Delineation and explanation of geochemical anomalies using fractal models in the Heqing area, Yunnan Province, China

Jun Deng, Qingfei Wang, Liqiang Yang, Yanru Wang, Qingjie Gong, Huan LiuJournal of Geochemical Exploration 2010 · DOI

TL;DR

This paper applies two fractal models (concentration-area and number-size) to geochemical data from the Heqing area in China to delineate and explain geochemical anomalies. The study demonstrates that different geological events and rock types leave distinct fractal signatures in element distributions, offering a novel way to understand geological processes and mineral exploration.

What Problem Does This Paper Address?

Identifying geochemical anomalies from background levels is crucial for geological interpretation and mineral exploration. Traditional methods for setting these thresholds (like iterative mean ±20 or box-plots) can be insufficient when dealing with the inherently irregular and inhomogeneous distribution of elements in geological samples. This irregularity is often a hallmark of complex geological processes. The paper addresses the challenge of effectively separating these anomalies from background noise and linking them to specific geological origins in a complex region like Heqing, which hosts diverse geological formations including alkaline porphyry, volcanic rocks, and sedimentary sequences.

Methods

The study employs two primary fractal modeling techniques:

  • Number-Size (N-S) Model: This model describes the relationship between the cumulative number of occurrences of a certain concentration (or higher) and the concentration value itself. It's expressed as $N(\ge r) = cr^{-D}$, where $r$ is the concentration, $N(\ge r)$ is the cumulative number of samples with concentration $\ge r$, $c$ is a constant, and $D$ is the fractal dimension. Plotting $\ln N(\ge r)$ against $\ln r$ yields a straight line with a slope of $-D$ for fractal distributions.
  • Concentration-Area (C-A) Model: This model relates the area covered by a certain concentration contour to the concentration value. It's expressed as $A(r \ge r_i) = cr_i^{-D}$, where $A(r \ge r_i)$ is the area with concentration $\ge r_i$, $r_i$ is the concentration threshold, $c$ is a constant, and $D$ is the fractal dimension. ArcGIS was used for this analysis.

Both models are used to determine thresholds for separating geochemical anomalies from background levels. The paper also utilizes Principal Component Analysis (PCA) to group the 39 analyzed elements into four categories based on their spatial distribution and likely geological control.

Datasets

The study utilizes a dataset of stream sediment samples collected during a regional geochemical survey at a scale of 1:200,000 in the Heqing area, Yunnan Province, China. A total of 1672 samples were analyzed for 39 elements, including Au, Ag, Zn, Cu, Co, Ni, and U. No new benchmark datasets were introduced; the study focuses on applying existing fractal modeling techniques to this specific regional geochemical dataset.

Key Insights and Ideas

  • Fractal Signatures of Geological Events: The core idea is that different geological processes (sedimentation, volcanism, metallogenesis) and rock types leave distinct "fractal fingerprints" on the spatial distribution of elements. This allows for the differentiation of geological objects and events based on their geochemical anomaly patterns.
  • N-S vs. C-A Model Complementarity: While both models are effective in separating anomalies, they offer different perspectives. The N-S model is useful for understanding the overall structure of geochemical data, while the C-A model is better for analyzing the spatial distribution on maps. Their comparison reveals nuances in how different elements behave.
  • Element Grouping and Geological Control: PCA successfully grouped elements that are spatially correlated and likely controlled by similar geological processes. For instance, Au, Pb, and Zn are concentrated in the Beiya gold orefield (alkaline porphyry related), while Cu, Co, and Fe are enriched in areas with Emeishan mafic rocks.
  • Mobility Differences: The study suggests that Pb is more mobile than Au in the surface environment, as indicated by the larger anomaly area for Pb compared to Au.
  • HFSEs and Multiple Controls: High field strength elements (HFSEs) like Y and Zr show concentrations that overlap with both porphyry-related and mafic rock-related anomalies, suggesting a combined influence from hydrothermal activities.

Key Results

  • The N-S and C-A fractal models were successfully applied to determine geochemical thresholds for 39 elements.
  • The thresholds obtained from the two models were generally similar, allowing for the division of element concentrations into three segments, each correlating with specific rock types (alkaline porphyry, mafic rocks, sedimentary rocks).
  • Gold (Au) and Lead (Pb): These elements, grouped together by PCA, show similar fractal patterns. The N-S model identified three segments for Au, with the third segment (small fractal dimension) corresponding to gold-mineralized areas around Beiya. The C-A model also showed three segments, with a second threshold indicating a smaller mineralization area.
  • Copper (Cu) and Cobalt (Co): Anomalies for these elements are primarily found in areas with Emeishan mafic rocks, suggesting hydrothermal activity related to these volcanic rocks.
  • Yttrium (Y) and Zirconium (Zr): These HFSEs show high concentrations overlapping with both Au/Pb and Cu/Co anomalies, indicating a dual influence from alkaline porphyry and mafic rock-related hydrothermal systems.
  • Lithium (Li) and Uranium (U): These elements are enriched in Triassic areas and boundaries between Triassic and Permian mafic rocks, often associated with sedimentary rocks.

Limitations and Future Work

  • Model Differences: While generally similar, the fractal dimensions obtained from the N-S and C-A models are different, indicating they reveal data distribution from distinct perspectives. The paper notes that the C-A model can denote a much smaller area of mineralization than the N-S model.
  • Element Mobility: The study highlights differences in element mobility (e.g., Pb vs. Au) but doesn't delve deeply into the specific mechanisms.
  • Future Work: The authors suggest that understanding the fractal patterns of element distribution can help characterize various geological events and provide valuable insights for ore exploration in similar geological settings. Further research could explore the application of these models to other complex geological regions and investigate the specific geological controls on the observed fractal dimensions in more detail.

Summary updated

BibTeX
@article{deng2010delineation,
  author = {Jun Deng and Qingfei Wang and Liqiang Yang and Yanru Wang and Qingjie Gong and Huan Liu},
  title = {{Delineation and explanation of geochemical anomalies using fractal models in the Heqing area, Yunnan Province, China}},
  journal = {Journal of Geochemical Exploration},
  year = {2010},
  doi = {10.1016/j.gexplo.2010.04.005},
}
Figure