| 14 Oct 2025
U-Pb dating of chrysocolla from supergene copper deposits in the Coastal Cordillera of northern Chile, Atacama Desert
Abstract. The dating of supergene copper minerals has been widely used as a proxy to investigate the evolution and onset of hyperaridity in the Atacama Desert. However, investigation of supergene copper mineralisation in the Atacama Desert has been restricted to two physiographic units favourable for the industrial extraction of copper: the Central Depression and the Precordillera. Furthermore, these studies dated the timing of supergene mineralisation by secondary non-copper minerals like alunite. In this study, we present new results of LA-ICP-MS U-Pb dating of chrysocolla from supergene deposits hosted in the western part of the Coastal Cordillera of northern Chile. The obtained U-Pb ages range from 8.4 ± 1.2 Ma to 0.046 ± 0.027 Ma. Supergene mineralisation ages point to significantly reduced precipitation, necessary for leaching and mineral precipitation process, since the Late Miocene to Pleistocene in the Coastal Cordillera, later than the secondary supergene mineralisation ages from the Precordillera. Post-Mid-to-Late Miocene ages point to repeated phases of sufficient moisture along the Coastal Cordillera that promotes chrysocolla mineralisation during the Pliocene and Pleistocene. We propose that due to the position of the study areas near the coastal escarpment, and the predominant hyperarid environment in this part of the Coastal Cordillera since at least the Mid-Miocene, pluvial periods and/or intensification of coastal fog events caused alternating phases of supergene activity.
Competing interests: One of the author is editor of Geochronology (Dunai)
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
This manuscript presents interesting data on the ages of copper-bearing minerals from hypogene ore deposits in the Atacama desert in Chile. The results have important implications for regional climate history and so should be of interest to environmentalists as well as economic geologists. The writing is not bad considering that English is not the primary language of the authors. I have made minor suggestions for improvement on an annotated copy of the text.
The analytical work is extensive but important documentation is lacking and error propagation is questionable. Since all the Pb and U measurements were made together, the 207Pb/206Pb and 238U/206Pb data must be correlated, but correlation coefficients for all the U-Pb data are set to zero in the tables. Errors are reported on only 207Pb/206Pb and 238U/206Pb ratios, making it impossible to calculate the correlation values, which could be calculated from the formula: rho = (%SigX^2 + %SigY^2 - %Sig(X/Y)^2) / (2 * %SigX * %SigY) although it is better to calculate them in the data reduction program from the variance in X*Y. This will have the effect of rotating the error ellipses.
The data sets are presented on Tera-Wasserburg concordia plots blown up to the scale of the data spread. This does not give a very good feel for the reliability of the results. Most data sets have a limited spread near the Y axis, meaning that Pb is weakly radiogenic and the scope of each diagram is different. The ages are very young meaning that the intersections with the concordia curve, which define the ages, are well beyond the scope of the figures (the intersection would be at infinite 238U/206Pb for age zero). I suggest that the authors include at least one large-scale composite plot so the reader can get a better idea of the distance between the data and the concordia intersections.
Significance of the Y-axis intercepts is worthy of discussion in the manuscript. Regression on the T-W plot defines an initial 207Pb/206Pb ratio as well as an age. The initial ratio is more constrained than the age. Many, if not most, initial ratios from carbonates and phosphates accord with the simple crustal Pb growth model of Stacey and Kramers, which was defined using Pb from large ore deposits with diverse ages. This model predicts a 207Pb/206Pb ratio of about 0.84 for the late Cenozoic, which accords within error with most of the data sets. Significant deviations are almost always toward lower 207Pb/206Pb values as seen here for 3 of the samples. These could be due to mixing with radiogenic Pb released from the breakdown of other minerals during formation of the dated minerals. Are there any coeval high-Pb minerals in the assemblages that could be used to measure a more precise initial Pb ratio that could anchor the isochrons? The ratios that agree within error could be averaged and used to anchor individual isochrons giving more precise ages. Even using an artificial number with no error should result in ages that are accurate relative to each other, provided one can assume the same initial ratio.
I do not understand the rationale of expanding the data errors so as to get MSWD values near 1 (lines 167-169). Errors on the individual ratios should be determined based on the numbers of counts. MSWD values serve to check whether scatter can be accounted for by measurement errors alone or whether it may be affected by other factors such as late disturbance. A moderately high MSWD does not invalidate data, it just serves as a warning of possible complications.
It is not logical to add the dispersion of results on the standards to each other or to the sample. Adding in quadrature assumes that deviations are random, whereas biases are systematic. The sample data should be corrected linearly for the Pb/U bias determined using a matrix-matched standard. The ablation biases on monazite, titanite and zircon chosen as standards should be quite different from each other as well as the samples. Although there are no chrysocolla standards there are many carbonates that qualify (although there are also differences between carbonates like calcite and dolomite). The normal procedure is to use NIST only for correcting 207Pb/206Pb mass bias. A primary matrix matched standard is then used to correct 206Pb/238U, which is affected by both mass and ablation bias. A different secondary matrix matched standard is then run to confirm the correction of the first. This is especially important for spot analyses where ablation biases can be tens of percent and vary with pit depth. Ablation bias should be minimized (although not eliminated) by performing scans rather than spot analyses (Dix et al. 2021 doi.org/10.1016/j.chemgeo.2021.120582). I suggest that the authors analyze some carbonate standards, such as WC1, under the same conditions as their samples. Since the ages are so young, they can tolerate relatively large errors and still be useful but it is a shame to degrade such extensive results by poor calibration. In any case, the results on standards should be explicitly reported (e.g. give the measured 238U/206Pb ablation bias found for them).
Raw data files (usually CSV files) should be reported as Supplementary Data so results can be recalculated in future when we might have a better understanding of bias correction.
Don Davis