A statistical model for predicting carbon monoxide
levels

Abdul‐Wahab, Sabah A.; Al-Rubiei, Raid; Al-Shamsi, Ali

doi:10.1504/ijep.2003.003318

Cited by 7 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…8 -11 Most standard statistical procedures, such as linear regression analysis and analysis of variance, deal with mean values of data sets. 12 However, PCA does not fall into this category. PCA is a multivariate statistical technique that considers the structure of the variability and interdependence between random variables in a multidimensional data set.…”

Section: Pca Techniquementioning

confidence: 99%

See 1 more Smart Citation

Source Characterization of Atmospheric Heavy Metals in Industrial/Residential Areas: A Case Study in Oman

Abdul‐Wahab

2004

Journal of the Air & Waste Management Association

Self Cite

View full text Add to dashboard Cite

Total suspended particulate matter (TSP) was collected from two districts: the Sohar industrial estate (SIE) and a residential area in the case study area in Oman. The TSP collected from SIE was taken from 19 different industrial activities, whereas those collected from the residential area were taken from 12 houses. The samples were analyzed for nine heavy metals: chromium (Cr), copper (Cu), manganese (Mn), nickel, lead (Pb), zinc (Zn), cadmium, vanadium, and molybdenum. The results were used to assess the source classes responsible for TSP and heavy metal levels in the atmosphere of the two districts. At each district, principal component analysis was applied to the concentrations of TSP and heavy metals to obtain the number of principal components. At SIE, three groups of elements were determined. The first component contained large factor loadings in Cu and Mn (Cu industrial processes and reinforcement steel production). The second presented most of the variance of Cr, Pb, and Zn (black and galvanized iron pipes production, mechanical industries, and vehicle construction). The third component was composed of Zn and Pb that probably reflected contributions from motor-vehicle tire wear or was associated with municipal incineration. On the other hand, results from the village showed that only one factor was able to explain the main part of the data variance, and the industrial site was polluting air quality in the village with Cu, Cr, Pb, nickel, and Mn. INTRODUCTIONEmission of heavy metals associated with particulate matter (PM) is a point of concern and care because of the harmful effects of such metals on humans and the environment. 1 Measurement of heavy metals in PM is, therefore, of immense importance for toxicological, environmental, and occupational health studies. 2 Because of their chemical stability, primary heavy metals are good fitting species for mathematical purposes. They are not subjected to transformations during transport from source to receptor. Therefore, the concentrations of metals in atmospheric particles are a function of their sources. In addition, important information relating to sources of PM pollutants can be derived by knowing the chemical composition of airborne PM. 3 For the effective management of air quality, great importance is attached to the identification of the sources of suspended particulates. This is the main objective of source apportionment modeling. 4 Source apportionment, also called receptor modeling, provides an estimate on the PM contribution of various sources to the levels at the receptor. It is a key component necessary for developing and achieving desired air-quality objectives. 5 The method relies on the principle that if a group of chemical constituents have a common origin, they should show a similar variation. It is applied to describe the relationships among different heavy metals. The results of source apportionment can be used to evaluate emission reduction on the PM levels and to devise more efficient emission reduction strategies. 5 Hence, e...

show abstract

Section: Pca Techniquementioning

confidence: 99%

“…PC methods also are used for selecting subsets of variables for a regression equation. 12 PCs are ordered in such a way that the first PC explains most of the variance in the data. Each subsequent one accounts for the largest proportion of variability that has not been accounted for by its predecessors.…”

Section: Pca Techniquementioning

confidence: 99%