Fourth Principal Component Research Articles

Soil Property Analysis using Principal Components Analysis, Soil Line, and Regression Models

Past research has attempted to relate surface characteristics of soils to reflectance from remotely sensed images to provide a means for quantifying spatial heterogeneity. Existing procedures have proven valuable, but no research has been performed to compare these techniques. The objective of this research was to compare existing methodologies, that is, principal components analysis (PCA), Chen et al.'s regression model, and the soil line Euclidean distance (SLED) technique, for quantifying spatial heterogeneity in soil surface organic matter (OM) and cation exchange capacity (CEC). The three existing techniques were compared using five bare soil images from three different silt loam to loam fields in the Midwest USA. At the same time as image acquisition, surface (upper 2.54 cm [1 in]) soil properties were measured in situ. Organic matter and CEC were highly correlated (R2 > 0.70) to the first principal component (PC1) for three bare soil images, moderately correlated (R2 > 0.40) for one image, and only slightly correlated (R2 < 0.25) for the final image. The lower correlations were hypothesized to be because of the range in the soil OM and CEC and image exposure. Principal Component 1 accounted for approximately 95% of the total variance in all the fields; therefore, no correlation was observed between the upper 2.54 cm (1‐in) surface soil properties and the second, third, or fourth principal components (PC2, PC3, and PC4, respectively). All three techniques equivalently predicted OM and CEC. However, PCA does not require field‐specific regression or soil lines parameters. It is also suggested that PC1 can replace the soil line in a technique for identifying soil‐sampling locations.

주성분분석에 의한 재래종 옥수수의 해석

육종재료를 얻기 위하여 부산·경남지역에서 수집된 재래종 옥수수 49 계통을 선발하여 본 실험을 실시하였다. 본 시료는 주성분분석을 이용하여 재래종 옥수수를 해석하고 계통분류를 실시하였던 바 다음과 같은 결과를 얻었다. 7 개의 형질을 이용하여 실시한 주성분분석에서는 제 4주성분까지를 가지고 전체 변동의 86.3%를 설명할 수 있었고, 제 2 주성분까지는 전체 변동의 67.4%를 설명할 수 있었다. 주성분에 대한 형질들의 기여율은 형질에 따라 달랐고 상위 주성분에서 켰으며 하위 주성분에서 작았다. 주성분과 형질과의 상관계수는 주성분의 생물학적 의의와 주성분에 대응한 식물체의 형을 명확히 하였는데 제 1 주성분은 식물체의 크기 및 생장기간에 관련된 주성분이었고, 제2주성분은 이삭수와 분얼수에 관련된 주성분이었다. 제 3주성분과 제 4 주성분에서는 형질간에는 유의성이 인정되지 않았다. This study was conducted to get basic information on the Korean local corn line collected from Busan City and Kyungnam Province, a total of 49 lines were selected and assessed by the principal component analysis method. In the result of principal component analysis for 7 characteristics, 67.4% and 86.3% of total variation could be appreciated by the first two and first four principal components, respectively. Contribution of characteristics to principal component was high at upper principal components and low at lower principal components. Biological meaning of principal component and plant types corresponding to the each principal component were explained clearly by the correlation coefficient between principal component and characteristics. The first principal component appeared to correspond to the size of plant and ear, and the duration of vegetative growing period. The second principal component appeared to correspond to the number of ear and tiller. But the meaning of the third and fourth principal components were not clear.

幼児の肥満要因と母親の食意識・食行動の関連について

We investigated the relationship between the primary factors for obesity in preschool children and the eating behavior and dietary consciousness of their mothers. Physical signs, biochemical data and frequency of food intake were collected, and a questionnaire on living conditions and life styles was conducted on 245 preschool children aged 4-5 years and their mothers who lived in suburban and rural areas.The physical characteristics of the children were a height of 105.2±4.5cm (mean±SD) and a weight of 17.7±2.7kg. The degree of obesity was 3.3±11.7%, and 26 (10.6%) of the children were considered to be obese. The rate of skipping breakfast was 25%, while the same rate for dinner was 7% among all the subjects. There were some significant differences between the non-obese and obese children regarding their eating behavior.For example, the rate of “eating quickly” in the obese group was higher than in the non-obese group. In respect of the dietary consciousness of the mothers, their attitude toward cooking and their interest in food labeling were not related to the childrens' obesity.To assess the relationship between the primary factors for obesity in the children and the dietary consciousness of the mothers, a principal component analysis was performed regarding the overall dietary consciousness and eating behavior of the mothers. The following four factors were identified: the method of providing meals (first principal component), the mothers' interest in nutrition (second principal component), the mothers' interest in calorie control (third principal component), and the frequency of eating out (fourth principal component).The score for the third principal component was significantly higher in the obese group than in the non-obese group. Some significant relationships were apparent between the individual factor score of the principal components and the eating behavior of the children. The variables of “have a snack after dinner”, “enjoyment while eating” and “eating quickly” showed a positive correlation, and “eating slowly” while “chewing slowly” showed a negative correlation with the third principal component; namely, carefully controlling the caloric intake.

