1

Wildfire thermochronology and the fate and transport of apatite in hillslope

and fluvial environments

Reiners, P.W.1, Thomson, S.N.2, McPhillips, D.2, Donelick, R.A.3, and Roering, J.J.4

1Department of Geosciences, University of Arizona, Tucson, AZ 85721 2Department of Geology & Geophysics, Yale University, New Haven, CT 06511 3Apatite to Zircon, Inc., 1075 Matson Rd., Viola, ID 83872 4Department of Geological Sciences, University of Oregon, Eugene, OR 97403

Index terms: 1140 Thermochronology, 1130 Geomorphological geochronology, 1625

Geomorphology and weathering, 1815 Erosion

Keywords: Wildfire, fission track, (U-Th)/He, thermochronology, detrital

2

Abstract Wildfire heating of the outer few centimeters of exposed rock or soil generates short duration, high temperature thermal events that produce characteristic thermochronologic signatures in minerals. Contrasting activation energies of fission-track annealing and He diffusion in apatite lead to a kinetic cross-over whereby wildfire heating resets fission-track (FT) ages much faster than (U-Th)/He ages, resulting in “inverted” FT-He ages in single grains. This can be used to trace wildfire-affected detritus at the Earth’s surface. We show that in exposed bedrock, inverted apatite FT-He ages vary systematically with depth to ~3 cm, and detrital clasts on hillslopes also show strong but heterogeneous wildfire-resetting signatures. In soils, colluvium, and low-order channel sediments, strongly wildfire-reset apatite grains are abundant, and in some cases dominate the population of detrital apatite, to depths at least as great as 10 cm. Wildfire-reset apatite is rare, however, in fluvial sediments sampled from larger basins, indicating a strong fractionation of apatite populations from hillslopes to rivers. Characteristic dissolution features in hillslope apatite and slower relative dissolution rates of other common minerals suggest that wildfire-reset apatite grains are rare or absent in rivers because they dissolve relatively rapidly in soil profiles. Apatite that does contribute to fluvial sediments is likely to be dominantly derived from bedrock landslides in steep regions or from large clasts containing grains protected from both wildfire heating and dissolution. This means that apatite in fluvial sediment is spatially fractionated with respect to its sources in the catchment, even if catchment erosion rates are spatially uniform. Introduction Cooling ages of low-temperature thermochronometers, such as those of the apatite fission-track (AFT) and apatite (U-Th)/He (AHe) systems, are commonly used to infer timing and rates of erosional or tectonic exhumation in the shallow crust (e.g., Gallagher et al., 1998; Farley, 2002). Most applications using apatite have focused on bedrock samples, but both AFT and AHe have also been applied to detrital apatite to study sediment provenance and exhumation histories of orogenic source regions (e.g., Hurford and Carter, 1991; Thomson, 1994; Carter, 1999; Garver et al., 1999; Bernet and Garver, 2005). Increasing interest in the detailed spatial and temporal patterns of erosion within drainage basins have also motivated studies of detrital AHe (Stock et al., 2006) and AFT (Vermeesch, 2007) ages in modern fluvial sediment, using approaches similar to those that have been used for detrital studies of higher temperature thermochronometric systems (Brewer et al., 2003; Ruhl and Hodges, 2005; Hodges, 2005; Stock et al., 2006; Huntington and Hodges, 2006). These approaches also hold potential for reconstructing paleotopography of catchments from ancient detritus (Stock and Montgomery, 1996; Reiners, 2007).

Low-temperature thermochronology using both apatite and other minerals has a long history that does not immediately suggest that surficial wildfire causes widespread or significant age resetting. However, both theory (Wolf et al., 1998) and observations (Mitchell and Reiners, 2003) have shown that wildfire can strongly partially reset (U-Th)/He ages of apatite, and to a lesser degree zircon, in the outermost ~3 cm of exposed bedrock, and in detrital pebbles on the soil surface. Although certain sample collection or preparation techniques can mitigate against

3

undesired detection of possible wildfire effects in bedrock samples, unless whole clasts greater than ~6 cm in diameter can be collected, this is not possible for detrital samples.

Whether the thermochronologic signatures of detrital apatite grains on hillslopes or in rivers have been affected by wildfire likely depends on several factors including the regional characteristics of wildfire and the way in which detrital apatite is liberated from bedrock and transported through soil, colluvium, and fluvial environments. If detrital apatite does bear a thermochronologic signature of wildfire, it may, in addition to confounding conventional exhumation-related studies, provide useful information as a tracer of fires over the landscape or of soil- or sediment-forming processes. It may also prove useful for detecting evidence of paleowildfire in some types of ancient materials.

In this paper we describe a diagnostic thermochronologic signature of wildfire resulting from different kinetics of fission-track annealing and He diffusion. We then examine the extent and abundance of this signature in detrital apatite from a variety of settings. Our results show that detrital apatite in soils and some low-order streams contain abundant and strong signatures of wildfire resetting, but that few grains in high-order fluvial environments do. This, combined with morphologic differences between apatites in hillslope and fluvial sediment, and considerations of apatite dissolution rates, suggest that most apatite in fluvial sediment has shortcut surficial hillslope transport pathways, and is instead derived from relatively large clasts delivered via landsliding along steep slopes. Apatite transported by slower soil creep along low slope regions is more likely to bear a wildfire resetting signature, but is also more likely to be dissolved along the way. Even if erosion rates in a catchment are spatially uniform, apatite yield is not, and fluvial detritus is spatially fractionated with respect to its bedrock source. Contrasting kinetics of He diffusion and fission-track annealing He diffusion Resetting of (U-Th)/He and FT ages, whether partial or complete, occurs by diffusive loss of He and annealing of fission tracks. The fractional loss f, of He resulting from a heating event at temperature T for duration t, is

( ) ( )∑∞

−−=1

22222 )exp(/1/61 t

aDnnf ππ (1),

(Crank, 1975) where D is the diffusion coefficient and a is the radius of the diffusion domain with assumed spherical geometry. The spherical domain assumption is arguably justified for most applications because actual crystal geometry makes a negligible difference as long as measured and modeled data use the same convention (Farley and Stockli, 2002; Meesters and Dunai, 2002a). Diffusion is assumed to follow an Arrhenius law of the form

[ ]RTEaD

aD

a /exp20

2 −= (2),

where D0 and Ea are the experimentally determined frequency factor and activation energy, respectively, and R is the gas constant. Shuster et al. (2006) have proposed a more complicated Arrhenius law to account for decreased diffusivity as a function of accumulated radiation

4

damage in apatite. This effect may be important in cases of old and high-U apatite, and although we do not include it here, could be incorporated in the modeling that follows. Two additional considerations simplify fractional loss calculations and make them more versatile for modeling. The first is that equation (1) can be approximated without the use of infinite sums, for example as

( ) )(/3)(/6 2222

1

222

3t

aDt

aDf ππππ −−⎟

⎠⎞⎜

⎝⎛≈ , f < 0.85 (3),

( ) )exp(/61 222 t

aDf ππ −−≈ , f > 0.85 (4),

(e.g., McDougall and Harrison, 1999). The second is recognition that Dt/a2 in equations 1, 3, and 4 can refer to not only a square-pulse heating event of constant temperature, but any arbitrary thermal history, if it is replaced with “reduced time” τ, where

')'(

1exp),(02

0 dttTR

EaDtT

ta∫ ⎥

⎦

⎤⎢⎣

⎡−=τ (5).

A heating event causing resetting can be uniquely associated with fractional loss through

Dt/a2 or τ, and contours of fractional loss, and therefore resetting of the (U-Th)/He system, can be plotted as a function of log time and inverse temperature (Figure 1). In this context, each fractional loss contour is a straight line with intercept β and slope Ea/R, where β is a function of D0/a2 and f.

For calculating He fractional loss from apatite, we used diffusion parameters determined for Durango apatite (Farley, 2000) [D0 = 31.6 cm2/s; Ea = 138 kJ/mol (32.9 kcal/mol)], and two different effective diffusion domain sizes of 50 and 100 μm (Figure 1). For zircon, we used diffusion parameters from Reiners et al. (2004) [D0 = 0.46 cm2/s; Ea = 169 kJ/mol (40.4 kcal/mol)], and the same diffusion domain sizes.

Fission-track annealing Compared with phenomenological models for He diffusion and fractional loss/resetting, annealing of fission tracks in apatite is complex. Apatite composition and other poorly understood characteristics affect the kinetics of annealing (e.g., Ketcham, 2005), but for the purposes of this study, it is sufficient to consider the typical annealing kinetics that characterize most natural apatite.

Most kinetic models describing fission-track annealing as a function of time and temperature bear some functional similarity to fundamental equations of atomic motion (Laslett et al., 1987; Carlson, 1990; Crowley et al., 1991; Ketcham et al., 1999; Ketcham, 2005), but in practice are largely empirical, containing five or six parameters fitted to experimental data. The models are based on track length shortening, which can be empirically related to track density, and therefore AFT age (e.g., Green, 1988). Here we show several of the most commonly used annealing models. These include the fanning linear model of Laslett et al. (1987), based on Durango apatite, and the fanning linear model of Crowley et al. (1991), based on fluorapatite in general. We focus largely on the fanning curvilinear and linear models of Ketcham et al. (1999),

5

which are based on the experiments of Carlson et al. (1999) using a variety of apatite specimens, laboratory experiments, and natural thermal history constraints. We use Ketcham et al.’s “RN” (Renfrew apatite), because it is most representative of typical apatite compositions and annealing behaviors. In the range of time-temperature conditions relevant to wildfire resetting, other kinetic models yield similar results.

Both the fanning linear and curviliear models describe shortening of fission tracks by r = l/l0 (length divided by pre-annealing length), as a function of log time and inverse temperature. In the Ketcham et al. (1999) model, lengths are reduced mean c-axis projected lengths (Ketcham, 2005), to account for anisotropic annealing. Contours of complete annealing can be considered to correspond to r values where track density (and thus age) rapidly approaches zero (r = 0.55), and contours of zero annealing correspond to shortening that occurs at room temperatures over geologic time scales (r = 0.93) (Ketcham, 2005).

The linear fanning model (after Laslett et al., 1987) describes track shortening as a function g(r), where

( )[ ] ( )

( ) ⎥⎦

⎤⎢⎣

⎡−−

+=−−

=3

210 /1

ln1/1)(cTctccrrg

αβ αβ

(6),

where α, β, c0, c1, c2, and c3 are empirical parameters based on experimental data. Equation 6 produces fanning linear trends of constant track length (Figure 1). The only mathematical difference between equation 6 and the fanning curvilinear model (Ketcham et al., 1999) is that the (1/T) term is replaced by ln(1/T) in the latter model, producing curvilinear annealing contours (Fig. 1). Parameters for Ketcham et al.’s (1999) linear and curviliear models were taken from their Table 5e. In order to specify annealing kinetics of Ketcham et al.’s (1999) most typical RN apatite, we replaced r in equation 6 with rmr, where

( )( ) 001

1 mrmrlrmr rrrr +−= κ (7), where rlr is the reduced fractional track length of interest (0.55 and 0.935 in Figures 1 and 2), and κ and rmr0 apply to RN apatite (Table 4b of Ketcham et al., 1999). Figure 2 shows resetting contour plots for fission tracks and the (U-Th)/He system in zircon. For zircon, we used a simplified fanning linear model:

( ) TctTccr 210 ln)1ln( ++=− (8), and constants for mixed-alpha damage zircon (Tagami et al., 1998) and zero-damage zircon (Rahn et al., 2004). Predictions of wildfire signatures from resetting contour plots

Fractional loss and annealing equations can be rearranged to highlight the relationship between explicit kinetic parameters like activation energy, and important features of resetting in Figures 1 and 2. For fractional He degassing, equations 3 and 4 can be rearranged to yield

6

( )⎟⎠⎞

⎜⎝⎛+

⎪⎭

⎪⎬⎫

⎪⎩

⎪⎨⎧

⎥⎥⎦

⎤

⎢⎢⎣

⎡ −−=

TRE

Daft a 1

3391ln)ln(

0

22

21

21

π

π (9),

and

( ) ⎟⎠⎞

⎜⎝⎛+

⎭⎬⎫

⎩⎨⎧

⎥⎦

⎤⎢⎣

⎡−−=

TRE

Daft a 11

6lnln)ln(

02

22

ππ (10),

respectively. Similarly, for fission-track annealing, equations 6 and 8 can be rearranged to yield

( ) ⎟⎠⎞

⎜⎝⎛−

+−−=Tc

crgcrgccct 1)()()ln(

1

00

1

32 (11),

and

( )⎟⎠⎞

⎜⎝⎛+−

+−=Tc

crgcct 1)(1ln)ln(

1

0

1

2 (12),

where g(r) = {[1-rβ)/β]α-1}/α (the left hand side of equation 6). For the curvilinear fanning model of Ketcham et al. (1999), the analogous equation is as in equation 11, except the last term goes with ln(1/T) instead of (1/T). Equations 9-12 describe lines or curves in Figures 1 and 2. For fractional He degassing, the intercepts are functions of constants, extents of degassing, and D0/a2, whereas slopes are Ea/R. For fanning linear models of fission-track annealing, both intercepts and slopes are functions of constants and extents of annealing (except for zircon fission-track, whose intercept depends only on constants). The slope of the curvilinear fanning model (Ketcham et al., 1999) is also a function of temperature [T(g-c0)/c1]. For fission-track annealing, these combinations of constants, and in some cases g and T, are analogous to a factor that is proportional to activation energy, as in the fractional degassing cases.

The fact that the slope of fractional loss and annealing contours are proportional to Ea underscores the role of activation energy in the time dependence of resetting. Systems with high activation energies (steep fractional loss contours) can experience fractional loss and partial resetting to a given extent if held only for short durations, but require relatively high temperatures. Systems with low activation energies (shallow fractional loss contours) can experience fractional loss and partial resetting to the same extent at much lower temperatures, but only over longer durations.

Except for very long timescales and low temperatures in the curvilinear model for apatite, contours of fission-track annealing are significantly steeper than those of fractional He degassing. This is most pronounced at extents of annealing approaching those where fission-track density and therefore age goes to zero (i.e., at higher temperatures). This contrast is reflected in higher activation energies of fission-track annealing relative to He diffusion. Activation energies for He diffusion in apatite and zircon are approximately 138 kJ/mol (32.9 kcal/mol) and 169 kJ/mol (40.4 kcal/mol), respectively, compared with those for 90% annealing of fission tracks of 161 kJ/mol (38.4 kcal/mol) and 221 kJ/mol (52.8 kcal/mol) to 339 kJ/mol

7

(80.9 kcal/mol) for the same minerals (Reiners and Brandon, 2006). These relative activation energies, combined with specific D0 and effective sizes (a) of typical apatite, and mechanistically equivalent terms in constants for fission-track annealing, leads to a triangle-shaped region encompassing time-temperature combinations of heating events resulting in complete fission-track age resetting but only partial He age resetting (Fig. 1). Given the expectation that fission-track ages should be at least as old as He ages for monotonically cooled rocks, this is a potentially diagnostic indicator of short-duration, high temperature heating such as might occur near rock or soil surfaces during wildfire. For apatite and heating timescales of ~1 hour, this results from heating to temperatures of 350-450 °C. At shorter timescales, a broader window of higher temperatures has the same effect (e.g., ~400-625 °C for timescales of ~1 minute); at longer timescales, a narrower window of lower temperatures can produce the same effect (e.g., ~225-250 °C for 1 year) (Fig. 1).

