Part 2 New Techniques and Modeling
Chapter 2 Detrital zircon U-Pb geochronology: current methods and new opportunities GEORGE GEHRELS Department of Geosciences, University of Arizona, Tucson, USA ABSTRACT Detrital zircon geochronology is rapidly evolving into a very powerful tool for determining the provenance and maximum depositional age of clastic strata. This rapid evolution is being driven by the increased availability of ion probes and laser ablation inductively coupled plasma (ICP) mass spectrometers, which are able to generate ages efficiently and with sufficient accuracy for most applications. Although large numbers of detrital zircon ages are generated each year, significant uncertainties remain in how data are acquired, which ages are used, how data are plotted, and how age distributions are compared. Improvements in current methods may come from enhanced precision/ accuracy of age determinations, better tools for extracting critical information from age spectra, abilities to determine other types of information (e.g., Hf, O, Li isotope signatures, rare earth element (REE) patterns, cooling ages, structural information) from the dated grains, and availability of a database that provides access to detrital zircon age determinations from sedimentary sequences around the world. Keywords: zircon; geochronology; detrital; provenance; analytical methods
IN TR ODUCTION Detrital zircon geochronology has evolved rapidly during the past 20 years, from a technique with apparently limited application to a nearly indispensible method of investigating sedimentary units and their source regions. This evolution, referred to informally as the “DZ revolution,” is largely a result of technical developments that have allowed researchers to efficiently determine U-Pb ages on individual zircon crystals (see Davis et al. [2003] and Kosler and Sylvester [2003] for excellent reviews of this history). Prior to these developments, most detrital zircon analyses were conducted on multigrain fractions. The prevailing strategy was to analyze groups of grains with similar characteristics (e.g., color, shape, rounding, etc.) in hopes that the grains in each fraction were of the same age (e.g., LeDent et al., 1964; Hart and Davis, 1969; Girty and Wardlaw; 1984; Erdmer and Baadsgaard, 1987; Gehrels et al., 1990; Ross and Bowring, 1990). While sufficient
to determine general age distributions, this procedure was generally inadequate to identify specific source ages. Analysis of individual zircon crystals became feasible during the 1980s and early 1990s utilizing both ion probes (e.g., Froude et al., 1983; Dodson et al., 1988; Ireland, 1992) and isotope dilutionthermal ionization mass spectrometers (ID-TIMS) (e.g., Davis et al., 1989). Technical developments that allowed for ID-TIMS analyses of individual crystals included preparation of a 205 Pb spike (Krogh and Davis, 1975; Parrish and Krogh, 1987) and design of Teflon microcapsules for low-blank grain dissolution (Parrish, 1987). During the late 1980s and early 1990s, ion probes and ID-TIMS techniques became increasingly used for detrital zircon geochronologic studies. A dramatic increase in detrital zircon geochronology occurred in the late 1990s through early 2000s, when laser-ablation ICP mass spectrometers (LA-ICPMS) were developed to the point that U-Pb analyses could be conducted in a robust fashion (Fryer et al., 1993;
Tectonics of Sedimentary Basins: Recent Advances, First Edition. Edited by Cathy Busby and Antonio Azor. Ó 2012 Blackwell Publishing Ltd. Published 2012 by Blackwell Publishing Ltd.
47
48
Part 2: New Techniques and Modeling
Machado and Gauthier, 1996; Horn et al., 2000; Kosler et al., 2001; Li et al., 2001; Machado and Simonetti, 2001; Horstwood et al., 2003). At present, detrital zircon geochronology is used primarily for four principle applications (Fedo et al., 2003; Anderson, 2005): .
. .
.
Provenance studies, where ages of detrital minerals are compared with ages of potential source terranes to determine ultimate sources of sediment Where provenance is known, determination of ages and characteristics of rocks in source terranes Correlation of sedimentary units, where ages of detrital minerals are compared in an effort to evaluate possible linkages between different sedimentary units Maximum depositional age, where the youngest age component in a clastic unit provides the earliest possible age of deposition
Although our research community is currently experiencing an explosion in the number of detrital zircon analyses that are conducted and published, it is disconcerting that we are not yet able to answer some basic questions about how to collect, display, and interpret U-Pb geochronologic data applied to detrital minerals. Some of these issues have been highlighted recently by Horstwood et al. (2009). Examples include the following: . . .
.
.
.
What is the optimal instrumentation used for a detrital zircon study? Which ages should be used, and how should these ages be evaluated and filtered? How many analyses should be conducted from each sample, and how should grains be selected for analysis? What is the most effective way to display detrital zircon data – Pb/U Concordia diagrams, TerraWasserburg diagrams, histograms, age-distribution diagrams, or cumulative probability plots? What is the best method of describing a set of detrital zircon ages (e.g., the youngest age component)? What is the best method for comparing age distributions of several samples?
This chapter is a progress report on attempts to address these questions, with an emphasis on current methods for gathering and interpreting U-Pb information and a brief look forward to future opportunities of gathering critical information
from detrital minerals. Readers are referred to Fedo et al. (2003) for an excellent summary of the history and applications of detrital zircon geochronology, and Gehrels (2000), Anderson (2005), and Nemchin and Cawood (2005) for a discussion of statistical analysis of detrital zircon data.
WHAT IS THE OPTIMAL INSTRUMENTATION USED F O R A D E T R I T A L Z I R CO N S T U D Y ? Three types of instruments are routinely used for determining U-Pb ages of detrital zircons. Following is a brief outline of the analytical methods used with each type of instrument, together with an evaluation of strengths and weaknesses for conducting U-Pb analyses of detrital zircons.
Isotope dilution-thermal ionization mass spectrometry (ID-TIMS) ID-TIMS analyses require dissolution of complete crystals or portions of crystals, addition of an isotopic tracer (commonly 205 Pb and 233 U), chemical separation of U and Pb, and isotopic analysis by TIMS (Bowring and Schmitz, 2003; Parrish and Noble, 2003; Mattinson, 2005). The chemical dissolution and separation portion of the analysis is quite time consuming and needs to be conducted in an ultra-clean environment to reduce contaminant Pb and U, but results in a very pure analyte that yields high-precision (0.1% at 2-sigma) isotope ratios and Pb/U concentrations. U-Pb ages determined by ID-TIMS are accordingly of the best possible precision and accuracy, and are essential for applications that require high temporal resolution. In most cases, however, this high precision is not necessary for a detrital zircon study.
Secondary ion mass spectrometry (SIMS or ion probe) SIMS analyses are generally conducted on the polished surface of a crystal that has been mounted in epoxy along with “standards,” which are crystals of the same mineral that are of known age and isotopic composition (Ireland and Williams, 2003). Because sputtering takes place at low temperature and high vacuum, backgrounds of Pb and U are
Detrital Zircon U-Pb Geochronology
low, and analyses can be conducted on very small volumes ( 1 ng) of material. This provides opportunities to determine U-Pb ages on small portions of crystals, with typical pits that are 10–30 microns diameter and 1 micron depth. Because it is not possible to add an isotopic tracer, ages are corrected for instrumental fractionation by standardsample bracketing (alternating between standards and unknowns, correcting the standards to the known age, and applying the same correction factor to the unknowns). This method yields ages with a precision and accuracy of 1–2% (2-sigma). Isotopic measurements are conducted by sequential analysis of U-Th-Pb (and other) isotopes, which requires a typical analysis time of 15 minutes. This technique is ideal for studies that require high spatial resolution (especially in the depth dimension), such as analysis of complex zircon crystals. Ion probes are also able to analyze other elements at the same time that U-Pb ages are determined, so it is now possible to also characterize detrital minerals for Ti and Zr concentrations, REE concentrations, and oxygen isotopes for constraints on petrogenesis (e.g., Mojzsis et al., 2001; Valley, 2003; Wooden et al., 2007). Laser-ablation inductively coupled plasma mass spectrometry (LA-ICPMS) LA-ICPMS methodology is similar to SIMS in that fractionation is determined by standard-sample bracketing, and that analyses are conducted on a polished crystal surface (e.g., Kosler and Sylvester, 2003). This method also yields ages with a precision and accuracy of 1–2% (2-sigma; Machado and Simonetti, 2001; Horstwood et al., 2003; Kosler and Sylvester, 2003; Chang et al., 2006; Gehrels et al., 2008; Horstwood, 2008). An advantage of LA-ICPMS is their much faster analysis time, facilitated with some instruments by having sufficient dispersion and enough collectors to be able to measure U and Pb simultaneously. A disadvantage, however, is that plasma ionization involves high flow rates of Ar gas, at atmospheric pressure, and at high temperature, all of which result in high background counts of Pb and Hg (which interferes with 204 Pb). Achieving high signal:background requires a fast rate of ablation and in most cases a larger volume of analyzed material (pits are typically 30 micron diameter by 10–20 micron depth). This faster rate of ablation makes analysis by LA-ICPMS very efficient, with typical analysis times of several minutes.
