Cy. Chork et R. Salminen, INTERPRETING EXPLORATION GEOCHEMICAL DATA FROM OUTOKUMPU, FINLAND - AMVE-ROBUST FACTOR-ANALYSIS, Journal of geochemical exploration, 48(1), 1993, pp. 1-20
A factor analysis (FA) rendered robust using the minimum volume ellips
oid (MVE) estimator was carried out on geochemical till data consistin
g of 622 observations on 21 variables (Al, Ba, Ca, Co, Cr, Cu, Fe, K,
La, Li, Mg, Mn, Ni, P, Sc, Sr, Th, Ti, V, Y, Zn) from Outokumpu, Finla
nd. The distribution of the majority of variables is slightly to moder
ately positively skewed. A robust, three-factor model accounting for 8
2. 1 % of the total variance was fitted. The fitted model delineates z
ones related to geology, ore deposits, mineralized horizons, and backg
round variation. The robust FA generally outperforms an otherwise iden
tical classical FA with clearer definition of lithology, background va
riation, and, in particular, greatly sharpens the anomaly contrasts ov
er mineralization. Both classical and robust factor scores successfull
y define regional patterns with respect to geology, mineralization and
surficial environments. However, both the number and magnitude of sco
res exceeding 2.0 has increased, from classical FA to robust FA, for a
ll three factors. Such enhancements facilitate the identification of g
eochemical anomalies, particularly subtle anomalies of the kind common
ly associated with till data. This study indicates that the presence o
f outlying data points can exert a strong masking effect on the identi
fication of geochemical anomalies in classical FA.