주성분 분석을 이용한 거세한우의 체형분류에 관한 연구

1997년부터 2001년까지 (사)한국종축개량협회의 전국한우능력평가대회에 출품된 642두의 거세한우의 도축 전까지 4회에 걸쳐 주요 10개 부위의 체형 측정치를 조사하여 주성분 분석을 이용하여 체형을 분류하였으며, 각 체형 측정치의 주성분 지수와 도체 형질과의 상관관계를 분석하였으며, 그 결과를 요약하면 다음과 같다. 1. 1차, 2차, 3차 및 4차 체형 측정치의 principal component(이하 P.C) 1의 분산은 각각 7.610, 8.297, 7.269 및 5.739로서 전체분산 중 76.0%, 83.0%, 72.7% 및 57.4%를 나타낼 수 있었고, P.C 1, P.C 2, P.C 3의 합은 전체분산 중 각각 86.69%, 90.49%, 84.62% 및 77.26%를 나타났다. 2. 각 차수별 체형 측정치의 P.C 1의 모든 계수가 양(+)이어서 일반적으로 좋은 체구의 크기를 나타내었으며, 체구의 크기에 가장 크게 영향을 미치는 형질은 흉심(0.328<TEX>${\sim}$</TEX>0.339)과 요각폭(0.325<TEX>${\sim}$</TEX>0.341)이였다. 또한 P.C 2, P.C 3 및 P.C 4는 각 차수별로 다양하게 나타났다. 3. 1차, 2차, 3차 및 4차 체형 측정치의 주성분 지수들과 도체형질간의 상관계수 추정에서 대부분의 분산을 차지하는 P.C 1은 도체중(0.539<TEX>${\sim}$</TEX>0.755), 일당증체량(0.256<TEX>${\sim}$</TEX>0.564), 등지방두께(0.227<TEX>${\sim}$</TEX>0.280) 및 배최장근단면적(0.187 <TEX>${\sim}$</TEX>0.344)간에 정(+)의 상관을 보였으며, 육량지수(－0.246<TEX>${\sim}$</TEX>－0.110)와는 부(－)의 상관을 보였고, 근내지방도(0.066<TEX>${\sim}$</TEX>0.099)와는 대부분 유의적인 상관을 보이지 않거나 낮은 상관을 보였고, P.C 2, P.C 3 및 P.C 4는 주성분 지수들의 값에 따라 도체형질과 다양한 상관관계를 보였다. 단일 체형형질과 도체형질간의 상관 그리고 P.C 1의 주성분 지수와 도체 형질과의 상관간에는 큰 차이가 없었다. Data were consisted of the ten body measurements (withers height, rump height, body length, chest depth, chest width, rump width, rump length, thurls width, hipbone width and chest girth) of 642 steers (Korean cattle), which was entered in the National Beef Quality Contest hosted by the Korea Animal Improvement Association from 1997 to 2001. A principal components analysis was used to classify the body types of the steers, and estimate the correlations between carcass traits and principal components for the body measurements of the first, second, third and fourth period, respectively. The first principal component of body measurements at the first, second, third and fourth period accounted for 76.0%, 83.0%, 72.7% and 57.4% of the total variance, respectively. The sum of first, second and third principal component at each period accounted for 86.69%, 90.49%, 84.62% and 77.26% of the total variance, respectively. At each period, all the first principal component of the body measurements were positive and it generally showed large framed body shape. The size of body was influenced mostly by chest depth(0.328<TEX>${\sim}$</TEX>0.339) and rump length(0.325<TEX>${\sim}$</TEX>0.341). The second, third and fourth principal component at the each period were various. There were positive correlations between principal components index of each period and carcass traits such as carcass weight(0.539<TEX>${\sim}$</TEX>0.755), average daily gain(0.256<TEX>${\sim}$</TEX>0.564), backfat thickness(0.227<TEX>${\sim}$</TEX>0.280), and eye muscle area(0.187<TEX>${\sim}$</TEX>0.344). The correlation with yield grade index(－0.246<TEX>${\sim}$</TEX>－0.110), however, was negative. The correlation with marbling score(0.066<TEX>${\sim}$</TEX>0.099) was low or statistically insignificant. According to principal component indexes of the second, third, and fourth components, the correlations with the carcass traits were various. There were no large differences between the correlations of the single body measurement trait with the carcass traits and the correlations of the first principal component indexes with the carcass traits.