49
LA-ICPMS instruments are also ideally suited to analysis of other elements such as Hf-Lu-Yb (for Hf isotope determinations) and trace/rare earth elements (e.g., Machado and Simonetti, 2001; Woodhead et al., 2004; Gerdes and Zeh, 2006; Flowerdew et al., 2007; Mueller et al., 2007; Yuan et al., 2008, Kemp et al., 2009). WHICH AGES SHOULD BE USED, AND HOW SHOULD AGES BE EVALUATED AND FILTERED? The U-Pb system is particularly powerful for geochronology because (1) there are two decay systems (238 UŁ206 Pb and 235 UŁ207 Pb), (2) half lives for the two systems are appropriate for use through all but the most recent portion of Earth time, (3) the two decay systems are linked because 238 U=235 U is constant (137.88; Steiger and J€ager, 1977) in nearly all crustal rocks, and (4) there is a non-radiogenic isotope of Pb (204 Pb) that can be used to account for Pb present in the crystal at the time of formation. These aspects allow graphical representation of U-Pb ages on a Pb /U Concordia diagram (Wetherill, 1956; Fig. 2.1), which plots 206 Pb =238 U versus 207 Pb =235 U as a function of age ( indicates that initial Pb has been subtracted). The Concordia diagram also shows 206 Pb =207 Pb , which is the slope of a line from the origin through the analysis (206 Pb =207 Pb ¼ 206 Pb =238 U=½207 Pb =235 U 137:88) (Fig. 2.1). Operationally, 206 Pb =238 U is determined by comparison with a tracer solution containing known amounts of Pb and U (e.g., 205 Pb and 233 U) for ID-TIMS, or by comparison with standards for SIMS and LA-ICPMS. 206 Pb =207 Pb generally requires only a minor correction because there is little instrumental fractionation of 206 Pb relative to 207 Pb. 207 Pb =235 U is usually not measured directly – instead this ratio is calculated from measured 206 Pb =238 U, measured 206 Pb =207 Pb , and known 238 U=235 U. Because 235 U is much (137.88 times) smaller than 238 U, measurement of 235 U would add significant uncertainty to the 207 Pb =235 U age. An analysis is plotted on a Concordia diagram based on 206 Pb =207 Pb (expressed as 207 Pb =206 Pb in many labs) and 206 Pb =238 U. The three available ages are then shown as the intersection of 206 Pb =238 U, 207 Pb =235 U, and 206 Pb =207 Pb lines with Concordia (Fig. 2.1). Uncertainties for 206 Pb =238 U, 207 Pb =235 U, and 206 Pb =207 Pb generally form a zircon-shaped(!)
Part 2: New Techniques and Modeling 1.2
Pb*/235U
Pb*/207Pb*
25
207
206
Pb*/238U
90
0
206
50
0
4.5
0.6
0
4.5 Age (Ga)
Age (Ga)
4.5 Age (Ga)
3000
206
Pb*/238U => 2900 Ma 2800 2600
2000
1800
a
dia
r co
00
n
co
1600
> *=
1400
2 */
1200 1000
6
20
07
M
29
Pb
Pb
207
0.2
Pb*/235U => 2900 Ma
0.4
206
Pb*/ 238U
2400 2200
0.0 0
4
8
12
16
20
207
Pb*/235U
Fig. 2.1. Pb /U Concordia diagram, showing the three chronometers that are commonly used. The Concordia line shows coincidence of the three chronometers as a function of age (Wetherill, 1956). Note that 206 Pb =238 U and 206 Pb =207 Pb are the two independent measurements; 207 Pb =235 U is calculated from these ratios and the modern 238 U=235 U (137.88).
polygon, which is commonly expressed as a continuous probability density function (e.g., Ludwig, 2008). The relative values of the three uncertainties vary as a function of age: for “young” samples, 206 Pb =238 U age is the most precise and 206 Pb =207 Pb age is the least precise, whereas for “old” samples the values are reversed, with 206 Pb =207 Pb age more precise than 206 Pb =238 U age (Fig. 2.2). The uncertainty of 207 Pb =235 U (in age) is always intermediate in value. In theory and in practice, all three uncertainties have the same value at about 1.4 Ga (Gehrels, 2000; Nemchin and Cawood, 2005; Gehrels et al., 2008). Because of these variations, the age resolution of the U-Pb system is poorest at 1.4 Ga, with improving precision of 206 Pb =238 U age for younger ages and improving precision of 206 Pb =207 Pb age for older ages. If all three ages (206 Pb =238 U, 207 Pb =235 U, and 206 Pb =207 Pb ) are similar within error, an analysis
plots on the Concordia line and is described as being “concordant” (Wetherill, 1956). Unfortunately, it is common for an analysis to lie below Concordia, in which case it is referred to as being “discordant.” This results in ages that increase from 206 Pb =238 U to 207 Pb =235 U to 206 Pb =207 Pb . Discordance most commonly occurs due to loss of Pb during younger thermal/hydrothermal activity (in which case the analysis moves down the 206 Pb =207 Pb line that existed at the time of disturbance), or due to inheritance of older material (in which the analysis moves up along a mixing line). “Reverse discordant” analyses, which plot above Concordia, are rare, and most likely result from inaccurate measurement of 206 Pb =238 U. The degree of discordance is best expressed as a percentage of the 206 Pb =238 U age divided by the 206 Pb =207 Pb age, with a perfectly concordant analysis having 0% discordance (discordance ¼ 100 100 [206 Pb =238 U age=206 Pb =207 Pb age]. Conversely, this ratio can be expressed
Detrital Zircon U-Pb Geochronology
51
Magnitude of uncertainties: 206/238 > 207/235 > 206/207
0.6
3000 2800 2600
206Pb*/238U
2400 2200
0.4 2000 1800
206/238 uncertainty
1600 1400
0.2
1200 1000
Magnitude of uncertainties: 206/238 = 207/235 = 206/207
un 206 ce /20 rta 7 in ty
207/235 uncertainty
Magnitude of uncertainties: 206/238 < 207/235 < 206/207
0.0
0
4
8
12
16
207Pb*/235U
as degree of concordance, with a perfectly concordant analysis equal to 100% (concordance ¼ 100 ½206 Pb =238 U age=206 Pb =207 Pb age. Because zircons are more susceptible to Pb loss with increasing age (due to increasing lattice damage during radioactive decay), it is common for Precambrian zircons to be discordant by a few percent to a few tens of percent. For example, Figure 2.3 is an example of a zircon that crystallized at 1500 Ma and experienced Pb loss at 500 Ma. The 206 Pb =238 U, 207 Pb =235 U, and 206 Pb =207 Pb ages are all younger than the crystallization age, with the 206 Pb =207 Pb age the closest to the true age. Because of this, for reasonable (10–30%) degrees of discordance it is generally more accurate to use 206 Pb =207 Pb ages for zircons that are younger than 1.4 Ga, even though the 206 Pb =238 U age is more precise. A common cutoff is 0.8–1.0 Ga (Gehrels, 2000; Gehrels et al., 2008), which is a balance between the more accurate 206 Pb =207 Pb age (assuming that discordance is due to Pb loss) and the more precise 206 Pb =238 U age. For analyses that are discordant due to Pb loss, it is important to realize that even the 206 Pb =207 Pb age is younger than the true crystallization age (Fig. 2.3). The inaccuracy of the 206 Pb =207 Pb age becomes worse as discordance increases, and as the age of the Pb-loss event increases, even though precision of the analysis does not vary. The 206 Pb =238 U, 207 Pb =235 U, and 206 Pb =207 Pb ages are also inaccurate if discordance results from inheritance, with potential inaccuracy increasing with increasing discordance. It is accordingly
20
Fig. 2.2. Example showing the orientation and magnitude of uncertainty ellipses for Phanerozoic, Proterozoic, and Archean ages. The relative magnitudes (in millions of years) of the three primary uncertainties are also indicated. Inset on right is a graphical depiction of the three primary uncertainties.