Figures 1 and 2 suggest that short duration thermal events may also partially reset fission-track ages to greater extents than He ages. This would in fact occur for t-T combinations covering most of the region between the r = 0.93 and r = 0.55 annealing contours, for timescales less than about 1 year. In other words, if a short-duration thermal event does not completely anneal fission tracks, it is likely to cause greater partial resetting of the fission-track than the He system, producing what we call inverted FT-He ages. As in the case of finite He age and zero fission-track age, this may provide strong evidence for a short-duration reheating event. Perhaps more interestingly, partial resetting of both systems uniquely constrains Dt/a2, which could be used to constrain the temperature, duration, or t-T histories of a wildfire. Alternatively, assuming an average wildfire thermal history, a unique Dt/a2 could potentially be used to estimate the number of wildfire events affecting a sample, or, with an additional estimate of the duration over which they occurred, their recurrence interval.

It should be noted that whereas annealing and diffusion predictions for typical exhumation paths require extrapolations across several orders of magnitude of both time and temperature, short duration wildfire heating events require no such extrapolation, as they reflect the same time and temperature conditions (Fig.s 1 and 2) used in experiments to derive the (empirically-based) kinetic models.

Assuming relationships between track length, density, and age suggested by Ketcham (2005), we calculated contours of constant time and temperature for square-pulse thermal histories in a plot of fractional resetting of AFT and apatite He ages (Figure 3). The contours in Figure 3 show constant time or temperature trajectories of resetting in these systems. At temperatures less than about 240 °C, significant partial resetting of these systems requires long durations and results in greater resetting of the He system than the FT system (e.g., point A in Fig. 3). Such conditions are characteristic of subsurface geologic processes, such as sedimentary burial. At higher temperatures, if the thermal event does not completely reset at least one of the systems, the duration of the event must be relatively small, and this results in inverted partially reset ages (e.g., point B in Fig. 3; ~2.7 hours at ~325 °C) or completely reset FT with a partially reset He age (e.g., point C; ~5 minutes at ~400 °C). As shown below, thermal histories characteristic of wildfires, and resulting FT-He ages, follow steep trajectories in this plot. Wildfire thermal histories Short-term thermal histories of shallow or exposed bedrock and soil during wildfire are not well known, and are likely quite variable, even within small areas and single burn events. Mitchell and Reiners (2003) reviewed several cases from the literature, noting estimates of rock and soil

8

temperatures ranging as high as 600-700 °C, and characteristic heating times with temperatures >100 °C on the order of minutes to tens of minutes.

Figure 4 shows time-temperature histories of the mineral-organic soil interface (0.5 cm depth) in several experimental summer fires in the Valencia region of the Mediterranean (Molina and Llinares, 2001). As an example of 1) conversion of arbitrary thermal histories to Dt/a2 (τ) values, and 2) stronger AFT resetting than AHe resetting for typical wildfire events, we estimated Dt/a2 (τ) and fractional AHe and AFT resetting for three of these curves. AFT ages were calculated using HeFTy (Ketcham, 2005). The history reaching the highest temperature completely reset the AFT system and nearly completely reset the AHe system. The case with the intermediate maximum temperature caused complete resetting of the AFT system, but only caused 65% He loss. The lowest temperature history caused 9% He loss and 15% fission-track age reduction. These simple simulations are consistent with predictions that typical wildfires should cause greater age resetting in the AFT system than in the AHe system. Samples We examined apatite from hillslopes, bedrock outcrops, and low-order and high-order channel sediments from several locations on the eastern flank of the central Washington Cascades and in the northern and central Sierra Nevada (Table 1).

Apatite and zircon (U-Th)/He ages from some of the Cascades samples (SGM samples) were reported in Mitchell and Reiners (2003); other SGM samples reported here were taken within several kilometers of these, all in the Icicle Creek drainage (Table 1). Most of the Icicle Creek drainage is underlain by granodioritic/dioritic intrusive rocks of the 95-Ma Mount Stuart batholith, though Jurassic schists and mafic/ultramafics are also present. The drainage was glaciated in the Pleistocene and moraines have modified channel morphology throughout the valley (Lorang and Aggett, 2005). All Cascades samples except 04WFC9 were taken from regions that burned with variable intensity in August 2001. The dominant vegetation is Ponderosa Pine and Douglas Fir, with estimated fire recurrence intervals of about 15-25 years (Agee, 1993; Everett et al., 2000). The mean annual temperature and precipitation are approximately 9 °C, and ~65 cm/yr, respectively.

SGM samples are exposed bedrock and scattered detrital chips of rock on the soil surface from a broad hillslope on the south side of Icicle Ridge (Table 1; Mitchell and Reiners, 2003). Sample 04WFC1 is sand and gravel from a low-order channel on a steep slope about 600 m above the SGM samples, and 04WFC1-6 are soil samples from a broad hillslope about 150 m below 04WFC1 (Table 1). Sample 04WFC9 is river sand integrating ~550 km2 of the Icicle Creek drainage just above the town of Leavenworth. Apatite (U-Th)/He ages of fresh bedrock from this part of Icicle Ridge range from ~18 to ~30 Ma at elevations of ~400 to ~2400 meters, respectively. A few AFT ages from samples at intermediate elevation in this transect (~900-1400 meters) range from 40-50 Ma (Reiners et al., 2002). Elsewhere in the Icicle Creek drainage, apatite He ages range from about 9-55 Ma (unpublished data), and zircon He ages range from 54-74 Ma (Reiners et al., 2002).

Samples 04WFS12 and 13 are bedrock and soil samples from a forested, low-relief part of the central Sierra Nevada (Table 1) underlain by Mesozoic Sierran batholithic granitoids. Samples 04WFS14 and 15 are fluvial sand sediments from the San Joaquin and Kings Rivers, in the western foothills. The drainages above each of these samples are 3480 – 3750 km2, and are

9

underlain primarily by Mesozoic Sierran batholithic granitoids, with some Mesozoic metavolcanics and other areally minor lithologies. Vegetation in this part of the western Sierra ranges widely but is dominantly mixed conifer and fir forests, with some alpine vegetation and chapparal at higher and lower elevations, respectively. These detrital samples are derived from basins covering a wide range of annual precipitation, from ~50-200 cm/yr.

All 05WFS samples are from Mesozoic Sierran batholithic rocks in the northeastern Sierra Nevada (Table 1; Fig. 5), from sites, or near sites, where erosion rates and their relationships with climatic and topographic variations have been determined with cosmogenic nuclides (Granger et al., 1996; Riebe et al., 2000; 2001a; 2001b; 2004). These areas were not glaciated in the Pleistocene. Erosion rates over the last 103-104 years in these regions, as measured by cosmogenic nuclides, range from approximately 0.02-0.2 mm/yr (Riebe et al., 2001). Mean annual temperatures for all 05WFS samples except Adams Peak are 9-12 °C; Adams Peak is 4 °C. Mean annual precipitation for Fall River and Grizzly Dome are 145-180 cm/yr, Adams Peak and Antelope Lake are 60-85 cm/yr, and Fort Sage is 25 cm/yr (Riebe et al., 2000). Vegetation at most 05WFS sites is mixed conifer or fir forests, but Fort Sage is sagebrush scrub. Samples 05WFS1 through 4 were collected in a region that burned in the 2001 “Stream Fire.” Most trees in the immediate vicinity of these samples were killed by this fire, suggesting relatively intense burning. The 05WFS samples are from a variety of geomorphic settings, including fresh interiors of bedrock tors or massifs, soils on broad hillslopes or colluvial hollows, and low-order channel sediment (Table 1). Methods Concentrated apatite fractions of each sample were prepared by standard crushing, sieving, magnetic, and (heavy-liquid) density mineral separation procedures. Apatite grains dated by (U-Th)/He methods (without accompanying FT dating) were selected by routine microscopic inspection and analytical methods (House et al., 2000; Reiners et al., 2003). Fission-track ages were measured in two different labs. The fission track samples analyzed at Yale followed the methodology outlined in Thomson and Ring (2006), using a CN5 glass to monitor neutron fluence during irradiation at the Oregon State University Triga Reactor, Corvallis, USA. A CN5 apatite zeta calibration factor (Hurford and Green, 1983) of 356.1±15.3 was used in age calculation. Some samples were analyzed at A2Z, Inc., following procedures outlined in Donelick et al. (2005).

A subset of apatite grains was dated by both He and FT methods, by plucking selected polished and FT-dated grains directly from epoxy and processing them through standard He dating procedures (e.g., Reiners et al., 2004).

Polishing and removing part of the apatite crystal in the FT-dating procedures removes part of the alpha-ejection affected rim, modifying the grain’s alpha-ejection correction for He dating. The modified alpha-ejection correction for the polished grain can be estimated by assuming that polishing was subparallel to the c-axis of the crystal and removed a thickness more than one stopping distance (~20 μm) and less than one-half of the crystal width. Under these conditions, the modified surface-area to volume ratio (β) of the polished grain (for an assumed pre-polishing cylindrical geometry with pinacoidal terminations and not including the surface area of the polished face) is

10

22 2)(22

drddrrr

l −−++=

ω

ωβ (13),

Where d is polishing depth, l is crystal length, r is c-axis perpendicular half-width (radius), and

⎟⎠⎞

⎜⎝⎛ −−= −

rd1cos 1πω (14).

We then assume that this β can be used in Farley’s (2002) standard polynomial relating β and FT for apatite, which we justify by noting the similarity in FT for a wide range of assumed morphologies, as long as β is sufficiently large. A similar and simpler approach is also possible for zircon, if the pyramidal terminations are ignored, so that

drrl −++=

21 2112β (15),

where r1 and r2 are the c-axis-perpendicular half-widths of the crystal parallel and normal to the polishing directions, respectively. This equation can be used with the standard polynomials relating β and FT for zircon (Farley, 2002; Hourigan et al., 2005).

Figure 6 shows that as long as polishing removes a thickness greater than one stopping distance (17-20 μm for zircon and apatite, respectively) and less than one half-width of the grain, its estimated effect on the alpha-ejection correction, relative to the whole unpolished grain, is generally less than about 2-5% (Figure 6) for typical crystal sizes. Because we don’t know the polishing depth for each grain precisely, we have assumed the standard FT correction here, based on the measured length and polishing-perpendicular width of each grain. We also do not report parent or daughter concentrations for these grains, because routine concentration estimates by AHe dating methods require a simplifying morphological assumption for each grain that is violated when polished partial grains are used (note that this has no effect on the age calculation, however).

Results Table 2 shows summaries of all apatite FT dating results. Individual grain data for all samples are in the Supplementary Tables. Table 3 shows results of combined FT and (U-Th)/He dating on single grains from a subset of these samples, and Table 4 shows results of single-grain (U-Th)/He analyses. Most of the single-grain He dates were performed on detrital grains from Icicle Creek (04WFC9), and the San Joaquin (04WFS14) and Kings (04WFS15) rivers, but several are from fresh bedrock in the northern Sierra (05WFS5) and hillslope and low-order channel deposits in both the Sierra and Cascades. Small-scale bedrock transect samples AFT ages from the 6-cm bedrock transect analyzed by Mitchell and Reiners (2003) show exclusively young or zero ages in the outermost section, old ages in the innermost section, and mixed age populations in the two intermediate-depth samples. Figure 7 shows probability density

11

plots and binomial peak fits (Brandon, 1992) to AFT grain ages. Apatite grains from the deepest sample (3-6 cm depth) have a unimodal age peak (but with a relatively poor chi-squared of ~1%) of 42.9+4.1

-3.8 Ma. Nearly 90% of apatite grains in the outermost 1 cm have AFT ages indistinguishable from zero, though a few have poorly-resolved ages younger than ~5 Ma. AFT ages from samples in the 1-2 cm and 2-3 cm portions show relatively large scatter. About 40-50% of these grains have zero ages or poorly resolved ages less than ~2 Ma. However, several grains with strongly shortened tracks from both samples have widely ranging ages, from ~5-40 Ma. The largest probability peaks in these intermediate depth samples are ~30-40 Ma, significantly younger than the central age of the deepest sample. AFT ages of grains from small chips on the forest floor (“SGMchips”) also yielded a wide range of ages, with nearly half indistinguishable from zero. The etch-pit feature Dpar, considered a proxy indicator for fission-track annealing kinetics in apatite (Donelick et al., 2005), shows a weak correlation with AFT age in the partially reset grains from 2-3 cm depth, and to a lesser degree in the sample from 1-2 cm depth (Supplementary Tables). He-FT “double dates” on the bedrock transect Combined He-FT “double dates” on the exposed bedrock transect samples are shown in Figure 8. Because of the low density of natural tracks in many of the apatite grains from these samples (Ns <5), we show 50% probability ages, rather than central ages, with 95% confidence intervals for a binomial parameter (Brandon et al., 1998) using the formulae given by Galbraith (2005, p. 50-52). The maximum apatite FT and He ages measured in these samples correspond well to those measured on grains analysed by only a single technique. Most grains in the deepest subsample having AFT ages of ~40-50 Ma and apatite He ages of ~16-19 Ma, whereas most grains from shallower and detrital subsamples range widely to younger He and FT ages, but only grains with He ages greater than ~14 Ma show FT ages indistinguishable from zero. Comparison of the measured He-FT double dates for most samples with model time-temperature trends for partial resetting are consistent with equivalent square-pulse heating temperatures and durations of ~270-400 °C and a few minutes to a few hundred hours. Bedrock, soil, colluvial, and low-order channel samples

AFT grain age populations from most samples of soil, colluvial hollows (or unchanneled valleys), exposed bedrock, and creek sand show a wide range of ages, with a large population of zero-age or near zero-age peaks. Representative photomicrographs of apatite grains in these detrital samples are shown in Figure 9. These images are from two samples with roughly subequal proportions of grains with ages matching those of fresh bedrock in the region and grains with zero or near-zero ages. The distinction in these populations can be clearly seen in spontaneous track abundances and their comparisons with mirror-image, induced-track abundances. In the following paragraphs we briefly survey the general results from each subregion, along with their regional contexts, focusing on the proportion of grains with zero or near-zero AFT ages.