Inter‐specific variation in Macropus crania: form, function and phylogeny

AbstractMorphological variation was studied in the crania of seven species of kangaroos and wallabies (Macropus). How function and phylogeny interact to influence cranial form amongst adults of diverse species was considered. Thirty‐six, three‐dimensional landmarks were acquired on each of 65 adult Macropus specimens from Western Australia. The different species occupy a wide range of habitats and latitudes and there are well‐known differences in diet among the species. Size and shape variability was examined using the methods of geometric morphometrics. Aspects of inter‐specific variation explained by principal components were visualized and examined for correlations with size, latitude, and diet. The first principal component correlates strongly with inter‐specific differences in latitude and it primarily contrasts species on the basis of the relative length of the snout anterior to the molar teeth, this being greater in species that live in cooler climates. Species scores on the second principal component correlate strongly with centroid size such that smaller crania have more flexed muzzles. Principal component three separates the short grass‐eating, red kangaroo M. rufus, from all other species (most of which eat tough herbage from trees and shrubs). This principal component represents differences in the masticatory apparatus. Variation on the fourth principal component also seems to be functionally interpretable in terms of further differences in the masticatory apparatus of M. agilis. The relationship between phylogeny and cranial form was investigated by comparing a phenogram based on the shape distances between taxa with a consensus phylogeny drawn from the literature on Macropus genetics, ecology and morphology. The phenogram shows considerable similarity to current taxonomic views, but, our results also indicate differences that are related to function. Thus, this study serves to unravel the interaction of functional and phylogenetic influences on cranial form amongst Macropus species and raises further issues relating to ontogeny and phylogeny.

Mince gels with hydrocolloids and salts: composition/function relationships and discrimination of functionality by multivariate analysis

Multivariate statistical methods were applied to gelling properties (mechanical, water holding capacity, and color) of blue whiting mince with varying combinations of hydrocolloids and different ranges of salts. Factor analysis identified five principal components which accounted for over 90% of the data variance. The role of the salts in the various functional properties was independent of the hydrocolloid used. The most important contributor was CaCl2, present in the first three factors, which negatively affected most of the target parameters and positively influenced elasticity and lightness. KCl and NaCl appeared to have a less drastic effect on the studied properties; they presented high factor loadings in the fourth and fifth principal components, respectively. Three clusters were obtained: 1) locust bean and guar gum; 2) carrageenans (iota and kappa), carboxymethylcellulose and sodium alginate; and 3) xanthan gum. Basically, breaking deformation and hardness were the variables that best discriminated in 96.4% of cases.

Relationships between meat quality measurements in rabbits fed with three diets of different fat type and content

Two hundred and eighty-four, genetically similar (a three-way cross), young rabbits were fed ad libitum, from weaning, either a commercial diet (group C, ether extract 2.6%) or a diet containing vegetable fat (group V, ether extract, 9.9%) or animal fat (group A, ether extract 11.7%). A principal component (PC) analysis was performed with the variables: ultimate pH at 24 h post mortem measured in the longissimus dorsi (LD) and in the biceps femoris (BF) muscles, colour measured on the surface of the loin, fatty acid composition of perirenal fat, meat fat content of the hind leg, water holding capacity and cooking losses of the meat, and sensory variables determined by a trained panel test. The four first PC explained 62% of the total variation (27, 13, 11 and 11%, respectively). The first PC grouped the fatty acids, the second PC grouped the sensorial variables, and the third and fourth PCs grouped the pHs and the water holding capacity. The projection of the data in the first two PCs showed three separate groups of points. Animals fed with diet V were on the left side of the graph, where the variable C18:2 lies, whereas animals fed with diets A and C lay on the right side of the graph, where the saturated acids were grouped. These were slightly separated by the higher content of oleic acid in the animals fed with diet A. The second PC, where the sensorial variables were grouped, did not separate the animals fed with diets A, V and C. The diets used in this experiment had only a slight influence on the organoleptic characteristics of rabbit meat.