common in a detrital zircon study to filter data based on degree of discordance. But what is the appropriate cutoff to use for discordance? Unfortunately, the appropriate level of discordance filter needs to be determined for each data set in light of the goals of the study and the complexities encountered. For example, if a study yields a mix of Phanerozoic and Archean ages, and the relative proportions of these ages are important, a generous (e.g., 30%) discordance cutoff might be appropriate so that most Precambrian ages are retained. If instead the main objective of a study is to test for a specific Late Archean age, it would be appropriate to use a tight discordance filter (e.g., 10%), or even use only concordant data, to achieve the best possible age resolution. And last, if a study yielded only young ages (e.g., <100 Ma), a discordance filter could not even be applied reliably because of the difficulty of determining reliable 206 Pb =207 Pb ages, and hence degree of discordance, for young ages. A discordance filter is also not assured of yielding robust ages because Pb loss soon after crystallization and/or inheritance of slightly older components will yield analyses that are analytically concordant but of inaccurate age. An alternative method is to place significance primarily on analyses that belong to a cluster, given that Pb loss and inheritance always scatter analyses away from their true age, A reasonable methodology is to filter for clustering, with a cluster defined, for example, as three or more analyses that overlap at 2-sigma uncertainty.
52
Part 2: New Techniques and Modeling 0.3 1600
crystallization age 206
207
0.2 206Pb*/ 238U
206
1400
Pb*/ 207Pb*age
ry
cto
Pb*/ 235U age
Pb*/ 238U age
je ra
t ss
1200
-lo
Pb
1000
800
age of Pb-loss
0.1
600
400 200
0.0 0
1
2
3
207Pb*/ 235U
Analyses that do not belong to a cluster may also be accurate, and are not discarded, but their true age significance remains uncertain. Figure 2.4 shows an example of a clustering filter applied to a data set with two groups of ages: an older group evaluated on the basis of overlapping 206 Pb =207 Pb ages, and a younger group evaluated on the basis of overlapping 206 Pb =238 U ages. Of the older analyses, three clusters are interpreted to record robust source ages, and less confidence is assigned to single analyses and pairs of analyses. For the younger analyses, two clusters with 3 constituent analyses are interpreted as robust and less confidence is assigned to single analyses and pairs of analyses. Unfortunately, filtering based on clustering becomes progressively more problematic with increasing discordance, and as the age of Pb loss becomes older (the Pb-loss trajectory diverges from the 206 Pb =207 Pb line; see Fig. 2.4). In the latter case, the 206 Pb =207 Pb ages would not define a cluster even if multiple grains are of the same crystallization age. Clustering analysis may also obscure true age variations within an apparent cluster. An alternative strategy, suggested by Nemchin and Cawood (2005) is to weight each analysis according to degree of discordance such that highly discordant analyses contribute little age probability to the final age-distribution curve. This would alleviate having to determine a specific level of discordance filter, but still requires determination of a weighting scheme that appropriately discounts for discordance but does not significantly bias the final age distribution.
4
Fig. 2.3. Diagram showing that, for a mid-Proterozoic analysis with Pb loss, the 206 Pb =207 Pb age is more accurate than 206 Pb =238 U and 207 Pb =235 U ages. This is why 206 Pb =207 Pb ages should be used for samples in the 800–1400 Ma age range (assuming that discordance is due to Pb loss), even though 206 Pb =238 U ages are generally more precise.
Because of these complexities, it is important to evaluate every data set individually to determine the robustness of each analysis. Clearly if the data shown in Figure 2.4 were acquired from a single sample, filtering based on discordance would bias the resulting age distribution toward young ages; the tighter the filter, the greater the bias. Likewise, a filter based on clustering may work well in a situation where Pb loss occurred in the recent past, but would fail to recognize coeval grains if the age of Pb loss is older. A reasonable initial strategy would be to use a moderate discordance filter (e.g., 10–30%) or an appropriate discordance weighting scheme (e.g., Nemchin and Cawood, 2005) to remove or deemphasize highly discordant grains but not significantly bias the data set, and then evaluate the accepted data in terms of clustering to highlight the most reliable age clusters. It is also becoming possible to increase the reliability of age assignments by conducting multiple analyses on each grain (e.g., Nemchin and Cawood, 2005; Simonetti et al., 2008; Johnston et al., 2009). For grains with Pb loss, it is possible to identify domains that have experienced little Pb loss (e.g., portions with low uranium concentration) and/or to use variable degrees of Pb loss to generate a Pb-loss trajectory with a robust upper intercept. This technique is also useful where there are multiple episodes of zircon growth, in which case a single grain would yield a crystallization history rather than a single age. As an example of this age-mapping
Detrital Zircon U-Pb Geochronology
53
Analysis belonging to a cluster Analysis of uncertain significance
0.6
3000
206Pb*/238Pb*trend
2800 10%
ance iscord 20% d ance rd o c is 30% d
2600
2200
0.4 2000
440
1800
206
Pb*/ 238U
2400
400
1600
360
1400
0.2
e
rdanc
disco
%
30
ce
an ord
c dis
320
1200 1000
280 240 200
0.0 0
4
8
12 207
16
20
235
Pb*/
U
Fig. 2.4. Concordia diagram showing strategy of using clustering and discordance filters to evaluate detrital zircon ages. For Precambrian ages, clusters are defined by analyses that yield similar 206 Pb =207 Pb ages (shown graphically with gray lines). Single analyses or pairs of analyses may yield useful information, but may be less reliable than analyses belonging to larger (e.g., n 3) clusters. Application of an appropriate discordance filter (e.g., 10–30%) may further improve reliability of the final ages. For younger data (lower right inset), clusters are defined on the basis of 206 Pb =238 U ages, with individual analyses or pairs of analyses interpreted to be less robust than analyses belonging to larger (e.g., n 3) clusters. Application of a discordance filter for young ages is generally not possible because of the large uncertainty of 206 Pb =207 Pb ages for young grains. It is important to realize that in some cases application of a clustering filter may be ineffective in identifying robust ages. For example, if analyses lie along a discordance trajectory with ancient upper and lower intercepts (e.g., analysis shown in Fig. 2.3), both 206 Pb =238 U and 206 Pb =207 Pb ages will be inaccurate and no clusters will be apparent. In such cases, application of a tight discordance filter (e.g., 10% or less) may be the only way to focus significance on the more robust analyses. As emphasized in the text, both discordance and clustering filters need to be applied with caution, and with full consideration of the complexities of the data and the objectives of the study.
approach, Figure 2.5 shows a crystal that has been analyzed in 140 different spot locations (adapted from Gehrels et al., 2009). The resulting age patterns, when superimposed on a cathodoluminescence (CL) image, yield a rich history that would be very useful in reconstructing provenance.
