was low and their values ranged from nil to0.94%.
In general, sodium and calcium were the dominant soluble cations in most soil samples followed by magnesium then potassium. Also, soluble anions were dominated by chloride in most cases, followed by sulphates then bicarbonates.
On the other hand, the results showed that considerable variations in soil available contents of nitrogen, phosphorus andpotassium in all studied sites. The soil available N content varied from nil to 46.7mg/kg with mean of 25.2 mg/kg, from 23.4 to 46.7 mg/kg with mean of 27.1 mg/kg, from 23.4 to 70.1 mg/kg with mean of 46.3 mg/kg and from 23.4 to 46.7 mg/kg with mean of 33.6 mg/kg, in sites A, B, C and D, respectively.
The data revealed that thesoil available P content ranges from 4.4 to 85.0 mg/kg with mean of 18.6mg/kg, from 2.2 to 41.0 mg/kg with mean of 16.8 mg/kg, from 0.4 to 25.5 mg/kg with mean of 6.1 mg/kg and from 5.5 to 29.6 mg/kg with mean of 16.4 mg/kg, in these respective sites A, B, C and D, respectively. While, the soil available K varied from 52 to 1883 with mean of 359, from 217 to 6204 with mean 1050, from 268 to 1986 with mean of 721 and from 11 to 762 with mean of 168 mg/kg in the studied sites A, B, C and D, respectively.
The spatial interpolation of the soil available N, P and K contentsin the surface soil samples (0-25 cm) of the study area showed thatJ-bessel, K-bessel, Exponential, Spherical, Pentaspherical, Tetraspherical and Rational Quadratic models were the best performed in describing the spatial dependency of N, P and K nutrients which was a strong spatial dependence.
Also, the spatial interpolation of some physico-chemical properties in the surface soil samples of the study areasrevealed that Circular, J-bessel, K-bessel, Tetraspherical, Hole effect and Rational Quadratic models were best performed for texture, pH and OM, while Rational Quadratic,Tetraspherical,Pentaspherical, Circular and Hole effectmodels were best performed for SP, ECe and CaCO3.
The cross-validation demonstrated that the ordinary kriging technique was the best in describing the spatial interpolation of these nutrients. Soil available N, P and K stocks showed a strong spatial dependence, indicating that soil available N, P and K contents were mainly controlled by intrinsic factors. Geostatistical analysis integrated with GIS provided an opportunity to assess the variability in the distribution of these nutrients.
This study also indicated that the possibility of producing the spatial distributions maps for the soil content N, P and K availability, as well assome physico-chemical properties of soil. These maps can facilitate the estimation of the fertilizers amount of N, P and K nutrients and help in make decisions for choosing appropriate fertilization policies for soils as well as to avoid adding too high fertilizer levels to get aclean environment, also to avoid adding fertilizers for sites where do not in need for it. On the other hand, the study proved that statistics and geostatistics analyses are powerful tools to assess, understand and mapping the spatial variability of the soil available nitrogen, phosphorus and potassium.