In the Fourth of July trail region of the Icicle Creek drainage of the Washington Cascades regional bedrock AFT ages are approximately 40-60 Ma (Reiners et al. 2002). Binomial peak fitting of AFT ages of detrital grains from the upper ~10 cm of soil and low-order channel sediments shows two dominant age peaks, with subequal proportions in each sample. One of these peaks is consistent with the regional bedrock ages (40-60 Ma), but the other one,

12

comprising approximately 40-60% of grains, has a zero age, or one statistically indistinguishable from it (Fig. 10). In the northeastern Sierra Nevada (Fig. 5), AFT age patterns are not known in detail, but He ages of fresh bedrock samples from two areas we sampled are similar to most of those in the northern Sierra as a whole (House et al., 1998; 2001; Cecil et al., 2006). A fresh bedrock sample from the Antelope Lake area yielded a well constrained central AFT age of 72.8 ± 4.5 Ma (1σ) and six single-grain apatite He ages averaging 66.8 Ma, with one standard deviation of 3.6 Ma (Tables 2, 4; Fig. 11). A single fresh bedrock sample from the base of Grizzly Dome in the canyon of the North Fork of the Feather River also yielded a similar central AFT age of 68.3 ± 4.5 Ma (1σ) (Fig. 12). Binomial peak fitting of AFT grain ages from soil and colluvial hollow samples from the Antelope Lake area of the Sierra show results similar to those in the Cascades. Two age peaks dominate (Fig. 11). The younger one, comprising between 45 and 92% of grains is indistinguishable from zero, and the older one is similar to the age of the fresh bedrock. A single sample of low-order channel sand from the region yielded a slightly smaller proportion of zero age grains: 29%. Results from the Adams Peak area (Fig. 13) are similar. Between 40 and 65% of grains from colluvial hollow sediments have AFT ages indistinguishable from zero, while the remainder fall into a peak similar in age to that of bedrock from the nearby Antelope Lake area. In the Grizzly Dome area (Fig. 12), AFT grain ages in two samples from the higher, low-relief areas show 32 and 34% zero-age peaks, with older peaks consistent with regional ages. Samples from low-order channel sand at the bottom of the Dome, in the Feather River Canyon however, show a much smaller proportion of zero-age grains: 2-10%. Notably, these are the largest creeks sampled in this study, and they also drain the steepest terrain. AFT grain ages from colluvial hollow samples in the Fort Sage area (Figure 14) show a much smaller proportion of zero- or near zero-age grains. In three out of four samples only one to two grains have zero ages (1-2% of the population), though in one sample, 16% of grains have zero ages. The older age peak is similar to but in some cases slightly younger than older-peak ages in the other regions. In the San Joaquin drainage, a single sample of exposed bedrock about 3 cm thick (04WFS12) yielded apatite with a wide range of AFT grain ages, from 0.9 to 13 Ma (Fig. 15). Though only a few grains were dated, these grains are significantly younger than the much older regionally consistent AFT ages of ~70-80 Ma. A nearby soil sample (04WFS13) also yielded apatite with a similarly wide range of AFT ages, several of which were much younger than the regional bedrock ages (Fig. 15). High-order channel (river) samples In contrast to most soil, colluvial hollow, and low-order channel samples, AFT grain-age populations from fluvial sediment from relatively large rivers show few or no zero-age or near-zero age grains. Out of 100 grains dated from the Icicle Creek sample, approximately 9% fell into the youngest grain-age peak, which is easily distinguishable from other peaks and close to zero-age (Fig. 10). This is much lower than the 50-61% and 40% from nearby soil and low-order channel samples, respectively. Similarly, none of the fifty grains dated from the San Joaquin river sample showed zero or near-zero ages (Fig. 15). This detrital sample yielded a single age peak of 75.7 +9.8

-8.8 (95% confidence interval) with a chi-squared probability of 99.9%. (U-Th)/He dating of a separate set

13

of 46 grains from this sample also failed to yield a single grain with anomalously young ages (Table 4). In the Kings river sample, 3% of grains fell into a distinct age peak much younger than the 66.1 +9.1

-8.0 Ma peak (95% confidence interval, chi-squared probability of 94%) of the other 97% of grains (Fig. 15). A few anomalously young grains was also confirmed by (U-Th)/He dating of a separate set of 52 grains from this sample, which yielded two grains with 0.14 and 2.6 Ma apatite He ages (Table 4). He-FT double dates on hillslope and fluvial samples Figure 16 shows He-FT double dates on single apatite grains from 1) soil, exposed bedrock, and fluvial samples from the San Joaquin drainage, and 2) fluvial samples from the Kings river, and 3) soil and fluvial samples in the Icicle Creek drainage in the Cascades. In the Sierran samples (Fig. 16A), apatite He-FT ages from the San Joaquin river (04WFS14) fall above or within uncertainty of the 1:1 He-FT age relationship, at ages similar to those of both AFT and apatite He age distributions that show no evidence for partial resetting of either system, consistent with the AFT peak fitting of this sample (Fig. 15). The Kings river sample (04WFS15) shows two grains with near-zero He and FT ages, consistent with nearly full resetting of both systems, and one with old FT and He ages, consistent with minor or no resetting. Two grains from the soil sample in the San Joaquin drainage (04WFS13) also show one grain with old FT and He ages (but an FT age younger than the He age) and one with near-zero ages for both systems. Three grains from the exposed bedrock sample in the San Joaquin drainage (04WFS12) show near-zero FT ages and He ages of 25-55 Ma, suggesting a shallow trend of increasing FT age with increasing He age, consistent with predicted resetting trends (Figs. 3, 8).

In the Cascades samples (Fig. 16B), two of the grains from the high-order (Icicle Creek) channel in the Cascades show FT ages older than He ages (at ages consistent with regional expectations based on bedrock samples in the basin), two of them have both FT and He ages close to zero, and two of them have near-zero FT ages and He ages of ~10 and 23 Ma (Fig. 16A). The low-order channel grains show near-zero FT ages with He ages of 0.6 and 28 Ma (Fig. 16A). These combined ages for both the high- and low-order channel Cascades samples are consistent with expectations for both unheated, slowly-cooled grains, or wildfire-reset grains (and inconsistent with volcanic airfall) from this area, as observed for many of the intermediate-depth and detrital samples from the 6-cm bedrock transect (Fig. 8)

The soil samples from the Cascades (Fig. 16B, 04WFC4, 5, and 6) also yield many grains with near-zero FT ages and a wide range of He ages that are younger than the expected age of bedrock in this area (20-30 Ma), as predicted for resetting trends (Figs. 3, 8). Notably, however, several apatite grains from each of these samples have old FT ages (35-50 Ma) and a wide range of He ages. A few of these are older than the 20-30 Ma expected for He ages in bedrock from this immediate area. Moreover, several have young He ages at old FT ages, which are not expected for either unheated bedrock in this area, or from predicted wildfire-resetting trajectories from the initial bedrock He and FT ages in the area (Fig. 8). Samples with apatite He ages only In order to 1) evaluate the extent to which routine (U-Th)/He dating procedures encounter anomalously young, wildfire-affected grains, 2) characterize the age populations of river samples, and, 3) in one case, characterize He ages of fresh bedrock, a subset of samples were analyzed by conventional (U-Th)/He methods (Table 4). Two grains from the 04WFC1 creek

14

sand sample in the Cascades (Fig. 10) showed ages consistent with fresh bedrock ages in the area. However among seven grains from three nearby soil samples (04WFC4, 5, and 6), one grain in each aliquot had anomalously young ages. Two grains from Sierra Nevada soil (04WFS11) also yielded one old grain consistent with regional fresh bedrock ages and one grain with an anomalously young age of 0.5 Ma. Two soil samples (04WFS12, 13) from the Adams Peak region (Fig. 13) showed one of three and two of four grains with anomalously young He ages. Six single grains from fresh bedrock in the Antelope Lake region (Fig. 13) yielded an average age of 66.8 Ma, with one standard deviation of 3.6 Ma (Table 4), slightly younger than the AFT age of 72.8 ± 4.5 Ma (1σ) (Table 2). Only one of fifty-eight grains dated by conventional (U-Th)/He methods from the Icicle Creek river sediment yielded an age much younger than found in bedrock samples from the drainage (Reiners et al., 2000; Reiners et al., 2003; Mitchell et al., 2003; Isaacson, 2005) (Fig. 17). Most of the grains (88%) in this detrital population have ages of 18-45 Ma, similar to most bedrock ages. Four grains yielded apatite He ages of 9-11 Ma, a distinguishable young age peak in the population, but fresh bedrock with this age has been found in the drainage and is likely common in the western part of it. Two grains gave He ages of 75 and 81 Ma, which are older than any found in bedrock in this area.

Forty-six apatite grains from sand of the San Joaquin river yielded ages between 54 and 81 Ma with an age population similar to, but less variable than, that of bedrock apatite He ages found in a large are of the western Sierra Nevada (House et al., 1998; House et al., 2001; Cecil et al., 2006) (Fig. 18). No grains with anomalously young ages were found in this sample.

Fifty-two apatite grains from sand of the Kings river yielded two grains with anomalously young ages and fifty with ages ranging from 27 to 74 Ma. This population extends to younger ages than samples from the San Joaquin or bedrock from farther north in the western Sierra, but ages nearly this young are found in bedrock of the southern Sierra, not far from the Kings drainage (Clark et al., 2005) (Fig. 18). Discussion Small-scale bedrock transect Apatite FT and He ages in the small-scale bedrock transect (Figs. 7, 8) show trends that agree with kinetic predictions of resetting for these two systems. Nearly all the grains in the outermost centimeter have lost all their fission tracks, and are essentially completely reset with respect to the AFT system. In contrast, He ages in this sample are reset by only 10-60%. In the intermediate depth (1-3 cm) and detrital chip samples, however, fission track annealing is highly variable; approximately half of the grains being completely reset and the other half extending to ages as old as that of interior bedrock. The He ages in the 1-3-cm depth and detrital chip samples are much more reproducible, though they are partially reset by up to 15%. No resetting by wildfire for either the AFT or apatite He systems is evident in rock greater than 3 cm from the surface. These data from a well-controlled sample suite lead to several points that guide expectations and interpretations of other samples. First, the observed resetting patterns follow kinetic predictions in that FT ages are commonly younger than He ages, at least in the outer few centimeters of exposed bedrock. Second, where thermal histories have led to partial resetting, as in the 1-2 and 2-3 cm depth samples, FT ages vary widely whereas He ages show restricted

15

ranges. Third, detrital rock chips can contain significant proportions (~50%) of grains with zero FT ages and finite He ages. Fourth, if the youngest AFT age peak (P1) of a grain age population is close to zero and clearly distinguishable from an older peak representing fresh bedrock age, then the proportion of grains comprising the P1 peak represents the proportion of grains fully reset by wildfire (Fig. 7). At least in the case of the samples shown in Figures 7 and 8, the scatter in apatite FT and He ages in the intermediate depth samples precludes robust conclusions about the number or intensity of wildfire events affecting the exposed bedrock. It may be that centimeter-scale sampling mixes grains over depth ranges too large to allow modeling of potential thermal histories. However, it is possible to place some constraint on the thermal dose in the outermost centimeter. Here the full resetting of AFT ages combined with a minimum apatite He resetting of about 35% requires temperatures greater than about 300 °C for durations less than about 30 hours (Fig. 8). This is likely a cumulative effect from several fires, rather than a single event, but this cannot be determined without additional constraints. Detrital samples As long as the FT age of unreset apatite grains (i.e., those from interior bedrock) is clearly distinguishable from zero, as it is in the cases shown here, then the fraction of grains in the near-zero FT age peak approximates the proportion of grains with AFT ages completely reset by wildfire. As shown in Figures 8 and 16, most of these grains are likely to have partially reset (U-Th)/He ages, and a few will also have completely reset He ages. Figure 19 shows the proportion of detrital apatite grains with near-zero, wildfire-reset FT ages, as a function of sample type, location, and the mean gradient and area of the basin in which it was collected. On average, soils contain the largest proportion of reset grains (50-90%), with decreasing proportions in colluvial hollow deposits (30-70%, excluding Fort Sage), followed by low-order channel sands (2-40%). In each location, sediments from large basins contain the fewest reset apatite grains (Figs. 15, 19).

The Fort Sage samples contain much lower proportions of reset grains (1-16%) than other colluvial hollow samples. This region is distinguished from all the others in this study by its aridity. Its mean annual precipitation is only about 13-40% of that in the other regions, and the vegetation is semi-desert scrub with sparse sagebrush. Although the region experiences some fire (abundant burn scars were found at low elevations in 2005), the scarcity of reset grains in this region is consistent with a strong control of vegetation on at least the intensity, if not frequency, of wildfire heating of surface detritus. If vegetation in the Fort Sage area was more abundant in the cooler, potentially wetter, climate of the Pleistocene, it was either not sufficient to have generated wildfire resetting signatures comparable to the other areas, or detrital apatite grains do not persist in surficial deposits long enough to record this history. Near-zero FT age components The high proportion of reset grains with zero or near-zero ages in many of these samples raises the question of why the near zero-age component of detrital AFT populations has not been widely recognized previously. One reason may be that detrital grains with zero or near zero ages (or lag times, in sedimentary samples) derived from tectonically or magmatically active regions could be interpreted as contributions from either volcanic or rapidly exhuming sources (e.g., Sachsenhofer et al., 1998; Issler et al., 1999; Kelley, 2002; Coutand et al., 2006). This study shows that young (and potentially syn-depositional) age peaks can represent surficial wildfire

16

resetting, rather than timing of monotonic exhumational or volcanic cooling. Another possible reason, at least in cases where only 2-4% of grains are reset and a single older fraction with a precise age dominates, is that the chi-squared probability statistic can pass at the 5% level (e.g. Table 2), leading to an “acceptably low” age dispersion for a single population. Finally, it is also possible that some detrital apatite samples truly lack any wildfire-reset grains. This may be because their source region simply does not experience wildfires, such as high alpine glacial regions or extreme deserts. A more likely reason, however, is that most detrital AFT studies have focused on fluvial samples from high-order channels, which contain the fewest wildfire-reset grains. As this work shows, even in areas where wildfire-reset grains are abundant in soil, colluvium, and low-order channels, the high-order channels may have few to no wildfire-reset grains. This raises the question of what controls the change in apatite populations between hillslopes and high-order channels. Glaciation in high-order channels Before proceeding, we acknowledge that our high channel network order sediment samples originated in catchments that were glaciated to varying extents in the Pleistocene. The distribution and abundance of wildfire-reset grains in these areas may be influenced by the legacy of glaciation, because glaciation may 1) strip away preexisting regolith containing wildfire signatures, 2) reduce or prevent burning during or immediately after ice coverage, 3) leave fresh till in valleys for later removal, and 4) alter the shape of valleys in such a way (e.g., by forming oversteepened walls) as to reduce vegetative fuel. Although we cannot confidently rule out an important role for any of these factors, all of which would be expected to reduce the abundance of wildfire-reset grains in high-order channels, we suggest that they likely play a minor role for the following reasons. First, in both unglaciated and previously glaciated settings in the northeastern Sierra Nevada, there is a consistent decrease in the abundance of wildfire signatures in apatite grains from soils, to colluvial hollows, to low-order channels. The formerly glaciated catchments extend this trend to high-order channels as we were unable to find high-order catchments unaffected by glaciation. This suggests that a common mechanism reduces the abundance of wildfire-resetting signatures from hillslopes to fluvial settings, and from low-order to higher-order channels. It also suggests that, if glaciers remove pre-glacial regolith, then most or all of the wildfire signatures, as well as dissolution features, in the formerly glaciated catchments, are Holocene in age. It is possible that the low abundance of wildfire-reset apatite in the Icicle Creek soils, relative to the northeastern Sierra soils (Fig. 19), is due to the relative youthfulness of the regolith in Icicle Creek that follows glaciation. But the consistent trends in fire signatures across sediment types in both glaciated and nonglaciated settings suggest that the 10-15 kyr of Holocene soil development and transport have generated sufficiently large wildfire signatures to reflect the paucity of wildfire-reset grains in high-order catchments as a general phenomenon. Paucity of wildfire-reset grains in fluvial samples The high abundance of wildfire-reset grains in soils, colluvial hollows, and, to a lesser degree, low-order channel deposits, contrasts with their near absence in sediment carried by larger rivers. Given that fluvial sediment is derived primarily from hillslopes, these results suggest that there is a process that fractionates apatite populations in detritus during its transfer from hillslopes to fluvial systems.