HOW MANY ANALYSES SHOULD B E C O N D U C T E D F R O M EA CH S A M P L E , AND HOW SHOULD GRAINS B E S E L E CT E D F O R A N A L Y S I S ? Two basic questions that commonly arise in detrital zircon studies concern the number of analyses
57 Ma 97 Ma 97 Ma 130 Ma 160 Ma Laser Pit (10 micron diameter)
100 microns
Fig. 2.5. Example of the power of using multiple small-volume analyses on a single grain to resolve a complex history of zircon growth. Data are from Gehrels et al. (2009); a cathodoluminescence (CL) image was acquired with assistance of Axel Schmitt (UCLA ion probe facility).
54
Part 2: New Techniques and Modeling
to be conducted, and how grains should be selected for analysis (e.g., Gehrels, 2000; Fedo et al., 2003; Anderson, 2005). The answers to these questions depend very much on the questions that are being addressed. If detrital zircon analyses are conducted in an effort to constrain maximum depositional age, a reasonable strategy would be to focus on crystals that have the lightest color and minimal rounding. Because pinkish color tends to deepen with increasing age (e.g., Silver and Deutsch, 1963), preferential analysis of crystals with the lightest color tends to increase the yield of the youngest ages. Degree of rounding would also tend to increase with age given the increased opportunities for recycling. The number of grains analyzed would depend on the proportion of the youngest grains in the sample, and on the degree of confidence needed for the maximum depositional age. If detrital zircons are analyzed in an effort to test for the presence of an expected component (or components), grains should be selected for analysis using the general guides noted above, that zircons tend to increase in color and in degree of rounding with age. Hence the probability of identifying a particular component can be maximized by selecting grains with the expected characteristics. The number of analyses again depends on the proportion of the expected component and on the degree of confidence needed for identification of the expected component. If detrital zircon analyses are conducted to determine general aspects of provenance, or to test for stratigraphic correlation, the objective would of course be to produce an age distribution that reflects the true ages (and proportions of ages) of zircon crystals in the original sample. Care must therefore be taken to avoid biasing the sample during processing, for example by magnetic separation, sorting by size (either intentionally or during handling), or hand picking (Gehrels, 2000; Sircombe and Stern, 2002; Fedo et al., 2003; Anderson, 2005; Nemchin and Cawood, 2005). During analysis, grains must be selected at random from the crystals available, irrespective of grain size, color, shape, degree of rounding, and so on. Unfortunately, exceptions to this approach must be made for analytical considerations: very small grains commonly cannot be analyzed, grains with fractures commonly yield unreliable ages (due to secondary minerals, Pb loss along fracture surfaces, or anomalous behavior of the incident laser or primary ion beam), inclusions
should be avoided, and grains with complex age zoning need to be avoided unless each component can be analyzed separately. Because of the latter, it is best to use CL images (e.g., Corfu et al., 2003; Nasdala et al., 2003) when locating analysis pits in detrital zircon crystals. The number of analyses needed for a robust provenance study depends on the number and proportions of different age groups present, whether crystals have been affected by Pb loss and/or inheritance, the precision of the analytical method used, and the confidence level needed to reach a particular conclusion or test a particular hypothesis. As discussed by Anderson (2005), it is possible to use the standard binomial probability formula (Dodson et al., 1988) to show that 60 analyses must be conducted to have a 95% probability of identifying one component that makes up 5% of the population of grains. Likewise, 117 analyses need to be conducted to have a 95% chance of identifying every 5% component in a population of grains (Vermeesch, 2004). In practice, however, these threshold values are underestimates because they assume that every age determination yields a robust age. As described above, this is commonly not the case due to complexities such as Pb loss and/or inheritance. If robustness is measured by clustering, a much greater number of analyses must be conducted to document that all minor age populations have been recognized. A reasonable initial approach for most provenance studies is to conduct 100 analyses, use a discordance filter that does not significantly reduce the number of “older” analyses (e.g., 10– 30%), and place greatest significance on analyses that belong to clusters (e.g., Gehrels et al., 2008). But such an approach will clearly not identify the minor components of a set of detrital zircons with certainty.
REPRESENTATION OF U-PB DATA ON CONCORDIA DIAGRAMS, RELATIVE AGE PROBABILITY D IA G R A M S , A N D C U M U L A T I V E P R O B A BI L I T Y P L O T S The most common format for presenting and evaluating detrital zircon data is with a Pb /U Concordia diagram, as described above and shown on Figure 2.6A. For a sample with a wide range of ages, however, the standard Pb /U format tends to obscure the younger grains. An alternative for
55
Detrital Zircon U-Pb Geochronology 1.0 206Pb / 238U
(A) 0.8 0.6
3400 3000 2600 2200
0.4
1800
0.2 207
0.0
0
20
10
Pb/ 235U
30
40
1.00 (B)
206Pb / 238U
3000 2400 1800 1400 1000 600
0.10
400 200 207
Pb/ 235U
1.00
0.10
10.00
0.4
100.00 (C)
3600
207
0.3
Pb / 206Pb
0.01 0.01
3000 2600
0.2
2000
(or relative age probability plot) because information about the uncertainty of each analysis is also included (Sircombe, 2004; Sircombe and Hazelton, 2004; Ludwig, 2008). An age-distribution curve is constructed by (1) assigning a normal (Gaussian) distribution to each analysis based on the reported age and uncertainty, (2) summing the probability distribution of every acceptable analysis into a single curve, and (3) if normalized, dividing the area under the curve by the number of analyses.Figure 2.7 shows a series of age-distribution curves that have been normalized for the number of constituent analyses (area under each curve is the same) in such a way that a sample can be compared against several reference curves. Age-distribution diagrams can be generated with routines from Ludwig (2008) and Sircombe (2004), while stacked and normalized probability plots can be made with a routine available from www.geo.arizona.edu/alc. For some applications, it is advantageous to show detrital age spectra on a cumulative age probability plot. Figure 2.8 shows a series of cumulative age probability plots for the same data shown in Figure 2.7. These two types of plots show exactly the same information, with age peaks on a relative age probability curve corresponding to steep segments on a cumulative probability curve. Graphically, cumulative probability plots can also be made with a routine available from www.geo.arizona.edu/alc. These plots are very useful if a large number of curves are to be
0.1 600
200 238
U/ 206Pb
0
20
40
60
80
100
Fig. 2.6. Example of a set of detrital zircon ages shown on three different Concordia diagrams: (A) a conventional Pb /U Concordia diagram (Wetherill, 1956), (B) a conventional Pb /U diagram plotted with log-log scale to emphasize young ages, and (C) a Tera-Wasserburg diagram (Tera and Wasserburg, 1972). All diagrams made with Isoplot (Ludwig, 2008).
such situations is to use a log scale for both axes (Fig. 2.6B). It is also possible to use a TeraWasserburg diagram (Tera and Wasserburg, 1972), which plots 207 Pb =206 Pb against 238 U=206 Pb (Fig. 2.6C). All of these plots are easily created with Isoplot (Ludwig, 2008). A detrital zircon age distribution can be plotted on a simple histogram, but it is more informative to portray the data on an age-distribution diagram
Sample
Relative age probability
0.0
Reference 5
Reference 4 Reference 3 Reference 2
Reference 1
0
500
1000 1500 2000 2500 3000 Detrital zircon age (Ma)
3500
4000
Fig. 2.7. Example of a normalized age-distribution (relative age probability) diagram, showing age distributions from a sample and several reference data sets. Diagram made from routine available at www.geo.arizona.edu/alc.