圧力処理炊飯米の老化の特徴

We researched into the characteristic of retrogradation of cooked rice treated by pressure after soaking. By β-amylase pullulanase (BAP method), the degree of gelatinization of the nonpressure treated rice immediately after cooking was 95.7% and 79.1% after 5 days. On the other hand, the degree of gelatinization of the pressure-treated rice immediately after cooking was 98.4% and 92.9% after 5 days. Retrograded rice was scanned by a Differential Scanning Calorimeter (DSC) .ΔH was 14.3 mJ/mg to the cooked rice immediately after cooking, and 10.4 mJ/mg to the rice treated by pressure. Furthermore, with the storage time ΔH in the rice came closer to ΔH in the rice flour with water. But in this case, d H in the pressure-treated cooked rice was more lower than d H in the nonpressure-treated cooked rice. The results of Near Infrared Reflectance (NIR) analysis were as follows: The first principal component score (coefficient of determination 90.5%) correlated to its storage term (R2=0.775). A difference between the pressure-treated and the nonpressure-treated ricewas observed by plotting the first and the fourth (c.d. 0.5%) principal component scores. With a value calculated by d H and a value measured by the BAP method as a purpose variable and the first through the fourth principal component scores as explanation variables, the multiple correlation analysis was calculated by the addition and reduction method. The fourth principal component score (R2=0.513)was primarily obtained, then the first principal component score (R2=0.818), and then the third principal component score (R2=0.865) significantly appeared. When the retrograded rice was heated with a microwave, the degree of gelatinization of nonpressure-treated cooked rice was 93.1 to 96.8%, but that of pressure-treated cooked rice was 100%. Even after 30 days, the retrogradation pressure treated rice showed a high degree of gelatinization by micro-wave heating.

Survey of environmental complex systems: pattern recognition of physicochemical data describing coastal water quality in the Gulf of Trieste.

A data set reporting temperature, salinity, dissolved oxygen, nitrogen as ammonia, nitrite and nitrate, silicate, chlorophyll a and phaeopigment values, determined in seawaters sampled during two years with a monthly frequency in 16 stations in the Gulf of Trieste, and at different depths of the water column, has been studied. In order to find synthetic descriptors useful for following the spatial and temporal variations of biogeochemical phenomena occurring in the considered ecosystem, the data set has been factorized using principal component analysis. A graphical display of scores, by means of boxplots and biplots, helped in the interpretation of the data set. The first factor conditioning the system is related to the input of freshwater from the estuary of the Isonzo River and to the stratification of the seawater (thermohaline discontinuity), while the second and third components describe interactions between biological activity, nutrients and physicochemical parameters; typical spring and autumn phytoplankton blooms were identified, in addition to an exceptional winter bloom conditioned by anomalous meteorological/climatic conditions. The fourth principal component explains the reducing activity of seawaters, which often increases when the decomposition of organic matter is relevant. The simple linear model proposed, and the related graphs, are shown to be useful tools for monitoring the main features of such a complex dynamic environmental system. The outlined approach to the considered complex data structure presents in a cognitive easy way (graphical outputs) the significant variations of the data, and allows for a detailed interpretation of the results of the monitoring campaign. Temporal and spatial effects are outlined, as well as those related to the depth in the water column.

Multivariate examination of data from gait analysis of persons with stroke.

Gait analyses yield redundant information that often is difficult to interpret. The purpose of this study was to show how principal-component analysis can provide insight into gait data obtained from persons with stroke. Twenty male and 11 female adults who were ambulatory were studied (mean age = 60.5 years, SD = 11.8, range = 24-79; mean time since stroke = 11.4 months, SD = 15.4, range = 2.0-88.0). Spatial data were used in a 4-segment link-segment model to calculate the kinematic and kinetic variables of gait. Principal components were constructed on the averages for 40 variables. The first principal component was related to speed and accounted for 40.8% of the variance. The second principal component was related to differences between the 2 limbs (symmetry) and accounted for 12.8% of the variance. The third principal component was related to adoption of a postural flexion bias and accounted for 10.2% of the variance. The fourth principal component, which was not interpretable, accounted for 6.8% of the variance. The principal-component analysis allowed clustering of related variables and simplified the complex picture presented by the large number of variables resulting from gait analysis. Examination of variables closely related to each principal component yielded insight into the nature of the strategies used in walking and their interrelationships. The method has potential for insight into similarities and differences in gait performances arising from different pathologies and for comparing the progress of individuals with similar pathologies.