17

Detrital apatite grains in transmitted-light and SEM photomicrographs (Fig.s 9 and 20) show several features distinguishing them from crystals from fresh bedrock. The most obvious of these is c-axis-parallel grooves and pits that give the grains a striated or wrinkly appearance on c-axis parallel faces, and either a spiky or swiss-cheese-like appearance on c-axis-perpendicular faces. Similar features due to preferential c-axis parallel dissolution of apatite have previously been observed in both laboratory and natural settings, including soils, saprolite, and shallow fluvial sandstones (Gleadow and Lovering, 1974; Morton, 1984; Banfield and Eggleton, 1989; Bouch et al., 2002; Welch et al., 2002). Detrital grains in this study also commonly have fractures at high-angles (most often perpendicular) to the c-axis, with dark coloration, and decorations of short c-axis-parallel pits along them, both on the surface and within the grains (Fig.s 9 and 20). Preliminary SEM examination and x-ray mapping of these cracks shows that many are filled with Fe-Si-Al-rich material, probably Fe-hydroxides and clays. Although within a given hillslope sample there is no obvious correlation between the presence of dissolution or decorated-crack features and the likelihood of reset He or FT ages, these dissolution features and decorated cracks are far more abundant in hillslope samples than in river sands. Thus these features may in general be diagnostic of near surface residence and the potential for wildfire resetting. Laboratory dissolution rates for apatite in soil-like conditions are two to three orders of magnitude faster than those of most other common rock-forming minerals (Fig. 21; Kowalewski and Rimstidt, 2002). Although laboratory dissolution rates for all minerals are typically much higher than observed in field settings (White and Brantley, 1995; Drever and Clow, 1995), if these proportional rate differences are preserved in natural soils, apatite would be expected to completely disappear far faster than most silicates in transport-limited erosional regimes. Rapid apatite dissolution relative to at least some other silicates is also consistent with the dominance of apatite-derived Sr on 87Sr/86Sr of soil and stream waters (Blum et al., 2002). Apatite dissolution has also been shown to occur at very shallow depths in sediments and sedimentary rocks, including in floodplain deposits, where it may play a role in the development of secondary porosity (Morton et al., 1986; Bouch et al., 2002).

Dissolution of apatite in hillslope environments provides a likely explanation for why grains with wildfire resetting signatures are rare or absent in river sediments. Transport-limited parts of landscapes with gentle slopes are often soil mantled and erode primarily by bedrock-to-saprolite conversion and downslope soil creep. Apatite in these areas are likely to reside in the upper few centimeters of the surface—susceptible to wildfire heating—for long periods of time. Slow “stirring” of soil profiles (Heimsath et al., 2000; Roering et al., 2002) also results in multiple near-surface exposures and wildfires over time. Apatite is also more likely to dissolve in these environments, however, removing it from the pool of detrital minerals bound for the fluvial system. Small (<3 cm) rock fragments containing internally bound apatite may occasionally survive transport through such regimes, preserving apatite from dissolution, but these grains would still be susceptible to wildfire heating, as our analyses of detrital rock chips on hillslopes show (Fig.s 7, 8). Transporting apatite that is unaffected by both wildfire heating and dissolution requires transport of larger clasts or an alternative transport mechanism.

Weathering-limited erosional regimes, such as steep hillslopes (and perhaps bedrock channels), erode primarily by threshold processes like landslides, rockfalls, or other episodic events (e.g., Roering et al., 2001). These processes can produce deep erosion and transport large cobbles or boulders or thick sections of regolith (e.g., Stallard, 1995). Larger transport units contain apatite grains less likely to have resided within a few centimeters of the surface,

18

susceptible to both wildfire heating and dissolution. A very rough constraint on the minimum size of rock clasts that must be delivered from hillslopes to fluvial systems comes from estimating the volume fraction of material in the outermost 1-3 cm of clasts, (Fig. 22), where wildfire resetting is most likely to occur (e.g., Fig. 8). For spherical clasts ~20 cm in diameter, ~ 27% of the volume resides in the outer 1 cm, and 66% in the outer 3 cm. To prevent delivery of material of less than 10% of grains in the outer 1 cm or 3 cm of clasts requires clasts larger than 0.6 m and 1.8 m diameter, respectively. Once in the fluvial environment, these clasts are likely to disaggregate, liberating unreset apatite grains or small clasts containing them. Effects of near-surface weathering on AFT and (U-Th)/He ages Given apatite’s relatively high dissolution rate and observations, from this and other studies, of morphologic and internal features in hillslope apatite suggesting corrosion and precipitation of other phases in cracks and along margins of the grains, a logical question is what effect partial dissolution or other kinds of weathering-induced changes have on apatite FT or (U-Th)/He ages. Gleadow and Lovering (1974) found that partially corroded apatite in a saprolite (taken from several meters below the surface) appeared to have lost ~25% its U, and ~40% of its spontaneous fission tracks, effects they attributed to low-temperature, fluid-assisted annealing accompanying U-loss and corrosion. In this study, dated apatite grains with the most obvious evidence of corrosion are those from soils. Although many of these grains have combined FT and He ages consistent with predicted trends of wildfire resetting, it is difficult to completely rule out the effects on weathering on some of the FT ages. However, a few grains from the Cascades soils fall in a region of the FT-He age plot (old FT ages and young He ages) that is not easily reconciled with either wildfire-resetting or fresh bedrock ages from this area (Fig. 16B). At this point we have no good explanation for these He-FT double dates except to suggest that dissolution, overgrowths, and/or open-system behavior of parents and/or daughter products during soil residence may have compromised either thermochronometric system. We also note that some grains from soils/colluvial hollows (Fig. 9) appear to have zones of higher induced track densities (reflecting higher U concentrations) coinciding with the dark, throughgoing fractures. This may reflect high U concentrations in the Fe-hydroxides and clays formed along the cracks and edges of the grains. Implications for detrital studies, surficial mass transport, and near-surface apatite Although wildfire-reset apatite is common in hillslope and low-order channel sediment, its relative rarity in high-order channel sediment probably means that wildfire resetting is not a major concern for thermochronologic studies of clastic sediments and sedimentary rocks that represent accumulation of sediment from medium to large rivers. Nonetheless, this study shows that even in some high-order channels, as much as 10% of apatite grains may have strongly disturbed (completely reset for FT) thermochronologic ages. This should be considered as a source of age dispersion or as an alternative to volcanic or rapid exhumation age component peaks in some populations. It is possible that fluvial sediment in some environments may contain higher proportions of reset grains, particularly if local circumstances lead to minimal sediment delivery from high-slope regions and minimal dissolution in hillslopes. Also, some types of sediments and sedimentary rocks that formed by means other than accumulation of sediment from high-order fluvial channels, such as paleosols, are likely to contain larger proportions of wildfire affected grains.

19

Derivation of most fluvial apatite from steep hillslopes has implications for studies that combine detrital cooling age populations with catchment hypsometry to deduce spatial or temporal variations in erosion rates within or among catchments (Brewer et al., 2003; Ruhl and Hodges, 2005; Hodges, 2005; Stock et al., 2006; Huntington and Hodges, 2006; Vermeesch, 2007). If most apatite exhumed in low-slope parts of landscapes dissolves in soils, then fluvial apatite populations will be dominated by steep regions of catchments. This means that populations of cooling ages from fluvial sediments may be misinterpreted as representing higher erosion rates in high slope areas, or possibly as erosion rate changes with time. For example, we predict that fluvial cooling-age populations sampled from an area draining the Grizzly Dome region (Fig. 12) would show a disproportionately large population of apatite grains with ages representing the lower parts of the age-elevation relationship where slopes are high on the margins of the dome, even if long-term erosion rates were uniform everywhere. This could be tested by comparing detrital cooling age distributions with those predicted from hypsometrically weighted age-elevation relationships from bedrock samples in the catchment. This may also have implications beyond detrital thermochronology. It is well known that due to differential solubility and abrasion resistance of phases, fluvial detritus (and its dissolved solute complement) is mineralogically (and compositionally) fractionated relative to its bedrock source (e.g., Kowalewski and Rimstidt, 2003). We suggest that it is also spatially fractionated, in that the less chemically and physically resistant phases in fluvial detritus are derived preferentially from steep regions prone to transport via bedrock slope instability. The conclusion that most apatite in fluvial sediment comes from large clasts or landslides derived from high slope regions is similar in some ways to observations that cosmogenic nuclide concentrations in fluvial sediment depend on sediment grain size (Brown et al., 1995) and are biased by landslides (Niemi et al., 2006). Preferential transport of material from hillslopes to rivers via landslides delivers detritus that has not resided in the upper few meters or centimeters for as long as it would if overland flow or soil creep were the only transport mechanisms. Detritus shielded from cosmic rays will also be shielded from wildfire heating. Thus we predict that the magnitude of thermochronologic wildfire signatures in detrital apatite, if present, should scale with cosmogenic nuclide concentrations. For a given erosion rate, the highest abundance of cosmogenic nuclides and the largest fraction of wildfire-reset grains should be found in fluvial sediment from landscapes characterized by hillslopes of moderate slope—steep enough to transport small clasts that prevent apatite dissolution but allow for wildfire heating, but shallow enough to prevent deep (>3 m) landslides. Paleowildfire signatures and other potential applications of He-FT age inversions As noted above, sediments or sedimentary rocks that do not form by accumulation of high-order fluvial sediment but contain a terrestrial clastic component may contain significant proportions of wildfire-affected grains. Some types of deposits, such as paleosols, may preserve evidence of paleowildfire and its variations in the geologic record. Typically such evidence comes from either fossil charcoal or polycyclic aromatic hydrocarbons (e.g., Scott, 2000; Finkelstein et al., 2006), which may be preferentially transported or preserved in some environments. Extrapolation of the results of this study suggests that many apatite grains in paleosols, at least those formed in vegetated areas, and those not dissolved during soil residence, should have FT ages approximating the soil formation age and older He ages on the same grains. Variations in abundance of grains with this diagnostic signatures through paleosol sequences could potentially illuminate paleowildfire dynamics.

20

Surface or near-surface processes other than wildfire can also produce transient thermal histories that produce inverted FT-He ages in both apatite and zircon. Processes capable of producing such thermal histories, at least theoretically, include small scale magmatic events (e.g., country rocks adjacent to small dikes or lava flows), shock heating, shear heating along brittle faults, and, at least potentially, lightning strikes. Conclusions Contrasting activation energies of FT annealing and He diffusion lead to a kinetic crossover in the apatite (and zircon) FT and (U-Th)/He thermochronometric systems. Short duration, high temperature thermal events characteristic of wildfire, and a few other processes such as transient heating along brittle faults, resets the FT system to greater extents than the (U-Th)/He system in apatite, resulting in inverted He-FT age relationships. Apatite in a 6-cm depth profile of exposed bedrock, as well as detrital clasts at the ground surface, confirm these predictions, and suggest a characteristic depth of penetration of wildfire heating effects of ~3 cm. Detrital apatite in soil, colluvium, and creek sand from a wide variety of settings shows that crystals with zero FT ages are common and comprise a large fraction of, or dominate, hillslope apatite populations. Combined single-grain FT-He double-dating confirms that many grains from these settings have inverted FT-He age relationships, and many grains are completely reset for the FT system but not (U-Th)/He system. Whereas wildfire-reset grains are common in hillslope environments, they are rare to absent in fluvial sediments from high-order channels. This, combined with the common presence of characteristic dissolution pits in hillslope apatite grains, suggests that most apatite in high-order channels is derived from material shielded from wildfire heating, probably inside large clasts and landslide material derived from high slope areas. This in turn means that apatite in fluvial systems does not sample low-slope regions of landscapes in proportion to their areal abundance, even if erosion rate and apatite abundance in bedrock are uniform. Acknowledgments We thank Stefan Nicolescu for analytical assistance, and Sara Mitchell for collection of the small-scale bedrock transect samples. We appreciate constructive reviews by Greg Stock, Rich Ketcham, and Andrew Gleadow, as well as helpful discussions with Joel Blum, Mark Brandon, and Rich Ketcham. Acknowledgment is made to the Donors of the American Chemical Society Petroleum Research Fund for support of this research. This work was also supported by NSF grant EAR-0236965 to PWR.

21

Figure Captions Figure 1. Pseudo-Arrhenius plot for He diffusion (blue lines) and FT annealing (green, orange, red, grey lines) in apatite. Solid and dashed blue lines for He represent 1% and 99% He degassing for grains with diffusion domains with equivalent spherical sizes (a) of 100 and 50 μm, respectively. He diffusion parameters are from Farley (2000). FT contours are shown for various annealing models, and normalized track lengths of 0.93 and 0.55, corresponding to negligible annealing at room temperature and essentially complete annealing where track density (and therefore age) go to zero, respectively. The yellow shaded region represent the approximate time-tempeature combinations for an eqivalent square pulse heating event that lead to complete FT age resetting with only partial (U-Th)/He age resetting. Figure 2. Pseudo-Arrhenius plot for He diffusion (blue lines) and FT annealing (green, red, lines) in zircon. He diffusion parameters are from Reiners et al. (2004). Essential features of this figure are analogous to those in Figure 1. Figure 3. Model contours of constant time (black lines, in natural log seconds) and temperature (red lines, °C) for partially reset apatite FT and (U-Th)/He ages. The track-length-density relationship of Ketcham (2005) was used. Labelled points A and C represent post-event FT-He ages for thermal histories of long duration and low temperature (~3.1 yr, 140 °C) and short duration and high temperature (~5 minutes, 400 °C), respectively. Labelled point B represents 325 °C for 2.8 hours. Figure 4. Representative wildfire thermal histories for 0.5 cm soil depths in Meditteranean maquis fires (Molina and Llinares, 2001). Fourier indices (Dt/a2) and fractional resetting extents for apatite FT and (U-Th)/He systems, calculated for the three histories with solid lines, are shown next to each line (histories with dashed lines were not translated to Dt/a2 or resetting values). HeFTy (Ketcham, 2005) was used for FT modeling. The lowest temperature thermal history results in post-heating FT track lengths 12.7 ± 1.14 μm. Apatite He diffusion parameters are from Farley (2000), with a = 75 μm. Figure 5. Northern Sierra Nevada sample locations. More detailed maps are shown with results in Figures 11-14. Figure 6. Percent difference between alpha-ejection correction for a whole apatite crystal and a crystal with one (c-axis-perpendicular) side polished, as a function of crystal radius (c-axis-perpendicular half-width; denoted by text next to each line), and polishing depth, assuming polishing depth is greater than one alpha-stopping distance and less than crystal half-width. Crystals polished exactly halfway through have an alpha ejection correction the same as the whole crystal. Crystals polished more than 20 μm but less than halfway through have an alpha ejection correction (FT of Farley et al., 1996) that is between zero and ~4.5% larger than that of the whole crystal for these crystal sizes, depending on crystal size and polishing depth. See text for details.

