Volume 12, Issue 46 (1-2009)
jwss 2009, 12(46): 609-622 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Ayoubi S, KHormali F. Spatial Variability of Soil Surface Nutrients Using Principal Component Analysis and Geostatistics: A Case Study of Appaipally Village, Andhra Pradesh, India. jwss 2009; 12 (46) :609-622
URL: http://jstnar.iut.ac.ir/article-1-1147-en.html
URL: http://jstnar.iut.ac.ir/article-1-1147-en.html
, ayoubi@cc.iut.ac.ir
Abstract: (34867 Views)
Understanding distribution of soil properties at the field scale is important for improving agricultural management practices and for assessing the effects of agriculture on environmental quality. Spatial variability within soil occurs naturally due to pedogenic factors as well as land use and management strategies. The variability of soil properties within fields is often described by classical statistical and geostatistical methods. This research was conducted to study what factors control the spatial variability of soil nutrients using an integration of principal component analysis and geostatistics in Appaipally Village, Andra Pradesh, India. 110 soil samples were randomly collected from 0-30 cm and prepared for laboratory analyses. Total N, available P, Ca, K, Na, Mg, S, B, Mn, Fe, Zn were measured using standard methods. Statistical and geostatistical analysis were then performed on raw data. The results of PCA analysis showed that 4 PC's had Eigen-value of more than 1 and explained 71.64 % of total variance. The results of geostatistical analysis revealed that three PC's had isotropic distribution based on surface variogram. Spherical model was fitted to all PC's. Ranges of model were 288 and 393 m for PC1 and PC3 respectively. On the other hand the range for PC2 was significantly different (877m). The most important elements in PC2 such as Fe, Mn, and Zn probably had similar range of effectiveness (700-900m). The comparison of PC's distributions indicated that PC1 and PC3 including total N, available Mg, K, Cu, Ca and P, were in accordance with farming plots dimensions and management practices. Therefore, it is necessary to improve the appropriate fertilizers used by farmers. The pattern of PC2 distribution was not consistent with farmer's plots, but had the best concordance with soil acidity. Therefore, the most correlated elements with this PC including Fe, Mn, and Zn are mainly controlled by soil acidity and not affected by management practices. However, spatial variability of these elements in areas lower than critical values should be considered for site-specific management.
Keywords: Spatial variability, Principal component analysis, Geostatistics, Soil nutrients, Appaipally village, India
| Rights and permissions | |
 | This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