Southwest monsoon rainfall in India: Part I-Spatial variability

Results of Principal Component Analysis (PCA) in spatial mode (S-Mode) applied on a 10 years' (1977-86) data set of weekly rainfall anomalies are presented in this study. The rainfall activity has been below average in the period under study with high standard deviations mostly over low rainfall areas. Inter - subdivisional correlation values suggest predominance of broad- scale weather systems over most parts of the country .The first principal component (PC) resemblance with the mean pattern. The second PC has been associated with active monsoon condition. The third and fourth PCs have been related to the northward progression of the Madden-Julian oscillation and the weak monsoon condition respectively.&#x0D; Resemblance of spatial patterns of alternative PCs with typical strong and weak monsoon activities suggest significant contributions from certain parts of the country (northeast, southeast) towards overall rainfall activity during weak monsoon situation.

Visible and near-infrared spectral changes in plasma of psychiatric patients

In the present study, the possible difference in visible and near-infrared (Vis-NIR) spectra of plasma in psychiatric patients compared to healthy control subjects was investigated. Mean Vis-NIR spectra of 60 psychiatric patients showed several specific peaks in the 630- to 930-nm region. Principal component analysis (PCA) and soft independent modeling of class analogy (SIMCA) were applied to manifest the spectral characteristics of each group. The PCA and SIMCA models demonstrated successful discrimination of the two groups using a score plot of the third versus fourth principal component and a Coomans plot, respectively. The PCA loadings were generally consistent with the discriminating power of the SIMCA model, indicating a specific pattern of Vis-NIR spectra in the plasma of psychiatric patients. The specific pattern possibly indicates altered superoxide, amine, and primary and secondary amides in plasma and may provide a possible indicator for the diagnosis of heterogeneous psychiatric disorders. This is the first step toward a clinical tool which has the potential to be applied in the diagnosis of psychiatric disease. Yet, there remain obstacles to its practical usage.

Principal components analysis of morphological measures in the Québec family study: Familial correlations.

Principal components analysis was carried out on 13 morphological dimensions collected in the first phase of the Québec Family Study (weight, height, fat mass (FM), fat-free mass (FFM), body surface area, six skinfolds, arm and calf girths). The first four principal components (PCs) account for 85.9% of the total phenotypic variance in the anthropometric dimensions. The first PC has almost equal loadings on the 13 variables, indicating its role as an overall body size and adiposity measure. This interpretation is supported by its high intraindividual correlations with other composite indices such as the BMI and sum of six skinfolds (SF6). An interesting finding compared to other studies dealing with PCs of only skinfolds is that the second component appears to contrast measures of body fat with those that are fat free or that represent overall body size, thus indexing a different dimension to the fat patterning. It explains 17% of the total phenotypic variance. The commonly observed component for the trunk-extremity contrast is represented as the fourth principal component. The third component primarily indexes height, although an extremity-trunk contrast is also weakly represented. Together, the 3rd and 4th components explain ∼10% of the total variance. The intraindividual correlation of PC3 with the trunk-extremity skinfold ratio (TER) is moderately negative (-0.307), whereas that of PC4 is highly positive (0.764). Analysis of familial correlations (parent-offspring and sibling) shows that the familial effect (transmissibility) ranges 0.43-0.75 for the first four PCs. The lack of significant spouse correlation for most of the PCs suggests that these familial effects may be primarily genetic. Am. J. Hum. Biol. 9:725-733, 1997. © 1997 Wiley-Liss, Inc.