22

Figure 7. Probability density and histogram plots of apatite FT ages of grains from samples taken from different depths from the surface of exposed bedrock. Apatite and zircon (U-Th)/He ages were reported in Mitchell and Reiners (2003); apatite He ages are also shown as a function of depth in central plot for context. Apatite in the deepest sample shows a unimodal AFT age of 42 Ma, whereas ages are strongly reset in the outermost sample and show a large range of partial resetting in the intermediate depth and detrital chips samples. Figure 8. A. Apatite FT and (U-Th)/He double dates on single grains from the same samples shown in Figure 6. Most of the oldest ages (found in the deepest sample) are consistent with those observed in single-method dated grains for both techniques. Shallower and detrital samples range widely to younger He and FT ages, but only samples with He ages greater than ~14 Ma have FT ages indistinguishable from zero. B. Comparison of model resetting trends with double-dated grains, assuming initial He and FT ages of 20 and 50 Ma, respectively. Model trends were created as in Figure 3, but using initial FT and He ages of 50 and 20 Ma, respectively. Figure 9. Representative photomicrographs of apatite grains from colluvial hollow sand samples with both reset and unreset FT ages, with and without c-axis-parallel dissolution pits and decorated fractures characteristic of detrital apatite in soils, colluvium, and low-order channel sand. Figure 10. Sample locations and probability density and histogram plots of apatite FT ages of grains from samples in the Icicle Creek catchment (outlined in red) of the central Washington Cascades. 40-61% of apatite grains in low-order channel sand and soil samples fall into an FT age peak that is indistinguishable from zero and far younger than the fresh-bedrock apatite FT age of 40-50 Ma for samples collected at the same elevation. In contrast, less than 9% of apatite grains in the high-order (river) sand sample (04WFC9A) are reset. A second analysis of the same sample (04WFC9B), but many fewer grains, identified only one reset grain. Figure 11. Sample locations and probability density and histogram plots of apatite FT ages of grains from samples a region just northwest of Antelope Lake in the northern Sierra Nevada [see text and Riebe et al. (2000; 2001a; 2001b; 2004) for more location details]. Fresh bedrock from this region shows a unimodal apatite FT age of 73 Ma. Soil and colluvial hollow samples in this region contain 46-92% wildfire-reset apatite grains, and the creek sand 29%. The red dashed line is the approximate limit of the 2001 Stream Fire (the fire occurred within the area to the east-northeast of the line). Figure 12. Sample locations and probability density and histogram plots of apatite FT ages of grains from samples at the top of Grizzly Dome and the bottom of it, from mouths of tributary creeks to the North Fork of the Feather River, in the northern Sierra Nevada [see text and Riebe et al. (2000; 2001a; 2001b; 2004) for more location details]. Fresh bedrock from this region shows a unimodal apatite FT age of 69 Ma. Colluvial hollow samples in this region contain 32 and 34% wildfire-reset apatite grains, and the creek sand samples 2 and 10%. Figure 13. Sample locations and probability density and histogram plots of apatite FT ages of grains from samples in the Adams Peak area of the northern Sierra Nevada [see text and Riebe et

23

al. (2000; 2001a; 2001b; 2004) for more location details]. Colluvial hollow samples in this region contain 40-65% wildfire-reset apatite grains. Figure 14. Sample locations and probability density and histogram plots of apatite FT ages of grains from samples in the Fort Sage area of the northern Sierra Nevada [see text and Granger et al. (2000) and Riebe et al. (2000; 2001a; 2001b; 2004) for more location details]. Three colluvial hollow samples in this region contain 1-2% wildfire-reset apatite grains, and one contains 16%. Figure 15. Sample locations and probability density and histogram plots of apatite FT ages of grains from samples in the San Joaquin drainage and directly from the San Joaquin and Kings rivers, central Sierra Nevada. Apatite (U-Th)/He ages of samples from the region within and around the San Joaquin drainage from House et al. (1998; 2001) are shown in the upper left panel. A 3-cm-thick slab of exposed bedrock shows essentially complete resetting of the apatite FT system. Shallow soil shows a wide range of resetting. High-order channel sand from the two rivers show zero reset grains and 3.2% reset grains; separate aliquots of the same samples with fewer grains show consistent results. The catchment areas of the San Joaquin and Kings rivers above the fluvial sand samples taken in this study are outlined in green and purple, respectively. Figure 16. Apatite FT and (U-Th)/He double dates on single grains from samples in the San Joaquin and Kings river drainages (Fig. 15) and Icicle Creek drainage (Fig. 10). A. San Joaquin river grains show ages consistent with no resetting, and consistent with independent He and FT apatite age populations from this area (Fig.s 15, 18). Exposed bedrock and soil show strongly reset and mixed reset signatures, and Kings river grains show both unreset and completely reset signatures. B. Apatite grains from both high-order (Icicle Creek) and low-order channels show either unreset He-FT ages corresponding to those of local interior bedrock, or zero FT and partially reset He ages characteristic of strong resetting. Some grains from the soil samples show age combinations similar to those of the channels, but each soil sample also contains some grains with old FT ages (30-60 Ma) and a wide range of He ages, including some significantly younger than the FT ages. These latter age-combinations are not expected from either unheated bedrock in this area, nor from predicted wildfire heating trajectories from the initial bedrock ages. Figure 17. Probability density and histogram plot of ages of 58 detrital apatite (U-Th)/He grains from Icicle Creek. Ages from ~9 to ~45 Ma are well represented in bedrock samples. It is not clear whether the two oldest grains are true ages or fliers explainable by errors in grain selection or analysis. The one grain with the youngest age is probably reset by wildfire, consistent with the ~9% reset population indicated by the FT results (Fig. 10). Figure 18. Probability density and histogram plot of ages of apatite (U-Th)/He ages of bedrock from the western Sierran Nevada (A., House et al., 1998; 2001; Cecil et al., 2006), 46 detrital grains from the San Joaquin river (B), and 58 detrital grains from the Kings river (C). San Joaquin river apatite grains show no wildfire resetting and similar He ages as bedrock. Kings river apatite grains range to younger He ages, consistent with regional bedrock data (Clark et al., 2005), and also show a small number of reset grains, consistent with the independent FT data on this sample (Fig. 10).

24

Figure 19. Summary diagram of wildfire resetting signatures in detrital apatite as a function of location, sample type, and basin gradient, as measured by percentage of grains with completely reset FT ages. Soil and colluvial hollows (with the exception of Ft. Sage) contain the largest proportion of reset grains, followed by low-order channels, and high-order channels. Figure 20. Photomicrographs showing dissolution features of apatite grains common in detrital samples. A. Transmitted light photograph showing grains with progressively greater extents of c-axis parallel dissolution. B. SEM image of apatite grain showing c-axis parallel dissolution pits and c-axis perpendicular fractures. High contrast regions are image artifacts caused by charging of the surfaces. Scale bars in each panel are 100 μm. Figure 21. Compiled laboratory measurements of dissolution rates of common rock-forming minerals. At pH ~6, apatite dissolution rates are about 100-1000 times faster than common silicate minerals. Anorthite is not typically present at endmember concentrations in natural settings. Also see Kowalewski and Rimstidt (2002) figure 1. Figure 22. Volume fraction of a clast comprising the outermost shells of the clast 1 cm, 1-2 cm, 2-3 cm, or 0-3 cm deep. To prevent less than 10% of material from clasts delievered to rivers from being in the outermost 3 cm requires clasts nearly 2 m in diameter.

25

References Agee, J.K. (1993), Fire ecology of Pacific Northwest forests: Washington, D.C., Island Press,

493 p. Banfield J.F., and R.A. Eggleton (1989), Apatite replacement and rare earth mobilization,

fractionation and fixation during weathering, Clays Clay Miner., 37,113–127. Bernet, M., and J.I. Garver (2005), Fission-track analysis of detrital zircon, in Low-Temperature

Thermochronology: Techniques, Interpretations, and Applications, Rev. Mineral. Geochem., vol. 58, edited by P.W. Reiners and T.A. Ehlers, pp. 205-238, MSA, Chantilly, VA.

Blum, J.D., A. Klaue, C.A. Nezat, C.T. Driscoll, C.E. Johnson, T.G. Siccama, C. Eagar, T.J. Fahey, and G.E. Likens (2002), Mycorrhizal weathering of apatite as an important calcium source in base-poor forest ecosystems, Nature, 417, 729-731.

Blum, A.E., and L.L. Stillings (1995), Feldspar dissolution kinetics, in Chemical weathering rates of silicate minerals, Rev. Mineral. Geochem., vol. 31, edited by A.F. White and S.L. Brantley, pp. 291-351, MSA, Chantilly, VA.

Bouch, J.E., J.H. Malcolm, N.H. Trewin, S. Chenery, and A.C. Morton (1984), Authigenic apatite in a fluvial sandstone sequence: Evidence for rare-earth element mobility during diagenesis and a tool for diagenetic correlation, J. Sed. Res. 72, 59-67.

Brandon, M.T. (1992), Decomposition of fission-track grain-age distributions, Am. J. Sci. 292, 535-564

Brandon, M.T., M.K. Roden-Tice, and J.I. Garver (1998), Late Cenozoic exhumation of the Cascadia accretionary wedge in the Olympic Mountains, northwest Washington State. Geol. Soc. Am. Bull., 110, 985-1009.

Brantley, S. L., and Y. Chen (1995), Chemical weathering rates of pyroxenes and amphiboles, in Chemical weathering rates of silicate minerals, Rev. Mineral. Geochem., vol. 31, edited by A.F. White and S.L. Brantley, pp. 119-172, MSA, Chantilly, VA.

Brewer, I.D., Burbank D.W., and Hodges K.V. (2003), Modelling detrital cooling-age populations: insights from two Himalayan catchments, Basin Res., 15, 305-320.

Brown, E.T., R.F. Stallard, M.C. Larsen, G.M. Raisbeck, and F. Yiou (1995), Denudation rates determined from the accumulation of in situ-produced 10Be in the Luquillo Experimental Forest, Puerto Rico, Earth Planet. Sci. Lett., 129, 193-202.

Carlson, W.D. (1990), Mechanisms and kinetics of apatite fission-track annealing, Am. Mineral., 75, 1120–1139.

Carter, A. (1999), Present status and future avenues of source region discrimination and characterization using fission-track analysis, Sed. Geol., 124, 31-45.

Cecil, M.R., M.N. Ducea, P.W. Reiners, and C.G. Chase (2006), Cenozoic exhumation of the northern Sierra Nevada, California, from (U-Th)/He thermochronology, Geol. Soc. Am. Bull., 118, 1481-1488.

Clark, M.K., G. Maheo, J. Saleeby, and K.A. Farley (2005), The non-equilibrium landscape of the southern Sierra Nevada, California, GSA Today, 15, doi:10:1130/1052-5173(2005)015<4:TNELOT>2.0.CO;2.

Coutand, I., B. Carrapa, A. Deeken, A.K. Schmitt, E.R. Sobel, M.R. Strecker (2006), Propagation of orographic barriers along an active range front: insights from sandstone petrography and detrital apatite fission-track thermochronology in the intramontane

26

Angastaco basin, NW Argentina, Basin Res. 18, 1–26, doi:10.1111/j.1365-2117.2006.00283.x

Crank J. (1975), The Mathematics of Diffusion. Oxford University Press, 414 pp. Crowley K.D., M. Cameron, and R.L. Schaefer (1991), Experimental studies of annealing etched

fission tracks in fluorapatite, Geochim. Cosmochim. Acta, 55, 1449-1465. Donelick, R.A., P.B. O’Sullivan, and R.A. Ketcham (2005), Apatite fission-track analysis, in

Low-Temperature Thermochronology: Techniques, Interpretations, and Applications, Rev. Mineral. Geochem., vol. 58, edited by P.W. Reiners and T.A. Ehlers, pp. 49-94, MSA, Chantilly, VA.

Everett, R.L., R. Schellhaas, D. Keenum, D. Spurbeck, and P. Ohlson (2000), Fire history in the ponderosa pine/Douglas-fir forests on the east slope of the Washington Cascades, For. Ecol. Management. 129, 207–225.

Farley K.A. (2000), Helium diffusion from apatite: General behavior as illustrated by Durango fluorapatite, J. Geophys. Res. 105, 2903-2914.

Farley K.A. (2002), (U-Th)/He dating: Techniques, calibrations, and applications, in Noble Gases, Rev. Mineral. Geochem., vol. 47, edited by D.P. Porcelli, C.J. Ballentine, and R. Wieler, pp. 819-844, MSA, Chantilly, VA.

Farley K.A., and D.F. Stockli (2002), (U-Th)/He Dating of Phosphates: Apatite, Monazite, and Xenotime, in Phosphates, Rev. Mineral. Geochem., vol. 48, edited by M.L. Kohn, J. Rakovan, and J.M. Hughes, pp. 559-577, MSA, Chantilly, VA.

Farley K.A., R.A. Wolf, and L.T. Silver (1996), The effects of long alpha-stopping distances on (U-Th)/He ages, Geochim. Cosmochim. Acta, 60, 4223-4229.

Finkelstein, D.B., L.M. Pratt, and S.C. Brassell (2006), Can biomass burning produce a globally significant carbon-isotope excursion in the sedimentary record?, Earth Planet. Sci. Lett., 250, 501-510.

Galbraith, R.F. (2005), Statistics for Fission Track Analysis, Chapman & Hall/CRC, Boca Raton, 219 pp.

Gallagher, K., R., Brown, and C. Johnson (1998), Fission track analysis and its applications to geological problems, Ann. Rev. Earth Planet. Sci., 26, 519-572.

Garver, J.I., M.T. Brandon, M.K. Roden-Tice, and P.J.J. Kamp (1999), Exhumation history of orogenic highlands determined by detrital fission track thermochronology, in Exhumation processes: Normal faulitng ductile flow, and erosion, edited by U. Ring, M.T. Brandon, S.D. Willett, and G.S. Lister, G.S, Geological Society [London] Special Publication 154, pp. 283-304.

Gleadow, A.J.W., and J.F. Lovering (1974), The effect of weathering on fission track dating, Earth Planet. Sci. Lett. 22, 163-168.

Granger, D.E., C.S. Riebe, J.W. Kirchner, and R.C. Finkel (2001), Modulation of erosion on steep granitic slopes by boulder armoring, as revealed by cosmogenic 26Al and 10Be, Earth Planet. Sci. Lett., 186, 269-281.

Green, P.F. (1988), The relationship between track shortening and fission track age reduction in apatite: Combined influences of inherent instability, annealing anisotropy, length bias and system calibration, Earth Planet. Sci. Lett. 89, 335-352.

Heimsath, A.M., J. Chappell, N.A. Spooner, and D.G. Questiaux, D.G. (2002), Creeping soil, Geology, 30, 111-114.

Hodges, K.V., K.W. Ruhl, C.W. Wobus, and M.S. Pringle (2005), Apatite fission-track analysis, in Low-Temperature Thermochronology: Techniques, Interpretations, and Applications,

27

Rev. Mineral. Geochem., vol. 58, edited by P.W. Reiners and T.A. Ehlers, pp. 239-257, MSA, Chantilly, VA.

Hourigan, J.K., P.W. Reiners, M.T. Brandon (2005), U-Th zonation-dependent alpha-ejection in (U-Th)/He chronometry, Geochim. Cosmochim. Acta, 69, 3349-3365.

House, M.A., B.P. Wernicke, and K.A. Farley (1998), Dating topography of the Sierra Nevada, California, using apatite (U-Th)/He ages, Nature, 396, 66-69.

House M.A., K.A. Farley, and D. Stockli D (2000), Helium chronometry of apatite and titanite using Nd-YAG laser heating, Earth Planet. Sci. Lett. 183, 365-368.

House, M.A., B.P. Wernicke, and K.A. Farley, (2001), Paleo-geomorphology of the Sierra Nevada, California, from the (U-Th)/He ages in apatite, Am. J. Sci., 301, 77-102.

