Non-linear principal components analysis: an alternative method for finding patterns in environmental data

2005 ◽  
Vol 17 (1) ◽  
pp. 1-11 ◽  
Author(s):  
R. N. Ellis ◽  
P. M. Kroonenberg ◽  
B. D. Harch ◽  
K. E. Basford
Keyword(s):  
Principal Components Analysis ◽  
Principal Components ◽  
Environmental Data ◽  
Alternative Method ◽  
Non Linear ◽  
Components Analysis

scholarly journals Multivariate Statistical Analysis Applied to Water Quality of a Tropical Coastal Lagoon, Cartagena, Colombian Caribbean

2019 ◽  
Vol 20 (1) ◽  
pp. 141
Author(s):  
Ildefonso Baldiris-Navarro ◽  
Juan Carlos Acosta-Jimenez ◽  
Angel Dario Gonzalez-Delgado ◽  
Alvaro Realpe-Jimenez ◽  
Juan Gabriel Fajardo-Cuadro
Keyword(s):  
Water Quality ◽  
Principal Components Analysis ◽  
Principal Components ◽  
Coastal Lagoon ◽  
Coastal Lagoons ◽  
Oxygen Demand ◽  
Habitat Destruction ◽  
Environmental Data ◽  
Fecal Coliforms ◽  
Components Analysis

Coastal lagoons are one of the most threatened ecosystems in the world, because of population growth, habitat destruction, pollution, wastewater, overexploitation and invasive species which are the main causes of their degradation. The objective of this paper was to evaluate the water quality behavior in a stressed coastal lagoon in Cartagena, Colombian Caribbean. Environmental data was analyzed using hypothesis testing, confidence intervals, and also Principal components analysis (PCA). The study was focused on water parameters such as dissolved oxygen (DO), biochemical oxygen demand (BOD5), chemical oxygen demand (COD), salinity, pH, total dissolved solids, total coliforms (TC), Fecal coliforms (FC), ammonium (NH4+) and total phosphorus (TP). The analysis was conducted in line with the Colombian national water standard. Results showed that BOD5, COD, phosphorus, and coliforms are out of the limits for these variables in Colombia and are reaching levels that may be a threat to human health. Principal components analysis detected five components that explained 79.4% of the variance of data and showed that anthropogenic and temporal factors might be affecting the variation of the parameters.

Non-linear principal components analysis using genetic programming

1999 ◽  
Vol 23 (3) ◽  
pp. 413-425 ◽  
Author(s):  
H.G. Hiden ◽  
M.J. Willis ◽  
M.T. Tham ◽  
G.A. Montague
Keyword(s):  
Genetic Programming ◽  
Principal Components Analysis ◽  
Principal Components ◽  
Non Linear ◽  
Components Analysis

Vegetation studies at Polar Bear Pass, Bathurst Island, N.W.T. II. Vegetation–environment relationships

1983 ◽  
Vol 61 (6) ◽  
pp. 1637-1646 ◽  
Author(s):  
J. W. Sheard ◽  
Dorothy W. Geale
Keyword(s):  
Correlation Analysis ◽  
Principal Components Analysis ◽  
Canonical Correlation Analysis ◽  
Principal Components ◽  
Canonical Correlation ◽  
Environmental Gradient ◽  
Environmental Influence ◽  
Environmental Data ◽  
Components Analysis ◽  
Soil Moisture Status

Vegetation–environment relationships are defined with the aid of principal-components analysis and canonical correlation analysis. In both the uplands and lowlands a moisture gradient, determined by measuring gravimetric moisture and indicated by organic carbon, is the most important environmental influence on the vegetation. In the uplands this gradient is also associated with snow depth (drifting) and in the lowlands with conductivity. The second environmental gradient in the uplands is associated with depth to permafrost and its soil textural correlates. Thus soil texture, independent of its effect on soil moisture status, influences the distribution of plant communities. In the lowlands the second environmental gradient is less clear but is also associated with depth to permafrost and, in addition, elevation and CaCO3 equivalent. Canonical correlation analysis shows that the components extracted by principal-components analysis of the vegetation data did not conform to the important trends of variation in the environmental data. Principal-components analysis is nevertheless an essential means of data reduction prior to the application of canonical correlation. The statistical model used in the study has potential advantages over the independent use of either principal-components analysis or canonical correlation.

Sign in / Sign up

Export Citation Format

Share Document