コンピュータグラフィックスによる質感表現の研究

We have compared differences in texture between computergraphics displays and the actual materials. We evaluated the appearance of 35 actual samples (building materials, exterior materials etc.) and 40 samples generated by computer graphics using the Semantic Differential method.Using Principal Component Analysis, we found that approximately 94% of appearance characteristics can be explained by five principal components. The appearance characteristics that can be described using the Phong model are the first principal component (gloss), aspects of the second principal component (density and weight), and the fourth principal component (clarity of appearance). The balance of the second principal component (density and weight) and the fifth principal component (metallic appearance), cannot be explained by the Phong model.Glossiness is proportional specular reflectance divided by diffuse reflectance and specular reflectance, and the cubicroot of the exponential variable representing the surface glossiness. These relationships show how texture changes with diffuse or specular reflectance, transmittance and the exponent representing the surface glossiness.

Identification of the nature and source of atmospheric aerosols near the Taj Mahal (India)

The chemical composition of aerosol samples collected at Agra near the Taj Mahal during April 1991-June 1992 was identified by wet chemical analysis. The average concentration of suspended particulate matter (SPM) was 368.5 µg m(-3), ranging between 83 and 1305 µg m(-3), depending upon the season. Elevated levels of Na, SO4, Mg, NO3 and Cl compared to levels reported worldwide were attributed to the suspension of soil particles, as well as industrial emissions. Geometric mean enrichment factors of elements indicated two groups; one having enrichment factors less than 7 which were comprised of Na, K, Ca, Mg, Fe, Al, Mn and Si (crustal) and the other having enrichment factors greater than 13 and which were comprised of Ni, Cu, Zn, Pb, and Cd (non-crustal). Principal component analysis revealed the association of the first principal component with soil-derived elements while the second, third and fourth principal components were associated with industrial processes, wood combustion and brick kilns respectively. The study indicates that near the Taj Mahal the dominance of natural sources may enhance the degree of deterioration of the marble surface if micro-climatological conditions favour its wetting.

Spatial and temporal variations of the january circumpolar vortex over the northern hemisphere

AbstractTemporal and spatial variations of the Northern Hemisphere's circumpolar vortex are analysed for each January from 1947 to 1990. Our data set is developed by digitizing the 546‐dam isoheight from 50‐kPa charts and recording the latitude of this contour at each 10° meridian. Principal components analysis is applied to the resulting 699‐day by 36‐variable matrix to examine the temporal and spatial evolution of the circumpolar vortex.The first principal component is comparable to the Pacific‐North America (PNA) teleconnection pattern. Over the 44 Januarys analysed, the frequency of the positive phase of the PNA pattern has increased. Principal component 2 is similar to the Eurasian pattern and varies between axially asymmetric zonal flow and a three‐wave pattern centred over the pole. The asymmetric pattern, which is characterized by higher heights over western Europe and Scandinavia, has increased in frequency over the study period. With the third principal component, which is similar to the East Atlantic teleconnection pattern, the strength of the Icelandic Low varies markedly. Time series of scores on the third component indicate that 50‐kPa heights over the North Atlantic are lower over the past 20 years. The fourth principal component is related to vortex area. The northern vortex has occupied a more expanded position since the mid‐1960s. These first four principal components explain half of the total variance in the data.In toto, these results indicate that the January circumpolar vortex has expanded over the past two decades. Most of the expansion has occurred in the vicinity of the Aleutian Low and has produced increased meridional flow throughout much of the Western Hemisphere. This circulation change could account for the observed increased occurrence of warm air masses in winter over eastern Alaska and western Canada, increased cyclone occurrence in the north‐western Atlantic, and both the increased frequency of Arctic outbreaks and declining temperatures in the south‐eastern USA. The overall vortex expansion is contrary to general circulation model forecasts that predict a decrease in the meridional temperature gradient of the Northern Hemisphere as trace‐gas concentrations increase.

Application of an Experimental Design to the Detection of Foreign Substances in Milk