Huntington, K.W., and K.V. Hodges, (2006), A comparative study of detrital mineral and bedrock age-elevation methods for estimating erosion rates, J. Geophys. Res. 111, F03011, doi:10.1029/2005JF000454

Hurford, A.J., and A. Carter, (1991), The role of fission track dating in discrimination of provenance, in Developments in sedimentary provenance studies, edited by A.C. Morton, S.P. Todd, and P.D.W. Haughton, pp. 67-78, Geological Society [London] Special Publication vol. 57.

Hurford, A.J. and P.F. Green (1983). The Zeta-Age Calibration of Fission-Track Dating, Iso. Geosci., 1, 285-317.

Isaacson, P.J. (2005), Relief change and erosion history of the Icicle Creek area, Washington Cascades, using (U-Th)/He thermochronometry, unpublished BS thesis, Yale University, 123 pp.

Issler, D.R., S.D. Willett, C. Beaumont, R.A. Donelick, and A.M. Grist (1999), Paleotemperature history of two transects across the Western Canada Sedimentary Basin: Constraints from apatite fission track analysis, Bull. Canad. Petrol. Geol. 47, 475-486.

Kelley, S.A. (2002), Unroofing of the southern Front Range, Colorado; a view from the Denver Basin, Rocky Mount. Geol. 37, 189-200.

Ketcham, R.A. (2005), Forward and inverse modeling of low-temperature thermochronometry data, in Low-Temperature Thermochronology: Techniques, Interpretations, and Applications, Rev. Mineral. Geochem., vol. 58, edited by P.W. Reiners and T.A. Ehlers, pp. 275-314, MSA, Chantilly, VA.

Ketcham RA, R.A. Donelick, and W.D. Carlson (1999), Variability of apatite fission-track annealing kinetics: III. Extrapolation to geological time scales. Am. Mineral., 84, 1235-1255.

Kowalewski, M., and J.D. Rimstidt (2003), Average lifetime and age spectra of detrital grains: Toward a unifying theory of sedimentary particles, J. Geol., 111, 427-439.

Laslett G.M., P.F. Green, I.R. Duddy, and A.J.W. Gleadow (1987), Thermal annealing of fission tracks in apatite, 2. A quantitative analysis, Chem. Geol. (Isot Geosci Sect), 65, 1-13.

Lorang, M.S. and G. Aggett (2005), Potential sedimentation impacts related to dam removal: Icicle Creek, Washington, USA, Geomorph., 71, 182-201.

McDougall I, and T.M. Harrison (1999), Geochronology and Thermochronology by the 40Ar/39Ar Method. Oxford: Oxford Univ. Press. 269 pp. 2nd ed.

Meesters A.G.C.A., and T.J. Dunai (2002)l, Solving the production-diffusion equation for finite diffusion domains of various shapes; Part I. Implications for low-temperature (U-Th)/He thermochronology, Chem. Geol., 186, 333-44.

28

Mitchell S.G., and P.W. Reiners (2003), Influence of wildfires on apatite and zircon (U-Th)/He ages, Geology, 31,1025-1028.

Molina, M.J., and J.V. Llinares (2001), Temperature-time curves at the soil surface in maquis summer fires, Int. J. Wildland Fire, 10, 45–52.

Morton, A.C. (1986), Dissolution of apatite in North Sea Jurassic sandstones: Implications for the generation of secondary porosity, Clay Minerals 21, 711-733.

Nagy, K.L. (1995), Dissolution and precipitation kinetics of sheet silicates, in Chemical weathering rates of silicate minerals, Rev. Mineral. Geochem., vol. 31, edited by A.F. White and S.L. Brantley, pp. 173-233, MSA, Chantilly, VA.

Niemi, N.A., M. Oskin, D.W. Burbank, A.M. Heimsath, and E. Gabet (2005), Effects of bedrock landslides on cosmogenically determined erosion rates, Earth Planet. Sci. Lett., 237, 480-498.

Rahn, M.K., M.T. Brandon, G.E. Batt, and J.I. Garver (2004), A zero-damage model for fission-track annealing in zircon, Am. Mineral., 89, 473-484.

Reiners, P.W. (2007), Thermochronologic approaches to paleotopography, in Paleoaltimetry: Geochemical and Thermodynamic Approaches, Rev. Mineral. Geochem., vol. 6X, edited by M.J. Kohn, MSA, Chantilly, VA, in press.

Reiners, P.W., R. Brady, K.A. Farley, J.E. Fryxell, B.P. Wernicke, and D. Lux (2000), Helium and argon thermochronometry of the Gold Butte block, South Virgin Mountains, Nevada, Earth Planet. Sci. Lett., 178, 315-326.

Reiners, P.W., and M.T. Brandon (2006), Using thermochronology to understand orogenic erosion, Ann. Rev. Earth Planet. Sci., 34, 419-166.

Reiners, P.W., T.A. Ehlers, J.I. Garver, S.G. Mitchell, D.R. Montgomery, J.A. Vance, and S. Nicolescu (2002), Late Miocene exhumation and uplift of the Washington Cascade Range, Geology, 30, 767–770.

Reiners P.W., Z. Zhou, T.A. Ehlers, C. Xu, M.T. Brandon, R.A.Donelick, and S. Nicolescu (2003), Post-orogenic evolution of the Dabie Shan, eastern China, from (U-Th)/He and fission-track dating, Am. J. Sci., 303, 489-518.

Reiners, P.W., T.L. Spell, S. Nicolescu, and K.A. Zanetti (2004), Zircon (U-Th)/He thermochronometry: He diffusion and comparisons with 40Ar/39Ar dating, Geochim. Cosmochim. Acta, 68, 1857-1887.

Riebe, C.S., J.W. Kircher, and R.C. Finkel (2004), Erosional and climatic effects on long-term chemical weathering rates in granitic landscapes spanning diverse climate regimes, Earth Planet. Sci. Lett., 224, 547-562.

Riebe, C.S., J.W. Kirchner, D.E. Granger, and R.C Finkel (2000), Erosional equilibrium and disequilibrium in the Sierra Nevada, inferred from cosmogenic 26Al and 10Be in alluvial sediment, Geology, 28, 803-806.

Riebe, C.S., J.W. Kirchner, D.E. Granger, and R.C Finkel (2001a), Minimal climatic control on erosion rates in the Sierra Nevada, California, Geology, 29, 447-450.

Riebe, C.S., J.W. Kirchner, D.E. Granger, and R.C Finkel (2001b), Strong tectonic and weak climatic control of long-term chemical weathering rates, Geology, 29, 511-514.

Roering, J., P. Almond, J. McKean, and P. Tonkin (2002), Soil transport driven by biological processes over millennial time scales, Geology, 30, 1115-1118.

Roering, J.J., J.W. Kirchner, L.S. Sklar, and W.E. Dietrich (2001), Hillslope evolution by nonlinear creep and landsliding: An experimental study, Geology, 29, 143-146.

29

Ruhl, K.W., and K.V. Hodges (2005), The use of detrital mineral cooling ages to evaluate steady state assumptions in active orogens: An example from the central Nepalese Himalaya, Tectonics, 24, TC4015, doi:10.1029/2004TC001712, 2005.

Sachsenhofer, R.F., I. Dunkl, C. Hasenhüttl, and B. Jelen (1998), Miocene thermal history of the southwestern margin of the Styrian Basin: vitrinite reflectance and fission-track data from the Pohorje/Kozjak area (Slovenia), Tectonophys., 97, 17-29.

Scott, A.C. (2000), The Pre-Quaternary history of fire, Palaeogeog. Palaeoclim. Palaeoecol., 164, 281-329.

Shuster, D.L., R.M. Flowers, and K.A. Farley, (2006), The influence of natural radiation damage on helium diffusion kinetics in apatite, Earth Planet. Sci. Lett., 249, 48-161.

Stock, G.M., T.A. Ehlers, and K.A. Farley (2006), Where does sediment come from? Quantifying catchment erosion with detrital apatite (U-Th)/He thermochronometry, Geology, 4, 725-729, DOI: 10.1130/G22592.1.

Stock, J.D., and D.R. Montgomery (1996), Estimating paleorelief from detrital mineral age ranges, Basin Res., 8, 317-327.

Tagami T, R.F. Galbraith, R.Yamada, and G.M. Laslett (1998), Revised annealing kinetics of fission tracks in zircon and geologic implications, in Advances in Fission-Track Geochronology, edited by P Van den Haute and F De Corte, pp. 99-112. Dordrecht: Kluwer Academic Publishers.

Thomson, S.N. (1994), Fission-track analysis and provenance studies in Calabrian Arc sedimentary rocks, southern Italy, J. Geol. Soc. Lond., 151, 463-471.

Thomson, S.N. and U. Ring, U. (2006), Thermochronologic evaluation of postcollision extension in the Anatolide orogen, western Turkey, Tectonics, 25, TC3005, doi: 10.1029/2005TC001833.

Valsami-Jones, E., K.V. Ragnarsdottir, A. Putnis, D. Bosbach, A.J. Kemp, and G. Cressey (1998), The dissolution of apatite in the presence of aqueous metal cations at pH 2-7, Chem. Geol., 151, 215-233.

Vermeesch, P. (2007), Quantitative geomorphology of the White Mountains (California), using detrital apatite fission track thermochronology, J. Geophys. Res., doi:10.1029/2006JF000671, in press.

Welch, S.A., A.E. Taunton, and J.F. Banfield (2002), Effects of microorganisms and microbial metabolites on apatite dissolution, Geomicro. J., 19, 343-367.

Wolf R.A., K.A. Farley, and D.M. Kass (1998), Modeling of the temperature sensitivity of the apatite (U-Th)/He thermochronometer, Chem. Geol., 148, 105-114.

30

Table 1. Sample Locations and Notes. Sample name Region Subregion Lat. (Deg.

N, DM) Long. (Deg. W, DM)

Elev. (m)

Sample type Notes Basin area (km2)

Mean gradient

SGM01 Cascades Icicle Crk. 47°34.540 120°47.599 680 bedrock 0-1 cm depth into rock surface na na SGM12 Cascades Icicle Crk. 47°34.540 120°47.599 680 bedrock 1-2 cm depth into rock surface na na SGM23 Cascades Icicle Crk. 47°34.540 120°47.599 680 bedrock 2-3 cm depth into rock surface na na SGM3+ Cascades Icicle Crk. 47°34.540 120°47.599 680 bedrock 3-5 cm depth into rock surface na na SGM chips Cascades Icicle Crk. 47°34.078 120°47.190 658 rock chips on top of soil on broad slope na nd 04WFC1 Cascades Icicle Crk. 47°35.265 120°47.201 1234 low-order chan. steep, high creek channel 0.11 0.71 04WFC4 Cascades Icicle Crk. 47°35.144 120°47.283 1182 hillslope soil 0-3 cm depth on broad slope 0.0002 0.76 04WFC5 Cascades Icicle Crk. 47°35.144 120°47.283 1182 hillslope soil 3-5 cm depth on broad slope 0.0002 0.76 04WFC6 Cascades Icicle Crk. 47°35.144 120°47.283 1182 hillslope soil 5-10 cm depth on broad slope 0.0002 0.76 04WFC9 Cascades Icicle Crk. 47°33.592 120°40.288 316 high-order chan. Icicle Creek, near Leavenworth 547 0.51 04WFS11 Sierra Kaiser 37°15 119°14 ~2900 low-order chan. creek just below Nellie lake nd nd 04WFS12 Sierra Kaiser 37°15 119°14 ~2400 bedrock Between Nellie and Huntington Lakes na na 04WFS13 Sierra Kaiser 37°15 119°14 ~2400 level soil Between Nellie and Huntington Lakes nd nd 04WFS14 Sierra San Joaquin 37°08.923 119°30.203 302 high-order chan. San Joaquin River 3750 0.16 04WFS15 Sierra Kings river 36°53.247 119°09.446 291 high-order chan. Kings River 3480 0.21 05WFS1 Sierra Antelope Lk. 40°11.005 120°38.233 1750 hillslope soil 0-3 cm depth; broad swale 0.012 0.15 05WFS2 Sierra Antelope Lk. 40°11.005 120°38.233 1750 hillslope soil 3-6 cm depth; broad swale 0.012 0.15 05WFS3 Sierra Antelope Lk. 40°10.738 120°38.161 1710 colluvial hollow 0-5 cm depth; steep sides to channel 0.087 0.36 05WFS4 Sierra Antelope Lk. 40°10.703 120°38.210 1700 colluvial hollow 0-5 cm depth; steep sides to channel 0.023 0.36 05WFS5 Sierra Antelope Lk. 40°10.676 120°38.247 ~1710 fresh bedrock freshly blasted roadcut material nd 05WFS6 Sierra Antelope Lk. 40°10.046 120°39.496 ~1800 low-order chan. coarse sand to gravel 1.13 0.18 05WFS7 Sierra Fort Sage 40°05.843 120°03.921 1300 colluvial hollow 0-5 cm depth; base of hill 0.63 0.31 05WFS8 Sierra Fort Sage 40°05.665 120°04.078 ~1394 colluvial hollow 0-5 cm depth; small low-relief channel 0.036 0.35 05WFS9 Sierra Fort Sage 40°05.454 120°03.744 1450 colluvial hollow 0-5 cm depth; low-relief channel 0.0075 0.24 05WFS10 Sierra Fort Sage 40°05.659 120°03.601 1350 colluvial hollow 0-5 cm depth; base of steep slope 0.23 0.35 05WFS11 Sierra Adams Peak 39°53.848 120°06.883 2053 colluvial hollow 0-5 cm depth; broad swale 0.094 0.44 05WFS12 Sierra Adams Peak 39°53.411 120°06.975 1986 colluvial hollow 0-5 cm depth; weak channel 0.079 0.44 05WFS13 Sierra Adams Peak 39°53.214 120°07.026 1943 colluvial hollow 0-5 cm depth 0.17 0.49 05WFS14 Sierra Fall River 39°53.235 121°21.754 535 low-order chan. pool at base of waterfall on steep slope 0.91 0.64 05WFS15 Sierra Fall River 39°53.307 121°21.696 ~535 fresh bedrock freshly blasted roadcut material na na 05WFS16 Sierra Fall River 39°53.356 121°21.610 ~535 low-order chan. channel at base of steep slope 0.96 0.66 05WFS17 Sierra Grizzly Dome 39°53.080 121°19.576 1418 colluvial hollow 0-5 cm depth; broad swale 0.38 0.25 05WFS18 Sierra Grizzly Dome 39°52.938 121°19.266 1486 bedrock fresh interior bedrock na na 05WFS19 Sierra Grizzly Dome 39°53.569 121°18.434 1500 colluvial hollow 0-5 cm depth; weak channel 0.042 0.41 Notes: na = not applicable; nd = no data.

31

Table 2: Summary of apatite fission track data Sample No.