56
Part 2: New Techniques and Modeling
Cumulative age probability
1 0.9 0.8 0.7 Reference 1 Reference 2 Reference 3 Reference 4 Reference 5 Sample
0.6 0.5 0.4 0.3 0.2 0.1 0 0
500 1000 1500 2000 2500 3000 3500 4000 Detrital zircon age (Ma)
Fig. 2.8. Example of a cumulative age probability plot, showing the same data as in Figure 2.7. Diagram made from routine available at www.geo.arizona.edu/alc.
compared, or if the samples of interest yield differing proportions of similar ages. WHAT IS THE BEST METHOD O F D E S C R I B I N G A SE T O F D E T R I T A L ZIRCON AGES (E.G., THE YOUNGEST AGE COMPONENT)? Because it is not feasible (or particularly useful) to discuss each individual age from a detrital zircon sample, it is generally necessary to describe ages of groups or clusters of analyses. This can be done easily by reporting the numbers of analyses within histogram bins, but this introduces artificial age boundaries and does not incorporate useful information about age uncertainties (e.g., Sircombe, 2004). A more informative approach is to use the grouping of ages shown on an age-distribution diagram (e.g., Fig. 2.7). A first attempt to do this systematically has been developed in a routine called “AgePick,” which is available from www. geo.arizona.edu/alc. This routine emphasizes the main groups of ages by identifying each age range that contains age probability from three or more overlapping analyses, and then reports the number of constituent analyses in each range (e.g., eight ages between 1682 and 1735 Ma). The routine also reports the peaks in age probability within each range and the number of analyses that contribute age probability to each peak (e.g., four ages contribute to a peak age of 1712 Ma). One of the shortcomings of reporting minimum and maximum ages of a range, or the age of a peak in age probability, is that there is no uncertainty associated with each age, even though there clearly
is uncertainty associated with each of the constituent analyses. It is accordingly common to calculate a weighted mean age and uncertainty of a group of ages, for example using Isoplot (Ludwig, 2008). This approach is built on the assumptions, however, that all included analyses are truly cogenetic and that the scatter of ages is due only to analytical uncertainty. Because there is no reason to assume that all detrital zircons of similar age are truly cogenetic, and because some ages may be inaccurate due to Pb loss and/or inheritance, this approach needs to be used with caution. This cautionary note applies to any averaging process, from calculating a simple average to more sophisticated analyses such as the “TuffZirc” and “Unmixing” routines of Isoplot (Ludwig, 2008). Fortunately, complementary information can be used to evaluate complexities of ages, and to determine whether a set of zircon grains may be cogenetic. For example, U concentration and U/Th can be used to identify unusual (e.g., non-cogenetic) crystals and to determine whether unusual ages may be compromised by Pb loss or metamorphic recrystallization/overgrowth. A useful method of evaluating this information is with plots of age versus U concentration and U/Th; correlations between age and U concentration would suggest that younger ages may have been affected by Pb loss, whereas correlations with U/Th would suggest that some analyses are compromised by metamorphic zircon growth/recrystallization. Plots of age versus analysis position (e.g., core versus rim) might reveal the presence of inherited cores or grain exteriors with Pb loss. Such plots can be generated with “AgePick” (available from www. geo.arizona.edu/alc). As noted below, this type of multidimensional analysis will be facilitated by future opportunities to gather additional isotopic, compositional, and structural information about dated crystals in efforts to determine whether groups really are cogenetic and to understand geochronologic complexities.
WHAT IS THE BEST METHOD FOR COMPARING AGE DISTRIBUTIONS O F S E V E R A L S A M PL E S ? Once a detrital zircon data set has been acquired, it is common to attempt comparisons with samples from the same study area and perhaps from other regions. Traditionally, comparisons have
been made visually, with either normalized age distribution (Fig. 2.7) or cumulative probability (Fig. 2.8) plots. For example, a visual comparison of the age-distribution curves shown in Figures 2.7 and 2.8 suggests that the sample analyzed is similar to Reference 2 but unlike the other reference age distributions. Such a conclusion is based on the fact that the sample and Reference 2 contain similar ages, in approximately the same proportions. But which is more important, overlapping ages versus similar proportions, and how can these criteria be quantified? An early attempt to quantify comparisons was made by Gehrels (2000) using numerical analyses of the degree to which two age-distribution curves contain overlapping ages. Values in this overlap analysis range from 0 if no ages occur in both samples to 1 if all ages occur in both samples. This analysis provides a measure of presence versus absence of specific ages. In an effort to factor in proportions of overlapping ages, a second criterion was developed that measures the degree to which overlapping ages exist in similar proportions (0 ¼ different proportions or non-overlapping ages, 1 ¼ overlapping ages in exactly the same proportions). Both of these criteria are descriptive, and they contain no information about the degrees of overlap and similarity that would be expected from two distributions based on the number of analyses and their analytical uncertainties. Overlap and similarity analysis routines are available from www.geo.arizona.edu/alc. A statistical comparison of multiple age-distribution curves is provided by the KolmogorovSmirnov (K-S) statistic, which measures the probability that two age-distribution curves are not drawn from the same original population. The fundamental criterion is the P value, which is the probability that two samples are not statistically different. For example, a P-value >0.05 yields >95% confidence that the two samples are not statistically different. This analysis has been used by Berry et al. (2001), DeGraaff-Surpless et al. (2003), and Dickinson and Gehrels (2009) as an effective means of comparing detrital zircon data from several sets of samples. It is important to note, however, that the K-S statistic is very sensitive to the proportions of ages present. As an illustration, Figure 2.9 shows age-distribution curves for two samples that contain exactly the same ages, but with a 20% difference in proportions of the two ages (40 ages of 100 Ma and 60 ages of 200 Ma versus 60 ages of 100 Ma and 40 ages of
Normalized relative age probability
Detrital Zircon U-Pb Geochronology
57
Sample 2
Sample 1
0
20 40 60 80 100 120 140 160 180 200 220 240 260 280 300
Detrital zircon age (Ma)
Fig. 2.9. Example of K-S analysis of a hypothetical data set in which 100 analyses are conducted on two different samples. Sample 1 yields 40 analyses that are 100 10 Ma and 60 analyses that are 200 10 Ma. Sample 2 yields 60 analyses that are 100 10 Ma and 40 analyses that are 200 10 Ma. K-S analysis indicates that these two samples were not drawn from the same population (at 95% confidence level), which, if used in a provenance analysis or to test for stratigraphic correlation, would indicate that these samples are unrelated. This example emphasizes the degree to which K-S analysis is sensitive to proportions of ages.
