Effect of water quality parameters on the distribution of Pleuromamma (Copepoda-Calanoida) species in the Indian Ocean: a statistical approach
You are viewing information about the paper Effect of water quality parameters on the distribution of Pleuromamma (Copepoda-Calanoida) species in the Indian Ocean: a statistical approach.
|Journal:||Environ Monit Assess 2008/08/21|
|Authors:||Jayalakshmy, K. V.;Saraswathy, M.;Nair, M.|
|Address:||National Institute of Oceanography (CSIR), Regional Centre, Cochin 18, P.B.No.1913, Kerala, India, email@example.com.|
The region between 10 degrees N and 10 degrees S latitude was known to be congenial for distribution of Pleuromamma species. Diel and ontogenetic migrations were observed for Pleuromamma xiphias. Multivariate analyses such as factor analysis on species' abundance and predictive step-up multiple regression models of water quality parameters: temperature, salinity, dissolved oxygen and their first order interaction effects on Pleuromamma species' abundance, were carried out in the regions, off 10 degrees N, 10 degrees N to 10 degrees S, 10 degrees S to 30 degrees S, along the Somali Coast (38 degrees S-40 degrees S) and a fifth region along the western boundary of the above four regions. Relation between Pleuromamma species' abundance and the water quality parameters showed a decreasing trend in the variability explained (VE) from region 1 (VE = 55.19%) to region 4 (VE = 31.15%) in the models, indicating that the influence of other ecological factors was of higher significance in the south than in the north, with a north south gradient. Indices of diversity (Shannon-Weaver) and evenness (Heips) were calculated and lognormal distributions fitted for these indices were found to be a good fit (p < 0.05). The five regions were compared, based on critical ratio of the diversity index. Shannon-Weaver diversity showed higher values during night collections than day collections, justifying the tendency for diurnal variations. Diversity and dominance were highly (r = 0.95) correlated. A multilinkage cluster analysis by group averaging method for the species, based on the standardised values of log(10) (X + 1) transformed species' abundance, showed that Pleuromamma indica, Pleuromamma gracilis and Pleuromamma abdominalis were not segregated in any of the five regions. Bray and Curtis (1957) coefficient of similarity for the species in the five regions combined together, showed 50% similarity for the cluster containing circum-global species, Pleuromamma piseki, P. gracilis, P. abdominalis and P. borealis. The species, P. indica and P. xiphias entered this cluster at 22% similarity level. Multivariate factor analysis by row normalisation (for species) and column normalisation (for stations), using varimax rotation to simple structure for unique grouping of species as well as stations, carried out in the five regions, showed that the maximum number of species' clusters were obtained in region 2, indicating a higher variability in the ecological conditions in this region than in the other four regions. The significance of the factor model and the differential factor groups of species were also determined in the studied regions. On a broader scale, based on the ecological aspect, the first four regions could be reduced to two regions, viz, region 1 including the north off 10 degrees N, Central Indian Ocean and the area of subtropical convergence, and region 2 consisting exclusively of the equatorial region between 10 degrees N and 10 degrees S based on distribution of Pleuromamma species. The plot of the species abundance with respect to latitude also showed such a demarcation.