# grains

Track Density (x 106 tr cm-2) Age Disper-

sion

Central Age (Ma)

BinomFit Best Fit Peaks Grain Point Counts (10 fields of view)

ρs (Ns)

ρi (Ni)

ρd (Nd)

(Pχ2) (±1σ) P1 age

P1 error (95% CI)

P1 %

P2 age

P2 error (95% CI)

P2 %

Tracks No Tracks

% Zero Tracks

05WFS1 50 0.1645 (237)

3.591 (5173)

1.164 (3632)

198% (0%)

7.0±2.1 3.5

-0.7 +0.5

92.0% 63.7 -11.6 +14.1

8.0% 95 72 76%

05WFS2 50 0.2069 (256)

3.148 (3884)

1.153 (3599)

200% (0%)

12.5±3.7 1.6

-0.5 +0.8

82.0% 79.1 -12.7 +15.2

18.0% 99 82 83%

05WFS3 50 0.5204 (469)

3.016 (2716)

1.143 (3566)

104% (0%)

31.3±5.1 0.8 -0.5 +1.1

45.9% 66.1 -8.5 +9.7

54.1% 103 25 24%

05WFS4 50 0.3199 (447)

3.691 (2239)

1.132 (3533)

161% (0%)

16.4±3.9 0.5 -0.2 +0.4

69.8% 65.3 -8.5 +9.8

30.2% 53 26 49%

05WFS5 20 1.944 (830)

5.308 (2266)

1.122 (3500)

1.8% (>99%)

72.8±4.5 72.8 -8.2 +9.2

100% - - - 84 0 0%

05WFS6 50 1.502 (752)

6.497 (3252)

1.111 (3467)

71% (0%)

38.9±4.7 1.8 -0.9 +2.0

28.5% 60.7 -6.9 +7.8

71.5% 124 18 15%

05WFS7 50 1.162 (638)

3.705 (2035)

1.100 (3434)

0.13% (90.8%)

61.1±4.0 0.0 - 2.1% 62.0 -7.3 +8.3

97.9% 150 5 3%

05WFS8 50 0.8804 (539)

3.324 (2035)

1.079 (3366)

43% (0%)

46.9±4.5 0.0 - 15.8% 59.4 -7.3 +8.3

84.2% 120 3 3%

05WFS9 50 0.8242 (607)

2.762 (2034)

1.067 (3330)

12.3% (43.8%)

56.1±3.8 3.1 -2.7 +19.4

2.0% 58.1 -6.9 +7.9

98.0% 69 4 6%

05WFS10 50 1.152 (474)

4.102 (1687)

1.056 (3295)

0.3% (95.9%)

52.6±3.7 0.0 - 1.1% 52.9 -6.7 +7.7

98.9% 329 0 0%

05WFS11 50 0.5013 (413)

3.092 (2547)

1.045 (3259)

90% (0%)

27.8±4.1 1.2 -0.8 +2.6

41.4% 51.1 -7.1 +8.3

58.6% 111 15 14%

05WFS12 50 0.3179 (268)

3.154 (2659)

1.033 (3224)

135% (0%)

16.3±3.4 1.4 -0.6 +1.1

65.1% 52.1 -8.0 +9.5

34.9% 100 15 15%

05WFS13 50 0.5189 (683)

2.737 (3602)

1.022 (3189)

91% (0%)

31.6±4.6 0.5 -0.3 +0.8

40.0% 59.2 -6.9 +7.8

60.0% 98 13 13%

05WFS14 50 1.465 (1044)

3.891 (2773)

1.011 (3153)

0.03% (97.6%)

67.4±4.0 0.0 - 1.9% 68.0 -7.3 +8.2

98.1% 60 0 0%

32

05WFS15 50 0.5536 (586)

1.455 (1540)

0.9992 (3118)

<0.01% (>99%)

67.3±4.5 67.3 -8.2 +9.3

100% - - - 237 4 2%

05WFS16 50 0.8499 (523)

2.715 (1671)

0.9878 (3083)

26% (1.1%)

54.1±4.3 7.0 -3.9 +8.6

10.4% 60.3 -7.6 +8.7

89.6% 102 15 15%

05WFS17 50 0.7834 (464)

2.977 (1763)

0.9765 (3047)

72% (0%)

39.4±5.0 2.0 -1.6 +7.9

32.0% 62.4 -8.8 +10.2

68.0% 118 12 10%

05WFS19 50 0.5238 (288)

2.226 (1224)

0.9652 (3012)

76% (0%)

37.5±5.1 0.0

- 33.6% 62.6 -9.3 +11.0

66.4% 87 11 13%

SGM1§ 25 0.00869 (8)

7.625 (7013)

3.417 (4075)

234% (0%)

0.3±0.2 0.0 - 87.1% 2.5 -1.3 +2.8

12.9% - - -

SGM3§ 25 1.178 (944)

5.318 (4261)

3.418 (4075)

0.9% (55.7%)

42.9±2.0 42.9 -3.8 +4.1

100% - - - - - -

SGM12§ 26 0.6238 (606)

7.808 (7586)

3.419 (4075)

106% (0%)

16.0±3.4 0.0 - 38.4% 9.6 -1.7 +2.1

23.1%* - - -

SGM23§ 30 0.6679 (672)

7.354 (7399)

3.420 (4075)

105% (0%)

14.2±2.8 1.9 -0.6 +0.8

53.3% 30.6 -3.0 +3.3

46.7% - - -

SGMchips§

30 0.6208 (356)

6.124 (3512)

3,421 (4075)

113% (0%)

17.7±4.5 0.0 -0.0 +35.0

47.6% 34.1 -4.1 +4.6

52.4% - - -

04WFC1§ 10 0.2681 (105)

1.108 (434)

2.878 (4188)

102% (0%)

31.0±10.7 0.0 - 40.0% 55.0 -11.2 +14.1

60.0% - - -

04WFC4§ 18 0.3437 (150)

2.646 (1155)

2.872 (4188)

137% (0%)

16.1±5.5 0.0 - 60.8% 43.0 -7.4 +8.9

39.2% - - -

04WFC5§ 19 0.2361 (170)

2.164 (1558)

2.866 (4188)

119% (0%)

21.9±6.3 0.0 - 52.6% 53.7 -9.0 +10.7

47.4% - - -

04WFC6§ 20 0.3484 (301)

2.441 (2109)

2.880 (4188)

111% (0%)

20.0±5.2 0.4 -0.3 +1.1

50.0% 40.0 -5.2 +6.0

50.0% - - -

04WFC9A 100 1.163 (2085)

4.202 (7534)

1.108 (3458)

35.7% (0%)

50.6±2.6 3.4 -1.6 +2.9

8.7% 49.3 -8.9 +10.9

43.6%**

04WFC9B§

19 0.4466 (319)

4.707 (3362)

2.854 (4188)

83% (0%)

38.7±8.3 0.1 -0.1 +0.4

27.6% 58.7 -7.7 +8.9

72.4% - - -

04WFS11§ 1 0.1302 (3)

0.5642 (13)

2.848 (4188)

n/a 37.3±23.9 n/a n/a n/a n/a n/a n/a - - -

04WFS12§ 12 0.00607 (2)

1.211 (399)

2.843 (4188)

23.1% (26.5%)

0.8±0.6 0.8 -0.6 +2.4

100% - - - - - -

04WFS13§ 8 0.5071 (37)

1.741 (127)

2.837 (4188)

13.6% (14.2%)

46.1±9.2 46.9 -14.5 +20.9

100% - - - - - -

33

04WFS14A

50 0.9984 (780)

2.527 (1943)

1.093 (3411)

<0.01% (99.9%)

75.7±4.8 75.7 -8.7 +9.8

100% - - - - - -

04WFS14B§

15 1.179 (547)

2.959 (1373)

2.831 (4188)

23.5% (0.06%)

65.2±5.5 23.7 -8.5 +13.3

6.7% 69.0 -7.6 +8.6

93.3% - - -

04WFS15A

50 1.005 (611)

2.958 (1797)

1.078 (3364)

0.04% (93.6%)

64.9±4.3 6.5 -5.9 +65.7

3.2% 66.1 -8.0 +9.1

96.8% - - -

04WFS15B§

19 0.7992 (379)

2.501 (1186)

2.825 (4188)

46% (0%)

49.1±6.3 1.0 -0.9 +6.2

15.7% 58.9 -7.2 +8.2

84.3% - - -

(i). Analyses by external detector method using 0.5 for the 2π/4π geometry correction factor; (ii). Ages calculated using dosimeter glass: CN5 with ζ CN5 = 356.1±15.3 (SNT); CN1 with ζ CN1 = 113.8±2.9 (RAD); (iii). P χ2 is the probability of obtaining a χ2 value for v degrees of freedom where v = no. of crystals - 1; * This sample has a P3 peak of 38.4 -4.1/+4.6, with fraction% of 38.5%. ** This sample has a P3 peak of 63.4 -9.4/+11.0, with fraction% of 47.7%. § Sample analyzed by RAD at A2Z, Inc. All other samples measured by SNT at Yale. See text for methods.

34

Table 3: Single-grain apatite fission track and (U-Th)/He data

Crystal # Ns Ni Ng Dpar Dper Age (Ma) 1σ

error Min Age

50% Age

Max Age

U (ng)

Th (ng) Sm (ng) He (pmol)

FT Age (Ma)

2σ error

(analyt) 04WFC1-1 0 39 100 2.57 1.16 0.00 n/a 0.00 2.94 13.06 0.0117 0.0193 0.1790 0.020 0.73 0.63 0.65 04WFC1-10 0 34 100 - - 0.00 n/a 0.00 3.37 15.07 0.0487 0.0708 0.2692 3.64 0.73 28.0 1.3 04WFC4-7 0 50 30 - - 0.00 n/a 0.00 2.28 10.08 0.0068 0.0154 0.0387 0.250 0.60 14.7 3.0 04WFC4-9 0 100 50 - - 0.00 n/a 0.00 1.14 4.97 0.0276 0.0546 0.101 0.652 0.69 8.58 0.70 04WFC4-10 24 117 30 2.06 0.39 33.43 7.56 20.62 34.46 52.13 0.0083 0.0169 0.0311 0.863 0.65 39.9 6.5 04WFC4-11 0 64 36 2.17 0.53 0.00 n/a 0.00 1.78 7.83 0.0193 0.0424 0.119 0.053 0.73 0.92 0.33 04WFC4-15 18 78 30 2.09 0.55 37.60 9.90 21.22 39.16 63.25 0.0250 0.0842 0.0534 0.686 0.51 11.1 0.93 04WFC4-17 55 225 50 2.19 0.42 39.82 6.11 29.12 40.36 53.63 0.0420 0.108 0.0919 3.91 0.65 32.5 1.6 04WFC5-1 0 252 80 - - 0.00 n/a 0.00 0.45 1.95 0.0582 0.131 0.178 2.93 0.76 15.8 0.91 04WFC5-4 8 33 60 2.46 0.62 39.41 15.58 15.76 43.13 86.71 0.0141 0.0341 0.153 0.840 0.71 19.6 2.0 04WFC5-5 0 122 60 - - 0.00 n/a 0.00 0.93 4.05 0.0175 0.0344 0.125 0.022 0.71 0.44 0.40 04WFC5-8 18 62 100 2.42 0.81 47.17 12.71 26.30 49.18 80.58 0.0066 0.0166 0.0534 0.258 0.66 13.6 2.9 04WFC5-9 0 90 30 1.75 0.66 0.00 n/a 0.00 1.26 5.52 0.0215 0.0433 0.328 1.40 0.77 20.8 1.6 04WFC5-15 44 150 50 2.20 0.56 47.66 8.29 33.29 48.49 66.99 0.0029 0.0047 0.0488 0.252 0.65 35.0 17 04WFC6-1 0 188 60 2.24 0.90 0.00 n/a 0.00 0.60 2.61 0.0197 0.0411 0.0624 0.459 0.65 8.87 0.57 04WFC6-2 22 104 100 2.54 0.49 34.33 8.12 20.65 35.49 54.69 0.0598 0.132 0.410 5.06 0.78 26.2 1.0 04WFC6-4 56 288 60 2.54 0.52 31.57 4.70 23.29 31.98 42.12 0.0358 0.0692 0.0742 3.15 0.69 32.2 1.6 04WFC6-5 0 65 24 - - 0.00 n/a 0.00 1.74 7.67 0.0102 0.0259 0.0415 0.043 0.56 1.75 0.87 04WFC6-8 55 190 100 2.68 0.52 46.94 7.32 34.15 47.59 63.59 0.186 0.758 0.182 2.83 0.75 3.81 0.17 04WFC6-15 0 114 60 3.00 0.49 0.00 n/a 0.00 0.99 4.33 0.0387 0.0914 0.136 0.413 0.70 3.61 0.33 04WFC9-9 44 120 30 3.06 0.95 59.27 10.59 41.03 60.33 84.21 0.0088 0.0222 0.0255 1.03 0.63 43.0 6.2 04WFC9-10 24 84 80 2.76 0.70 46.23 10.79 28.12 47.70 73.35 0.0152 0.0231 0.132 1.61 0.73 38.9 4.1 04WFC9-11 1 738 60 2.48 1.05 0.22 0.22 0.01 0.37 1.23 0.450 0.411 0.343 24.1 0.71 22.7 0.90 04WFC9-15 0 440 80 1.54 0.70 0.00 n/a 0.00 0.26 1.11 0.0547 0.133 0.117 1.66 0.73 9.79 0.51

35

04WFC9-16 0 850 100 2.60 0.97 0.00 n/a 0.00 0.13 0.57 0.475 0.303 0.420 0.183 0.80 0.16 0.02 04WFC9-17 0 456 60 2.14 0.48 0.00 n/a 0.00 0.25 1.07 0.358 0.494 0.412 0.432 0.77 0.44 0.03 04WFS12-4 0 42 50 - - 0.00 n/a 0.00 2.69 11.95 0.0007 0.0015 0.0349 0.103 0.69 50.3 50 04WFS12-6 0 128 60 1.91 0.42 0.00 n/a 0.00 0.88 3.83 0.0267 0.0559 0.4007 1.87 0.67 25.4 1.3 04WFS12-11 0 26 32 1.57 0.40 0.00 n/a 0.00 4.37 19.72 0.0066 0.0133 0.122 0.856 0.58 54.9 6.3 04WFS13-3 0 12 25 - - 0.00 n/a 0.00 9.59 45.61 0.0036 0.0050 0.0443 0.007 0.56 0.99 2.9 04WFS13-7 10 32 15 1.91 0.34 50.25 18.27 22.08 54.16 104.3 0.0086 0.0104 0.0069 1.06 0.56 63.5 6.8 04WFS14-1 60 120 36 1.72 0.48 80.04 12.88 57.81 81.15 109.7 0.0248 0.0165 0.0809 3.21 0.62 66.2 4.4 04WFS14-3 40 110 60 1.66 0.33 58.31 10.91 39.61 59.46 84.27 0.0226 0.0282 0.0579 2.87 0.65 55.6 3.2 04WFS14-7 30 66 30 1.86 0.57 72.81 16.18 45.74 74.79 113.4 0.105 0.0971 0.341 16.0 0.69 66.0 2.9 04WFS14-13 24 80 27 1.92 0.33 48.14 11.30 29.21 49.68 76.62 0.0145 0.0175 0.069 2.16 0.64 67.1 5.5 04WFS15-1 0 20 32 2.06 0.42 0.00 n/a 0.00 5.67 25.92 0.0049 0.0135 0.0223 0.008 0.54 0.70 1.6 04WFS15-10 0 75 100 1.47 0.46 0.00 n/a 0.00 1.49 6.55 0.0484 0.108 0.216 0.059 0.77 0.38 0.13 04WFS15-12 27 81 30 2.01 0.48 53.36 11.96 33.23 54.90 83.20 0.0947 0.206 0.0718 10.8 0.67 41.7 1.6 SGM1-3 2 246 100 2.16 0.63 1.58 1.12 0.19 2.12 5.77 0.131 0.227 nd 5.27 0.80 13.1 0.39 SGM1-9 0 618 60 - - 0.00 n/a 0.00 0.22 0.94 0.199 0.376 0.165 4.36 0.74 7.52 0.27 SGM1-13 0 752 80 3.35 0.94 0.00 n/a 0.00 0.18 0.78 0.127 0.316 0.135 4.01 0.75 9.77 0.35 SGM1-23 4 229 40 2.75 0.84 3.40 1.72 0.92 3.97 8.81 0.0535 0.0949 nd 2.34 0.67 17.0 0.52 SGM3-5 44 236 50 2.70 0.73 36.16 6.04 25.61 36.76 50.00 0.0586 0.104 0.0680 2.94 0.70 18.6 0.71 SGM3-7 45 200 36 2.94 0.71 43.61 7.31 30.86 44.33 60.44 0.0460 0.0853 0.0534 2.48 0.72 19.2 0.77 SGM3-14 26 119 25 2.56 0.54 42.35 9.26 26.62 43.56 65.05 0.0149 0.0326 0.0183 0.66 0.70 15.5 0.84 SGM3-18 40 205 50 1.90 0.87 37.84 6.64 26.28 38.53 53.24 0.0171 0.0320 0.0214 0.86 0.69 18.7 0.97 SGM12-2 0 210 36 2.30 0.80 0.00 n/a 0.00 0.64 2.79 0.0333 0.0645 0.0503 1.80 0.72 18.9 0.76 SGM12-5 0 505 100 1.32 1.78 0.00 n/a 0.00 0.27 1.16 0.220 0.405 0.288 14.0 0.79 20.6 0.74 SGM12-8 0 484 80 - - 0.00 n/a 0.00 0.28 1.21 0.0815 0.152 0.0871 3.53 0.72 15.3 0.57 SGM12-10 0 399 70 - - 0.00 n/a 0.00 0.34 1.47 0.0623 0.120 0.0748 2.49 0.72 14.0 0.94

