Correlation, Method of Least Squares, Linear Regression and the General Linear Model - Geomathematics: Theoretical Foundations, Applications and Future Developments - page 130

Table 4.5 Abitibi copper
hindsight study; Amounts
of copper for ten cells with
most copper in 2008

Rank Cell #,# Type Copper (kT)
1 20,4 1,0,0 4,516.1
2 37,9 11,0,0 1,710.1
3 37,8 1,0,0 1,436.8
4 48,10 4,0,0 576.1
5 41,9 5,0,0 241.2
6 47,10 1,0,1 166.5
7 36,8 2,0,0 114.9
8 38,7 2,0,0 92.6
9 35,7 2,0,0 90.3
10 34,1 1,0,0 79.9
Largest value (rank 1) for Kidd Creek Mine accounts for 47.5 %
of total Abitibi production and reserves. Cell numbers as in
Fig. 4.10 . Three numbers in column for “Type” are for VMS,
Ni-Cu and porphyry copper deposits per cell

10 6

10 5

10 4

10 3

10 2

10 1

10 0
10 −2 10 −1 10 0 10 1

Copper Grade (%)

Fig. 4.11 Log-log plot of Ore Tonnage versus Copper grade (2008 data). The three points on the
left are probably outliers. When these three points are deleted, the correlation coefficient
( r ¼ 0.079) is nearly zero suggesting lack of functional relationship between grade and ore tonnage
(Source: Agterberg 2011 , Fig. 3)

deposits in the Abitibi area had stabilized and further increases in production and
reserves (from 5.23 to 9.50 Mt) were for copper within the known deposits and for
new discoveries close to (and generally at greater depths than) the known deposits.
Average grade of total production and reserves is about 1.6 % copper in the original
1968 data set with 41 copper deposits as well as in the 2008 data set with 66 copper
deposits. Figure 4.11 is a log-log plot of copper grade versus amount of ore for the
corresponding 35 copper cells that also will be analyzed in more detail in the next
section.
Predictions made in Agterberg et al. ( 1972 ), such as the one for a “test area” in
the surroundings of Timmins, Ontario, were based on the assumption that the