NIRA was used in order to study interactions between temperature (T) and soluble substances and to detect solutes added to milk. A two-level factorial design was planned to estimate the effects of the main factors and of the most interesting interactions on spectral data. Water ( W), NaCl ( N), reconstituted skim milk powder ( P) and W with N were alternately added to milk. The plan was repeated on eight different milks. The interactions studied were W × N, W × T, N × T, P × T, W × N × T. The values assigned to the experimental design model were 1 and —1 for active and non-active factors, respectively. This model and the corresponding analysis of variance were assessed independently at each wavelength, making it possible to plot results versus wavelengths as a spectrum. Thus, absorbance (predicted variable) was considered as the response to experimental design variations (predictors). Principal Component Analysis (PCA) was applied to the overall data collection to identify the main variation factors. The ability of principal components to predict the presence of foreign substances was assessed by Principal Component Regression. Several PCRs were calculated on two sub-collections (HIGH and LOW temperature), obtained from the overall collection (TOTAL), in order to assess whether recording spectra at two temperatures was useful to predict the presence of foreign substances. PCA on the overall data collection identified T as the most important factor affecting the total spectra variation; N and P groups were also separated according to the third and the fourth principal components. The parameters of the analysis of variance confirmed that the main factors N, P and T and the T interactions were more significant. T and N induced a shift of water absorption bands around 1450 and 1940 nm. There was no prediction improvement in recording milk spectra at two temperatures. PCA on the LOW sub-collection identified W and P according to the first factor and N according to the second one.

Chemometric differentiation of raw and commercial milk by trace elements using principal component analysis

Nine trace elements (Cr, Mn, Fe, Ni, Cu, Zn, Mo, Cd, and Pb) were determined in the dissolved ash of 36 samples of raw milk. The distribution of the concentration of each element was first investigated by means of a test of normality. The matrix of the correlation between the concentrations of the elements was then used as a starting matrix for principal component analysis. Nine variables were reduced to four principal components, accounting for 75% of the total variance. The biophilic elements Mn-Fe and Cu-Mo were positively associated with the first two principal components, while Cr was correlated to the third and Ni and Cd with the fourth principal component. Pb and Zn are both negatively correlated to the first principal component. Comparison with 42 samples of a commercial milk, by using a two-dimensional plot of the principal component scores, rendered possible the differentiation between raw and commercial milk.

Petrologic Factors Controlling Coal Gas Contents of Fruitland Formation Coals, San Juan Basin, New Mexico

The petrologic factors controlling coal-bed gas content in coals of the Fruitland Formation in the central San Juan basin are abstruse, with many of the factors being complexly interrelated. Data from this study were derived from 78 samples collected from seven cored wells in Rio Arriba and San Juan counties, New Mexico, and include desorption gas content, proximate analyses, maceral modal analyses, vitrinite reflectance, x-ray diffraction mineralogy, and pyrolysis values. Although several of the measured parameters affect gas content in a general sense, no single petrologic factor is strongly correlated to desorbed gas content. Principal component analysis was used to group the measured parameters into associations of related variable. The first principal component, accounting for 34% of the total variation in the data set, is dominated by parameters that vary strongly with organic matter maturity, namely vitrinite reflectance (R{sub o}), volatile matter and fixed carbon contents, hydrogen index, and pyrolysis-generated thermally extractable hydrocarbons (S{sub 2}). Ash content and other measured parameters correlatable to it account for 19% of the total variation as the second principal component. The third principal component (13%) is controlled by variation in the maceral content and the fourth principal component (8%) is primarily comprised of thermallymore » extractable hydrocarbons (TEH or S{sub 1}), hydrogen index, and as-received moisture content. Reification and interpretation of the principal component analysis are dependent upon an understanding of the Fruitland depositional system, the burial diagenesis of the coal-bearing strata, and coal-bed gas desorption phenomena.« less

Multivariate data analysis of sea waters and mussels in relation to pollution sources of trace elements.

The total concentration of Cu, Zn and Pb in surface sea waters from the Bay of Muggia (Gulf of Trieste, Northern Adriatic Sea) was determined by anodic stripping voltammetry. The association of these trace elements in relation to the known sources of pollution was discussed. The content of eight trace elements (Mn, Co, Ni, Cu, Zn, Cd, Hg, Pb) in the soft part of mussels (Mytilus galloprovincialis Lamarck) is also considered. The wild molluscs were sampled in the harbour of Trieste, in the proximity of an important city sewer. Principal component analysis was used to analyse the correlation matrix obtained from an 8 x 43 data matrix after a logarithmic transformation of the concentration variables. Eight variables were reduced to four principal components, which explained 80% of the total variance. The orthogonally rotated factor matrix shows that Co, Ni, Cd, and Pb are associated with the first principal component, Cu and Zn to the second, Hg to the third and Mn to the fourth principal component. The results of this multivariate data analysis are compared with those already obtained from two sampling sites in the Bay of Muggia and the origin of some trace metals in the soft part of mussels from the Gulf of Trieste is discussed.