200 Ma). Because the P-value for this data set is less than 0.05%, K-S analysis indicates that these two samples were not drawn from the same population (at the 95% confidence level). If this comparison were conducted for a provenance analysis, a logical conclusion would be that the two samples were not shed from the same source region. Although statistically accurate, most researchers would question this conclusion based on the fact that the two samples contain exactly the same two age groups, only in slightly different proportions. Routines for conducting K-S analyses are available from www.geo.arizona.edu/alc. Comparison of detrital zircon data sets can also be accomplished through kernel function estimation (Sircombe and Hazelton, 2004). This technique generates a numerical value that describes the difference between two age-distribution curves, but this value is again highly sensitive to proportions of ages as well as age differences. Clearly there is a critical need for better tools with which to compare detrital zircon data! F U T U R E O PP O R TU N I T I ES Given that detrital zircon geochronologic analyses have been generated routinely for only the past
58
Part 2: New Techniques and Modeling
15 years, it is not surprising that analytical methodologies are evolving rapidly and that challenges remain. Prospects for future development can be divided into three main areas, as follows. Enhanced precision, accuracy, efficiency, and spatial resolution Many of the challenges outlined above could be addressed more effectively if it were possible to generate U-Pb ages more efficiently, with finer spatial resolution, and with improved precision and accuracy. For SIMS and LA-ICPMS, improving precision and accuracy to better than 1–2% would provide a more robust assessment of discordance and clustering, which would improve data quality in most cases. Such advances may be enabled by chemical abrasion of zircons (Mattinson, 2005) prior to analysis by in situ analysis (Horstwood et al., 2008). Improving spatial resolution and efficiency of these techniques, and perhaps automating data acquisition (e.g., Frei and Gerdes, 2009; Holden et al., 2009), would make it feasible to routinely conduct multiple analyses on each grain, which would provide a powerful tool for identifying grains that have experienced Pb loss or have multiple age domains. Although the presence of such complexities currently compromises a data set, full characterization of these complexities could provide a very powerful tool for determining provenance and for studying source terrane history. Examples of the power of enhanced spatial resolution for studying complex crystals and small zircons (e.g., in shale and siltstone) by LA-ICPMS have recently been reported by Simonetti et al. (2008), Cottle et al. 2009, 2008, Johnston et al. (2009), and Gehrels et al. (2009). Database tools for analysis and archival of detrital zircon data Given the large number of detrital zircon analyses that are conducted and reported each year, there is an urgent need for a database system in which data from around the world are archived and available for integration with other data sets. There are many database systems that operate within individual countries or even states/provinces (e.g., the National Geochronological Database in the United States [Sloan et al., 2003], the Canadian Geochronology Knowledgebase [http://gdr.nrcan.gc.ca/ geochron/index_e.php], the Brazilian National Geochronological Database [Silva et al., 2003]
and OZCHRON for Australia [www.ga.gov.au/oracle/ozchron/]), but given the global scope of provenance studies, an international database system would be a powerful research tool. Ideal would be a system in which geochronologic data could be analyzed using well-established database interrogation methods, and then integrated with all available geologic, geochemical, stratigraphic, and paleontologic data. Efforts to construct such a database are currently in progress as part of the StratDB initiative (Eglington, 2004) and the EarthChem initiative (www.earthchem.org/earthchemWeb/index.jsp). Integration with other isotopic, compositional, thermochronologic, and structural information One of the most exciting opportunities for future development involves the integration of geochronologic data with geochemical, isotopic, thermochronologic, and structural information gathered from a dated mineral. Geochemical data of interest currently includes Ti concentrations to determine crystallization temperature (e.g., Hanchar and Watson, 2003; Watson et al., 2006; Fu et al., 2008) and REE concentrations (e.g., Hoskin and Ireland, 2000; Hoskin and Schaltegger, 2003; Wooden et al., 2007). There are many isotopic systems that can be applied to zircon that reveal important constraints on petrogenesis, with Lu-Hf, Li, oxygen, and Sm-Nd being the most common (Kinny and Maas, 2003; Valley, 2003; Hawkesworth and Kemp, 2006; Scherer et al., 2007; Ushikubo et al., 2008). Thermochronologic information (e.g., (U-Th)/ He and fission-track ages) provides a very powerful companion to U-Pb age, with important implications for provenance and evolution of source areas (Rahl et al., 2003). Given that U-Pb, He, and fission-track ages can also be measured from other minerals (e.g., apatite; Carrapa et al., 2009), analysis of various phases in a sandstone could provide more robust provenance information and a rich time-temperature history of source regions. An additional area of great promise is the abundance of compositional and structural information that can be gained from spectroscopic methods (Corfu et al., 2003; Nasdala et al., 2003). Summary of future opportunities Given the exciting developments noted above, it is possible to imagine a future for detrital zircon geochronology in which the following occur:
Detrital Zircon U-Pb Geochronology .
. .
.
.
Multiple analyses are conducted on each grain in an effort to determine a robust crystallization age (or ages), and also to constrain patterns of complexity that may provide useful information about provenance and source terranes. Acquisition of U-Pb data is automated (e.g., Frei and Gerdes, 2009; Holden et al., 2009). Additional information (e.g., Hf, Li, and O isotopes, REE patterns, Ti concentrations, He and/ or fission-track cooling ages, etc.) is acquired from the dated grains to help resolve provenance and to constrain history of source terranes or host strata. Graphical and quantitative tools are available for interpreting and integrating this multidimensional information. A worldwide database exists that would allow a new data set to be integrated with existing detrital zircon data as well as complementary stratigraphic, faunal, geochemical, and other information.
CONCLUSIONS The development of new geochronologic methods for determining provenance, correlation, and maximum depositional age provides geologists with powerful new tools for studying sedimentary rocks, their sources, and the basins in which they accumulated. One of the most exciting developments is in the use of detrital zircons given that they are common in many crustal rocks; tend to survive multiple cycles of weathering, transport, and diagenesis; and generally provide robust ages of crystallization and cooling (e.g., Fedo et al., 2003). In spite of these strengths and the virtual explosion of detrital zircons studies, fundamental aspects of data acquisition, interpretation, presentation, and integration are as yet uncertain (e.g., Horstwood et al., 2008). Because of this, it is critical for researchers interested in using detrital zircon data to be fully involved in all aspects of processing samples, acquiring U-Pb data, interpreting the data (especially if complex), and evaluating the significance of the ages. Only when armed with a full understanding of each step in the analytical process will researchers have a full appreciation of the strengths and limitations of their data. Meanwhile, researchers who operate labs used to generate U-Pb data need to continue working on the development of more robust tools with which to evaluate, present, compare, and
59
integrate data, and new methods for gathering and integrating complimentary geochemical and structural information. Through these collective efforts, detrital mineral geochronology will evolve from a new technique with great promise to a fundamental tool in deciphering the rich history preserved in sedimentary rocks. ACKNOWLEDGMENTS I thank the editors of this volume for the invitation to prepare this chapter, and also wish to acknowledge extensive discussions with Bill Dickinson and Bill McClelland about detrital zircon methods and applications. Geochronologic research at the Arizona LaserChron Center is conducted in collaboration with Joaquin Ruiz; is enabled by capable assistance from Alex Pullen, Mark Pecha, Victor Valencia, Mark Baker, Scott Johnston, Gayland Simpson, and Roswell Juan; and is supported by the Instrumentation and Facilities Program of the National Science Foundation (EAR-0443387 and EAR-0732436). Reviewed by Matt Horstwood and Chris Fedo. R E F ER EN C E S Anderson, T. (2005) Detrital zircons as tracers of sedimentary provenance: limiting conditions from statistics and numerical simulation. Chemical Geology, 216 (3–4), 249–270. Berry, R.F., Jenner, G.A., Meffre, S., and Tubrett, M.N. (2001) A North American provenance for Neoproterozoic to Cambrian sandstones in Tasmania? Earth and Planetary Science Letters, 192, 207–222. Bowring, S.A., and Schmitz, M.D. (2003) High-precision U-Pb zircon geochronology and the stratigraphic record, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 305–326. Carrapa, B., DeCelles, P., Reiners, P., Gehrels, G., and Sudo, M. (2009) Apatite triple dating and white mica 40Ar/39Ar thermochronology of syntectonic detritus in the central Andes: A multiphase tectonothermal history. Geology, 37 (5), 407–410. Chang, Z., Vervoort, J.D., McClelland, W.C., and Knaack, C. (2006) U-Pb dating of zircon by LA-ICP-MS. Geochemistry, Geophysics, Geosystems, 7, Q05009. doi: 10.1029/ 2005GC001100. Corfu, F., Hanchar, J.M., Hoskin, P.W.O., and Kinny, P. (2003) Atlas of zircon textures, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 469–500. Cottle, J.M., Horstwood, M.S.A., and Parrish, R.R. (2009) A new approach to single shot laser ablation analysis and its application to in-situ geochronology. Journal of Analytical Atomic Spectrometry. doi: 10. 1039/b821899d.
