Please use this identifier to cite or link to this item: http://krishikosh.egranth.ac.in/handle/1/5810145672
Authors: Muhsina, A
Advisor: Brigit Joseph
Title: Multivariate analysis for the classification of locations using soil parameters in central districts of Kerala
Publisher: Department of Agricultural Statistics, College of Agriculture, Vellayani
Language: en
Type: Thesis
Pages: 137
Agrotags: null
Keywords: Manova, PCA, Organic carbon, Multivariate analysis, Cluster analysis, Mahalanobis, Correlation analysis
Abstract: The research entitled ‘Multivariate analysis for the classification of locations using soil parameters in central districts of Kerala’ was conducted with the objectives to find out appropriate multivariate method to classify the locations based on soil fertility parameters and also to find out the interrelations among the soil parameters. Data on thirteen soil parameters of 17 panchayats (678 samples) in Ernakulam (EKM) and 28 panchayats (789 samples) in Kottayam (KTM) as per “Revolving Fund Mode Project- Soil Testing Lab” of Department of Soil Science and Agricultural Chemistry, College of Agriculture, Vellayani were subjected to various multivariate techniques. Initial data analysis using box plot technique was carried out to remove the outliers present in soil fertility parameters of both districts. Descriptive statistics of each soil parameters in the different panchayats in both districts were worked out and it was found that pH ranged from 4.2-5.8 in EKM with a coefficient of variation (CV) of 10.6 per cent, whereas that of KTM ranged from 4.61-5.95 with a CV of 11.28 percent. Oxidisable organic carbon (OC) (range: 1.23-2.47 %), phosphorus (range: 16.51-113.59 kg ha-1) and potassium (range: 45.86-570.21 kg ha-1) content had a CV of 46.84, 78.80 and 86.54 percent respectively in EKM while KTM showed a variation of 56.92, 79.24 and 60.21 percent respectively for OC (range: 0.98-3.39 %), P (range: 16.89-104 kg ha-1) and K (range: 179.14-580.64 kg ha-1). Multivariate analysis of variance (MANOVA) was done in order to test whether the panchayat means were significantly different with respect to 13 soil parameters. Results of MANOVA revealed that there was significant difference between the mean vector of 17 panchayats in EKM and 28 panchayats in KTM. Principal Component Analysis (PCA) generated five PCs which together accounted for 80 per cent variation in data of EKM and in KTM. Sampling adequacy for PCA and factor analysis was tested by Kaiser-Meyer-Olkin measure and it was found that sample sizes were adequate to conduct PCA in EKM (0.571) and KTM (0.464). Correlation among the variables was tested using Bartlett’s chi square test of sphericity and it was found that variables were correlated to each other. Variables were plotted using factor loadings and it was observed that EC, S and B had high positive loadings on factor 1 and 2 in EKM while none of the variables had positive loadings on F1 and F2 in KTM. Score plot obtained from PCA in EKM showed that Chengamanadu, Manjapra and Thirumarady panchayats had high content of available S and B. In Kottayam, Score plot drawn showed that Cu and Zn were predominant in Akalakkunnam, Kadaplamattom, Meenachil, Melukavu, Poonjar and Ramapuram panchayats. Hierarchical clustering (HC) and K –means clustering were performed to group the panchayats in both districts based on soil fertility status and thereafter comparison of various clustering procedures was done using Davies – Bouldin (DB) index. Different dissimilarity measures- Euclidean, squared Euclidean, Chebychev distance and Mahalanobis D2 were determined and single linkage, complete linkage and average linkage methods were adopted under these measures. The results in EKM showed that Mahalanobis D2 was the better clustering procedure with seven clusters (DB index: 0.120) followed by average linkage method under Euclidean distance (DB index: 0.306) with seven clusters. Manjapra (C6) and Keerampara (C7) remained as individual clusters. Keerampara had strongly acidic soils (pH -5.17) with high available Mg (158 mg kg-1) while Manjapra soils had low Mg availability (19 mg kg-1) and high S content (57 mg kg-1). Kakkad, Kalady and Vengoor came under C1 which possessed approximately same EC (0.15-0.19 dS m-1), OC (2-2.4%) and Mg (71-73 mg kg-1) content. Chengamanadu and Vengola came under C3 while Ayyampuzha and Mudakkuzha came under C4. Clustering of panchayats in KTM using Mahalanobis D2 resulted in better clusters as DB index was 0.120. Twenty eight panchayats in KTM were classified into 8 clusters. Only Akalakkunnam panchayat remained as a separate cluster with pH less than 4.8 and EC 0.08 dS m-1. Aymanam and Elikulam came under a cluster while Pallikkathode and Puthuppally came under another cluster. Bharananganam, Chempu and Melukavu came under a cluster as they had similarity in pH (5.35 -5.48), Mg (46-49 mg kg-1), S (23-27 mg kg-1) and Fe (23-29 mg kg-1). Inter relations among the variables were determined by using Pearson’s correlation and it was found that Ca, Fe and P was positively and significantly correlated with EC in KTM whereas S and P were positively related with EC in EKM. Mg and B were negatively related (-0.495). Similarly Fe and Mn had a negative correlation with each other (-0.467) in KTM district. Ca and B was negatively correlated in EKM (-0.525) which indicated the antagonistic effect. The results of the study indicated that Mahalanobis D2 gave the optimum clusters in both EKM and KTM followed by Euclidean distance using average linkage.
Description: PG
Subject: Agricultural Statistics and Informatics
Theme: Multivariate analysis
These Type: M.Sc
Issue Date: 2018
Appears in Collections:Theses

Files in This Item:
File Description SizeFormat 
174349.pdf
  Until 2021-01-01
23.59 MBAdobe PDFView/Open Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.