Upload
ngohanh
View
215
Download
0
Embed Size (px)
Citation preview
Processing Stream Sediment Geochemical Data for Geology and Dilution:
An Example from Northern Vancouver Island
Dennis Arne, PGeo (BC), RPGeo (AIG), Principal Consultant ‐ Geochemistry, CSA GlobalYao Cui, PGeo (BC), Senior Geomatics Geoscientist, BCGS
Rob Mackie, PGeo (MB), Principal Consultant – Geology, CSA GlobalOlivia Brown, Associate Database Consultant, CSA Global
www.csaglobal.com 1Geoscience BC Day, October 8, 2015
• Statement of problem• Background to catchment analysis• Results from northern Vancouver Island• Conclusions
www.csaglobal.com
Outline
2
• The presence of regionally extensive basalts of the KarmutsenFormation having elevated Cu makes it difficult to detect Cu anomalies associated with porphyry Cu and other Cu‐bearing mineral deposits on northern Vancouver Island.
• How can we filter out the effects of high regional background Cu and so add value to re‐analyzed regional geochemical survey (RGS) samples and new stream sediment data?
www.csaglobal.com 3
Statement of Problem
Raw Cu Data
• Geology is interpreted in terms of lithology, rather than map units.• The distribution of Cu is clearly influenced by the Karmutsen basalt.
www.csaglobal.com 4
Importance of Lithology
Mt Hall Gabbro?
• The most important principal components from regional stream sediment geochemical data usually reflect bedrock lithology
• PC1 from northern Vancouver Island defines mafic lithologies
www.csaglobal.com 5
Principal Component Analysis
Data Cloud in 2D
From Peters, 1987
• PC1 provides a better discriminator for the Karmutsen basalt
www.csaglobal.com 6
PC1 vs Raw Cu Data
Raw Cu Data Inverse PC1
• Catchment size–Controls dilution
• Catchment geology– Introduces variable geochemical background and differential erosion
• Variable relief–Controls erosion rates in different parts of the catchment
• Scavenging of metals by various sample constituents–Adsorption onto secondary Fe and/or Mn hydroxides, clays or organics
• Stream water pH & Eh–Controls solubility of metals and secondary hydroxides
• Effects of exposed mineralization– Introduces anomalous data populations
• Sample representativeness–Historical RGS Au data is not very reproducible
www.csaglobal.com 7
Drainage Geochemistry Affected by:
• Raw data• Residuals following regression against control elements• Data levelled by dominant catchment lithology• Data levelled by presence/absence of a particular unit• Data levelled by catchment weighted background value• Multiple regression analysis using catchment geology• Residuals following regression against principal components• Productivity analysis• Weighted sums model• Weighted sums model adjusted for catchment area size
www.csaglobal.com 8
Data Processing Approaches
• Data compiled for 1835 samples in total• Catchment basins delineated for 1725 samples
www.csaglobal.com 9
North Vancouver Island Catchments
• Entire catchment area delineated for each sample (nested)• Locations validated against original maps and adjusted to TRIM hydrology
www.csaglobal.com 10
Sample Validation/Catchments
• Cu shows a positive correlation with Al• Regression against Al removes some of the lithological effects
www.csaglobal.com 11
Relationship Between Cu and Al
Cu Residuals
• Many elements show lithological control• Statistical outliers are preserved during Z‐score levelling
www.csaglobal.com 12
Levelling by Dominant Lithology
Raw Data Levelled Data
• Levelling by dominant lithology or presence/absence of basalt both reduce lithological effects
www.csaglobal.com 13
Levelling Cu Data for Lithology
Levelled by Dominant Lithology Levelled by presence/absence basalt
• Average background values have been estimated for most lithologies• These have been used to Z‐score level the Cu data
www.csaglobal.com 14
Weighted Background Values
Z‐score Cu “Residuals”
• Cu can be regressed against PC1 (lithological control)• High positive residuals represent Cu above background
www.csaglobal.com 15
Regression Against PC1
www.csaglobal.com 16
Weighted Sums Model
• The user assigns importance rankings to selected variables based on a‐priori knowledge of the mineralogy/chemistry of a target deposit type
• Positive rankings used for those elements that are expected to be high (greater than median) in the target deposit type
• Negative rankings used for elements expected to be low (less than median) in the target deposit type
For porphyry Cu example, using levelled data: Cu Mo Ag Au*Zn
*raw data 3 2 1 1 1
Garrett & Grunsky, 2001
• Geochemistry in catchment areas >10 km2 mainly reflects regional background• Catchments > 10 km2 have not been effectively sampled
www.csaglobal.com 17
Dilution Effects
Sampling Effectiveness
Regional Background
• Productivity analysis takes into account the effects of dilution
www.csaglobal.com 18
Productivity Analysis
Introduced to account for the downstream dilution of the geochemical signature from a mineral deposit exposed within a catchment.
MemAm = (Mea ‐Meb)Aa + MebAmwhereMea = the metal content at the mouth of catchment
Aa = the total area of the catchmentMem = the metal content of the eroding mineral deposit
Am = the area of the target exposed at surfaceMeb = background metal content in the catchment
However, for a small deposit size relative to the area of the catchment: MemAm = (Mea ‐Meb)Aa
This is also known as the Productivity
WBV Cu Productivity
(From Rose et al., 1979)
• Data processing methods were validated against known occurrences
www.csaglobal.com 19
Method Validation
Cu Occurrences No Cu Occurrences
>90t
hPercentile
<90t
hPercentile
TruePositives
FalsePositives
FalseNegatives
TrueNegatives
Single Element Methods
Cu Produ
ctivity
Weighted Sums M
odel
WSM
* Catchment A
rea
•The effects of elevated background Cu are easily filtered using a variety of methods of varying sophistication.
•Processing methods using several commodity and pathfinder elements weighted for the effects of dilution work best.
•Existing Cu occurrences & deposits on NVI are more readily apparent and new targets evident in the processed data.
•An empirical assessment of sampling effectiveness indicates there are still exploration opportunities in under‐sampled areas.
•There is still much value to be obtained from the interpretation of RGS data, especially where archived material is re‐analyzed by ICP‐MS.
•Our approach to geochemical data should be more like our approach to geophysical data where filtering and data enhancements are routine.
www.csaglobal.com 20
Conclusions
• 41 page report• Compiled data file with
geochemistry• Shapefile of catchments
attributed with derived values• PDF copies of report figures and
catchment maps
www.csaglobal.com 21
Final Products
Available from http://www.geosciencebc.com/s/Home.asp Example of thematic catchment basin map
Porphyry Cu WSM * Area