60
Part 2: New Techniques and Modeling
Cottle, J., Parrish, R., and Horstwood, M. (2008) A new method for analyzing thin (>2 micron) zircon rims by LA-ICP-MS. Geochimica et Cosmochimia Acta, 72 (12S) A183. Davis, D.W., Poulson, K.H., and Kamo, S.L. (1989) New insights into Archean crustal development from geochronology in the Rainy Lake area, Superior Province, Canada. Journal of Geology, 97, 379–398. Davis, D.W., Williams, I.S., and Krogh, T.E. (2003) Historical development of U-Pb geochronology, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 145–181. DeGraaff-Surpless, K., Mahoney, J.B., Wooden, J.L., and McWilliams, M.O. (2003) Lithofacies control in detrital zircon provenance studies: insights from the Cretaceous Methow basin, southern Canadian Cordillera. GSA Bulletin, 115, 899–915. Dickinson, W.R., and Gehrels, G.E. (2009) U-Pb ages of detrital zircons in Jurassic eolian and associated sandstones of the Colorado Plateau: evidence for transcontinental dispersal and intraregional recycling of sediment. Geological Society of America Bulletin, 121, 408–433. Dodson, M.H., Compston, W., Ireland, I.S., and Wilson. J.F. (1988) A search for ancient detrital zircons in Zimbabwean sediments. Journal of the Geological Society of London, 145, 977–983. Eglington, B.M. (2004) DateView: a windows geochronology database. Computers and Geosciences, 30 (8), 847–858. Erdmer, P., and Baadsgaard, H. (1987) 2.2 Ga age of zircons in three occurrences of Upper Proterozoic clastic rocks of the northern Cassiar terrane, Yukon and British Columbia. Canadian Journal of Earth Sciences, 24, 1919–1924. Fedo, C.M., Sircombe, K., and Rainbird, R. (2003) Detrital zircon analysis of the sedimentary record, in Hanchar, J. M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 277–303. Flowerdew, M., Millar, I., Curtis, M., Vaughan, A., Horstwood, M., Whitehouse, M., and Fanning, C.M. (2007) Combined U-Pb geochronology and Hf isotope geochemistry of detrital zircons from early Palaeozoic sedimentary rocks, Ellsworth-Whitmore Mountains block, Antarctica. Geological Society of America Bulletin, 119 (3–4), 275–288. Frei, D., and Gerdes, A. (2009) Precise and accurate in situ U–Pb dating of zircon with high sample throughput by automated LA-SF-ICP-MS. Chemical Geology, 261 (3–4), 261–270. Froude, D.O., Ireland, T.R., Kinny, P.D., Williams, I.S., Compston, W., Williams, I.R., and Meyers, J.S. (1983) Ion microprobe identification of 4,100–4,200 Myr-old terrestrial zircons. Nature, 304, 616–618. Fryer, B.J., Jackson, S.E., and Longerich, H.P. (1993) The application of laser ablation microprobe-inductively coupled plasma mass spectrometry (LAM-ICPMS) to in situ (U)-Pb geochronology. Chemical Geology, 109, 1–8. Fu, B., Page, F.Z., Cavosie, A.J., Fournelle, J., Kita, N.T., Lackey, J.S., Wilde, S.A., and Valley, J.W. (2008) Ti-in-zircon thermometry: applications and limitations. Contributions to Mineralogy and Petrology, 156 (2), 197–215.
Gehrels, G.E. (2000) Introduction to detrital zircon studies of Paleozoic and Triassic strata in western Nevada and northern California, in Soreghan, M.J. and Gehrels, G.E., eds., Paleozoic and Triassic paleogeography and tectonics of western Nevada and northern California. Geological Society of America Special Paper 347. Boulder, CO, Geological Society of America, 1–18. Gehrels, G.E., McClelland, W.C., Samson, S.D., Patchett, P.J., and Jackson, J.L. (1990) Ancient continental margin assemblage in the northern Coast Mountains, southeast Alaska and northwest Canada. Geology, 18, 208–211. Gehrels, G.E., Valencia, V., and Ruiz, J. (2008) Enhanced precision, accuracy, efficiency, and spatial resolution of U-Pb ages by laser ablation–multicollector–inductively coupled plasma–mass spectrometry. Geochemistry, Geophysics, Geosystems, 9, Q03017. doi: 10.1029/2007 GC001805. Gehrels, G., Rusmore, M., Woodsworth, G., Crawford, M., Andronicos, C., Hollister, L., Patchett, J., Ducea, M., Butler, R. Klepeis, K. Davidson, C., Mahoney, B., Friedman, R., Haggart, J., Crawford, W., Pearson, D., and Girardi, J. (2009) U-Th-Pb geochronology of the Coast Mountains Batholith in north-coastal British Columbia: constraints on age, petrogenesis, and tectonic evolution. Geological Society of America Bulletin, 121 (9–10), 1341–1361. doi: 10.1130/ B26404.1. Gerdes, A., and Zeh, A. (2006) Combined U-Pb and Hf isotope LA-(MC)-ICP-MS analyses of detrital zircons: comparison with SHRIMP and new constraints for the provenance and age of Armorican metasediment in central Germany. Earth and Planetary Science Letters, 249, 47–61. Girty, G.H., and Wardlaw, M.S. (1984) Was the Alexander terrane a source of feldspathic sandstones in the Shoo Fly Complex, northern Sierra Nevada, California. Geological Society of America Bulletin, 96, 516–521. Hanchar, J.M., and Watson, E.B. (2003) Zircon saturation thermometry, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon. Reviews in Mineralogy and Geochemistry, 53, 89–112. Hart, S.R., and Davis, G.L. (1969) Zircon U-Pb and wholerock Rb-Sr ages and early crustal development near Rainy Lake, Ontario. Geological Society of America Bulletin, 80, 595–616. Hawkesworth, C.J., and Kemp, A.I.S. (2006) Using Hf and oxygen isotopes in zircons to unravel the record of crustal evolution. Chemical Geology, 226, 144–162. Holden, P., Lanc, P., Ireland, T.R., Harrison, T.M., Foster, J.J., and Bruce, Z. (2009) Mass-spectrometric mining of Hadean zircons by automated SHRIMP multi-collector and single-collector U/Pb zircon age dating: the first 100, 000 grains. International Journal of Mass Spectrometry, 286 (2–3), 53–63. Horn, I., Rudnick, R.L., and McDonough, W.F. (2000) Precise elemental and isotope ratio determinations by simultaneous solution nebulization and laser ablationICP-MS: application to U-Pb geochronology. Chemical Geology, 164, 281–301. Horstwood, M.S.A. (2008) Data reduction strategies, uncertainty assessment, and resolution of LA-(MC)-ICP-MS isotope data, in Sylvester, P., ed., Laser ablation ICP-MS in the Earth sciences: current practices and outstanding
Detrital Zircon U-Pb Geochronology issues. Short Course Series, 40. Quebec, QC, Mineralogical Association of Canada, 283–303. Horstwood, M., Cottle, J., and Parrish, R. (2008) Improving the utility of detrital zircon studies through chemical abrasion. Geochimica et Cosmochimia Acta, 72 (12S) A392. Horstwood, M.S.A., Foster, G.L., Parrish, R.R., Noble, S.R., and Nowell, G.M. (2003) Common-Pb corrected in situ U-Pb accessory mineral geochronology by LA-MC-ICPMS. Journal of Analytical Atomic Spectrometry, 18, 837–846. Horstwood, M.S.A., Kosler, J., Jackson, S., Pearson, N., and Sylvester, P. (2009) Investigating age resolution in laser ablation geochronology. EOS, Transactions, 90 (6), 47. Hoskin, P.W.O., and Ireland, T.R. (2000) Rare earth element chemistry of zircon and its use as a provenance indicator. Geology, 28 (7), 627–630. Hoskin, P.W.O., and Schaltegger, U. (2003) The composition of zircon and igneous and metamorphic petrogenesis, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 27–62. Ireland, T.R. (1992) Crustal evolution of New Zealand: evidence from age distributions of detrital zircons in Western Province paragneisses and Torlesse greywacke. Geochimica et Cosmochimica Acta, 56, 911–920. Ireland, T.R., and Williams, I.S. (2003) Considerations in zircon geochronology by SIMS, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 215–241. Johnston, S., Gehrels, G., Valencia, V., and Ruiz, J. (2009) Small-volume U-Pb geochronology by Laser AblationMulticollector-ICP Mass Spectrometry. Chemical Geology, 259, 218–229. Kemp, A.I.S., Foster, G.L., Schersten, A., Whitehouse, M.J., Darling, J., and Storey, C. (2009) Concurrent Pb-Hf isotope analysis of zircon by laser ablation multi-collector ICP-MS, with implications for the crustal evolution of Greenland and the Himalayas. Chemical Geology, 261, 244–260. Kinny, P.D., and Maas, R. (2003) Lu-Hf and Sm-Nd isotope systems in zircon, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 327–341. Kosler, J., and Sylvester, P.J. (2003) Present trends and the future of zircon in U-Pb geochronology: laser ablation ICPMS, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 243–275. Kosler, J., Tubrett, M., and Sylvester, P. (2001) Application of laser ablation ICPMS to U-Th-Pb dating of monazite. Geostandards Newsletter, 25, 375–386. Krogh, T.E., and Davis, G.L. (1975) The production and preparation of 205 Pb for use as a tracer for isotope dilution analyses. Carnegie Institute of Washington Yearbook, 74, 416–417. LeDent, D., Patterson, C., and Tilton, G.R. (1964) Ages of zircon and feldspar concentrates from North American beach and river sands. Journal of Geology, 72, 112–122. Li, X., Liang, X., Sun, M., Guan, H., and Malpas, J.G. (2001) Precise 206 Pb=238 U age determination on zircons by laser ablation microprobe-inductively coupled plasma-mass spectrometry using continuous laser ablation. Chemical Geology, 175, 209–219.
61
Ludwig, K.R. (2008) Isoplot 3.6. Special Publication No. 4. Berkeley, CA, Berkeley Geochronology Center, 77 p. Machado, N., and Gauthier, G. (1996) Determination of 207 Pb=206 Pb ages on zircon and monazite by laser ablation ICPMS and application to a study of sedimentary provenance and metamorphism in southeastern Brazil. Geochimica et Cosmochimica Acta, 60, 5063–5073. Machado, N., and Simonetti, A. (2001) U-Pb dating and Hf isotopic composition of zircon by laser-ablationMC-ICP-MS, in Sylvester, P., ed., Laser Ablation ICPMS in the Earth Sciences: Mineralogical Association of Canada, Short Course Series, 29, 121–146. Mattinson, J.M. (2005) Zircon U-Pb chemical abrasion (“CATIMS”) method; combined annealing and multi-step partial dissolution analysis for improved precision and accuracy of zircon ages. Chemical Geology, 220, 47–66. Mojzsis, S.J., Harrison, T.M., and Pidgeon, R.T. (2001) Oxygen-isotope evidence from ancient zircons for liquid water at the Earth’s surface (4300) Myr ago. Nature, 409, 178–181. Mueller, P.A., Foster, D.A., Mogk, D.W., Wooden, J.L., Kamenov, G.D., and Vogl, J.J. (2007) Detrital mineral chronology of the Uinta Mountain Group: implications for the Grenville flood in southwestern Laurentia. Geology, 35 (5), 431–434. Nasdala, L., Zhang, M., Kempe, U., Panczer, G., Gaft, M., Andrut, M., and Plotze, M. (2003) in Hanchar, J.M., Hoskin, P.W.O., and Kinny, P. (2003) Atlas of zircon textures, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 427–467. Nemchin, A.A., and Cawood, P.A. (2005) Discordance of the U-Pb system in detrital zircons: Implication for provenance studies of sedimentary rocks. Sedimentary Geology, 182, 143–162. Parrish, R.R. (1987) An improved microcapsule for zircon dissolution in U-Pb geochronology. Chemical Geology, 66, 99–102. Parrish, R.R., and Krogh, T.E. (1987) Synthesis and purification of 205Pb for U-Pb geochronology. Chemical Geology, 66, 103–110. Parrish, R.R., and Noble, S.R. (2003) Zircon U-Th-Pb geochronology by isotope dilution-thermal ionization mass spectrometry (ID-TIMS) in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 183–213. Rahl, J.M., Reiners, P.W., Campbell, I.H., Nicolescu, S., and Allen, C.M. (2003) Combined single-grain (U-Th)/He and U/Pb dating of detrial zircons from the Navajo Sandstone, Utah. Geology, 31 (9), 761–764. Ross, G.M., and Bowring, S.A. (1990) Detrital zircon geochronology of the Windermere Supergroup and the tectonic assembly of the southern Canadian Cordillera. Journal of Geology, 98, 879–893. Scherer, E.E., Whitehouse, M.J., and Munker, C. (2007) Zircon as a monitor of crustal growth. Elements, 3 (1), 19–24. Silva, L.C., Rodrigues, J.B., Silveira, L.M.C., and Pimentel, M.M. (2003) The Brazilian National Geochronological Database: Chronobank, in da Silva Rosa, M., ed., IV South American Symposium on Isotope Geology: Salvador, 115–116.
62
Part 2: New Techniques and Modeling
Silver, L.T., and Deutsch, S. (1963) Uranium-lead isotopic variations in zircon: a case study. Journal of Geology, 71, 721–758. Simonetti, A., Heaman, L., and Chacko, T. (2008) Use of discrete-dynode secondary electron multipliers with faradays: a “reduced volume” approach for in situ U-Pb dating of accessory minerals within petrographic thin sections by LA-MC-ICP-MS, in Sylvester, P., ed., Laser ablation ICP-MS in the earth sciences: current practices and outstanding issues. Short Course Series, 40. Quebec, QC, Mineralogical Association of Canada, 241–264. Sircombe, K.N. (2004) AGEDISPLAY: an EXCEL workbook to evaluate and display univariate geochronological data using binned frequency histograms and probability density distributions. Computers and Geosciences, 30 (1), 21–31. Sircombe, K.N., and Hazelton, M.L. (2004) Comparison of detrital zircon age distributions by kernel functional estimation. Sedimentary Geology, 171, 91–111. Sircombe, K.N., and Stern R.A. (2002) An investigation of artificial biasing in detrital zircon U-Pb geochronology due to magnetic separation in sample preparation. Geochimica et Cosmochimica Acta, 66 (13), 2379–2397. Sloan, J., Henry, C.D., Hopkins, M., and Ludington, S. (2003) National Geochronological Database: US Geological Survey, Open-File Report 03-236. Steiger, R.H., and J€ ager, E. (1977) Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359–362. Tera, F., and Wasserburg, G.J. (1972) U-Th-Pb systematic in three Apollo 14 basalts and the problem of initial Pb in lunar rocks. Earth and Planetary Science Letters, 14, 281–304.
Ushikubo, T., Kita, N., Cavosie, A., Wilde, S., Rudnick, R., and Valley, J. (2008) Lithium in Jack Hills zircons: evidence for extensive weathering of Earth’s earliest crust. Earth and Planetary Science Letters, 272 (3–4), 666–676. Valley, J. (2003) Oxygen isotopes in zircon, in Hanchar, J.M., and Hoskin, P.W.O., eds., Zircon: Reviews in Mineralogy and Geochemistry, 53, 343–385. Vermeesch, P. (2004) How many grains are needed for a provenance study? Earth and Planetary Science Letters, 224, 441–451. Watson, E.B., Wark, D.A., and Thomas, J.B. (2006) Crystallization thermometers for zircon and rutile. Contributions to Mineralogy and Petrology, 151, 413–433. Wetherill, G.W. (1956) Discordant uranium-lead ages. International Transactions of the American Geophysical Union, 37, 320–326. Wooden, J.L., Mazdab, F.K., and Barth, A.P. (2007) Using the temperature and compositional characteristics of zircon and sphene to better understand the petrogenesis of Mesozoic magmatism in the Transverse Ranges, California. Proceedings of the Ores and Orogenesis Conference, Tucson, AZ, 154. Woodhead, J., Hergt, J., Shelley, M., Eggins, S., and Kemp, R. (2004) Zircon Hf-isotope analysis with an excimer laser, depth profiling, ablation geometries, and concomitant age estimation. Chemical Geology, 209, 121–135. Yuan, H., Gao, S., Dai, M., Zong, C., Gunther, D., Fontaine, G., Liu, X., and Diwu, C. (2008) Simultaneous determinations of U-Pb age, Hf isotopes, and trace element compositions of zircon by excimer laser-ablation quadrupole and multicollector ICP-MS. Chemical Geology, 247, 100–118.