36

SGM12-15 47 295 50 3.20 0.77 30.92 4.94 22.24 31.39 42.14 0.0738 0.134 0.0811 2.98 0.71 14.7 0.56 SGM12-19 40 177 36 3.50 0.81 43.81 7.78 30.31 44.63 61.96 0.209 0.319 0.211 8.18 0.76 14.0 0.52 SGM12-21 12 240 40 2.47 0.78 9.72 2.89 4.95 10.27 17.29 nd nd nd nd nd nd nd SGM12-22 14 446 100 2.67 0.76 6.10 1.67 3.31 6.40 10.35 0.225 0.388 0.271 13.3 0.80 19.3 0.70 SGM12-23 34 412 70 1.99 0.77 16.03 2.90 10.96 16.36 22.76 0.0461 0.0838 0.0512 2.42 0.72 18.8 1.37 SGM12-24 20 302 100 2.32 0.63 12.87 3.00 7.75 13.31 20.22 0.104 0.174 0.146 4.35 0.76 14.5 0.54 SGM12-25 23 679 70 2.01 0.83 6.59 1.41 4.15 6.78 9.96 0.190 0.365 0.157 8.21 0.76 14.4 0.52 SGM23-2 57 416 80 2.61 0.87 26.61 3.84 19.82 26.94 35.14 0.122 0.245 0.123 7.52 0.74 20.9 0.76 SGM23-5 65 400 50 2.63 1.06 31.54 4.32 23.90 31.89 41.05 0.0472 0.1000 0.0517 1.74 0.67 13.5 0.52 SGM23-16 16 439 70 2.17 0.92 7.09 1.82 4.02 7.39 11.64 0.0678 0.130 0.0560 3.52 0.71 18.7 0.69 SGM23-17 25 286 40 1.88 0.87 16.99 3.58 10.82 17.46 25.58 nd nd nd nd nd nd nd SGM23-20 3 160 30 2.01 0.60 3.65 2.13 0.74 4.47 10.85 0.0114 0.0270 0.0239 0.41 0.60 14.3 0.94 SGM23-24 0 285 50 - - 0.00 n/a 0.00 0.47 2.06 0.0403 0.0748 0.0454 1.20 0.66 11.5 0.49 SGM23-26 0 325 50 - - 0.00 n/a 0.00 0.42 1.80 0.0420 0.0728 0.0402 1.25 0.63 12.4 0.52 SGM23-27 0 352 40 - - 0.00 n/a 0.00 0.38 1.66 0.0279 0.0760 0.0544 1.95 0.64 24.6 1.00 SGMchips-2 0 195 40 - - 0.00 n/a 0.00 0.69 3.01 0.0287 0.0547 0.0339 1.38 0.60 20.3 0.86 SGMchips-4 0 186 36 0.85 0.59 0.00 n/a 0.00 0.73 3.16 0.0249 0.0338 0.0280 0.09 0.65 1.56 0.18 SGMchips-5 0 160 30 - - 0.00 n/a 0.00 0.85 3.68 0.0156 0.0242 0.0172 0.50 0.64 13.5 0.79 SGMchips-6 0 135 25 - - 0.00 n/a 0.00 1.00 4.37 0.0414 0.0596 0.0497 1.51 0.64 15.8 0.65 SGMchips-9 0 135 25 - - 0.00 n/a 0.00 1.00 4.37 0.0409 0.0927 0.0329 0.59 0.63 5.46 0.24 SGMchips-12 22 126 30 2.54 0.99 33.90 7.90 20.53 35.02 53.52 0.0338 0.0689 0.0524 2.13 0.61 25.8 1.06 SGMchips-13 33 130 25 2.50 1.11 49.22 9.71 32.57 50.34 72.47 0.0412 0.0507 0.0435 1.89 0.65 20.2 0.87

(i). Ns = Number of spontaneous tracks, Ni = Number of induced Tracks, Ng = Number of counting squares. (ii). ζCN1 = 113.8±2.9, Area of counting square = 6.4x10-7 cm2. (iii). na = not applicable; nd = no data.

37

Table 4. Apatite (U-Th)/He data on grains dated only by (U-Th)/He method. sample name mass

(μg) radius (μm)

U ppm

Th ppm

Sm ppm

4He nmol/g

Ft corr. age (Ma)

analyt. ± (2σ)

04WFC1aA 16.66 99 5.79 9.76 nd 0.834 0.85 22.5 0.9 04WFC1aB 4.69 57 21.88 40.12 nd 3.116 0.76 24.3 0.9

04WFC4aA 2.84 43 23.45 44.92 nd 0.508 0.70 3.95 0.2 04WFC4aB 5.50 64 17.85 34.11 nd 4.171 0.78 38.4 1.5 04WFC4aC 2.12 40 48.36 102.12 179.80 6.154 0.67 23.2 0.8

04WFC5aA 8.10 65 13.45 27.64 nd 1.969 0.79 23.1 0.8 04WFC5aB 8.81 75 50.02 103.01 nd 3.021 0.81 9.33 0.3

04WFC6aA 6.53 61 9.78 19.88 nd 1.160 0.78 19.1 0.7 04WFC6aB 6.60 61 17.17 31.92 nd 0.156 0.78 1.51 0.1

04WFS11aA 1.35 34 24.14 40.89 nd 7.291 0.63 63.3 2.5 04WFS11aB 1.24 32 24.00 41.83 nd 0.059 0.62 0.525 0.1

04WFS12aA 6.41 65 13.54 20.75 nd 2.399 0.79 30.7 1.1 04WFS12aB 4.04 51 17.10 27.63 nd 4.371 0.74 46.3 1.6 04WFS12aC 1.92 43 20.01 30.02 331.85 0.016 0.69 0.157 0.1 04WFS12aD 1.86 44 39.05 51.91 401.62 2.235 0.69 11.6 0.4

04WFS13aA 1.34 41 44.65 69.37 nd 0.000 0.66 0.0 0.0 04WFS13aB 3.89 52 30.02 62.44 nd 12.991 0.74 72.4 2.4 04WFS13aC 1.01 36 40.92 33.60 361.96 10.660 0.63 63.4 2.7

05WFS5a1 1.54 35 105.23 130.50 175.22 32.883 0.64 69.6 3.2 05WFS5a2 1.54 37 121.35 122.27 157.16 32.833 0.65 61.7 2.9 05WFS5a3 2.40 45 100.23 149.30 178.64 33.547 0.70 64.9 2.9 05WFS5a4 0.65 28 188.59 239.12 221.20 52.996 0.56 71.8 3.3 05WFS5a5 3.36 50 93.90 102.70 145.66 30.826 0.73 65.7 3.0 05WFS5a6 1.29 33 122.22 141.89 176.47 35.322 0.63 66.8 3.1

04WFC9a1 1.99 40 138.45 62.80 206.39 25.943 0.68 45.8 2.2 04WFC9a2 3.48 51 7.88 12.89 57.37 1.791 0.73 41.0 1.9 04WFC9a4 4.65 49 188.44 299.16 131.60 29.366 0.74 28.5 1.7 04WFC9a5 17.03 90 7.29 10.13 40.28 1.189 0.83 27.2 1.7 04WFC9a6 1.91 40 79.59 52.80 156.92 8.312 0.66 25.3 1.6 04WFC9a7 0.98 33 56.13 151.16 69.62 8.934 0.58 31.1 1.8 04WFC9a8 1.03 31 114.67 239.60 116.57 17.660 0.57 33.2 2.0 04WFC9a9 7.48 65 7.22 9.64 47.60 1.359 0.78 33.9 2.2 04WFC9a10 3.81 53 29.23 64.71 94.84 13.019 0.72 74.6 4.4 04WFC9a11 1.28 34 24.73 28.71 208.81 1.894 0.63 17.5 1.2 04WFC9a12 2.06 41 94.29 70.66 169.23 4.288 0.67 10.7 0.7 04WFC9a13 1.63 38 58.78 153.25 79.38 7.135 0.67 20.9 1.3 04WFC9a14 2.27 40 70.75 76.00 235.95 25.977 0.67 80.6 4.9 04WFC9a15 0.65 30 28.73 70.06 135.68 6.172 0.57 44.5 2.9 04WFC9a16 1.14 35 16.74 33.90 63.37 3.501 0.60 43.2 3.0 04WFC9a17 2.10 42 84.61 102.29 183.51 4.038 0.67 10.3 0.4 04WFC9a18 6.47 65 160.83 181.39 98.28 23.237 0.79 26.8 1.0

38

04WFC9a19 16.71 79 14.41 21.37 68.10 3.934 0.82 45.4 1.7 04WFC9a21 10.11 78 7.86 10.39 49.31 1.266 0.82 27.4 1.2 04WFC9a22 9.35 68 58.97 69.93 179.15 11.167 0.79 34.6 1.7 04WFC9a23 5.15 42 74.25 33.41 160.61 7.641 0.71 24.3 1.0 04WFC9a24 17.26 87 13.99 18.73 52.09 2.845 0.84 33.9 1.3 04WFC9a25 21.02 104 6.18 8.46 39.42 1.148 0.86 30.1 1.1 04WFC9a26 11.05 71 7.19 9.60 44.49 1.501 0.80 36.6 1.5 04WFC9a27§ 3.18 41 4.12 6.88 276.59 0.040 0.68 1.77 1.0 04WFC9a28 4.18 54 83.16 43.18 169.75 3.474 0.74 9.33 0.4 04WFC9a29 4.58 58 45.80 88.53 94.07 8.118 0.74 30.3 1.2 04WFC9a30 1.40 35 28.18 42.05 101.17 3.904 0.64 29.7 1.1 04WFC9a31 1.97 45 203.30 216.34 95.07 20.824 0.70 21.8 0.8 04WFC9a32 1.35 38 235.22 240.77 77.62 20.201 0.66 19.5 0.7 04WFC9a33 2.17 45 15.78 22.36 63.54 2.483 0.70 31.2 1.1 04WFC9a34 9.12 70 164.21 152.46 112.49 25.308 0.81 29.0 1.1 04WFC9a35 17.24 88 18.93 29.18 105.81 4.064 0.84 34.5 1.2 04WFC9a36 10.33 66 15.61 24.77 80.22 4.160 0.80 44.7 1.6 04WFC9a37 6.39 56 22.30 34.43 82.18 5.168 0.77 40.8 1.4 04WFC9a38 13.80 85 50.79 29.49 115.60 5.430 0.84 20.8 0.8 04WFC9a39 4.14 56 29.23 23.70 182.37 1.441 0.76 10.1 0.4 04WFC9a40 15.02 82 9.84 12.07 55.27 1.731 0.83 30.2 1.1 04WFC9a41 4.51 53 121.56 210.47 120.53 16.866 0.75 24.4 0.8 04WFC9a42 2.03 42 24.78 34.52 86.52 4.414 0.69 36.0 1.3 04WFC9a43 3.24 46 98.34 90.16 147.46 12.887 0.72 27.6 1.0 04WFC9a44 4.09 51 101.39 136.24 115.19 12.453 0.74 23.3 0.8 04WFC9a45 9.48 86 156.38 221.61 124.77 25.831 0.82 27.9 1.0 04WFC9a46 5.27 46 107.90 58.17 160.77 10.470 0.74 21.7 0.8 04WFC9a47 2.58 43 137.78 177.33 98.40 16.046 0.70 23.6 0.8 04WFC9a48 3.74 49 114.83 135.19 113.56 25.479 0.73 43.8 1.6 04WFC9a49 3.48 49 27.95 55.86 103.32 4.249 0.73 26.2 0.9 04WFC9a50 4.64 51 179.93 215.26 102.60 28.836 0.74 31.1 1.1 04WFC9a51 3.20 44 23.05 34.57 74.01 3.489 0.71 29.1 1.0 04WFC9a52 1.92 41 165.12 117.33 107.79 16.647 0.68 23.4 0.9 04WFC9a53 1.61 39 55.19 29.64 225.99 5.831 0.67 25.9 1.0 04WFC9a54 2.91 52 20.95 29.14 87.27 4.973 0.73 45.0 1.6 04WFC9a55 2.24 39 24.98 38.78 106.94 4.130 0.67 33.1 2.0 04WFC9a56 1.99 38 9.25 11.92 45.71 1.451 0.67 33.3 1.2 04WFC9a57 2.26 40 80.29 64.52 153.48 11.231 0.69 31.6 1.2 04WFC9a58 2.56 49 138.62 144.15 159.09 22.356 0.72 33.3 1.2 04WFC9aA 8.59 73 13.46 17.45 nd 2.514 0.81 32.8 1.2 04WFC9aB 6.71 57 120.52 177.79 nd 28.251 0.77 41.8 1.4

05WFS15a1 3.46 43 23.66 43.29 72.84 5.571 0.71 42.9 1.5 05WFS15a2 5.72 53 66.46 105.18 94.50 26.047 0.75 69.8 2.4 05WFS15a3 1.26 29 130.49 100.26 88.43 25.897 0.60 51.9 1.9 05WFS15a4 2.81 45 63.62 95.72 67.80 15.265 0.71 46.1 1.6 05WFS15a5 1.73 37 15.97 33.33 33.89 3.419 0.65 40.6 1.8 05WFS15a6 1.54 36 20.48 28.52 158.09 3.243 0.65 33.8 1.5 05WFS15a7 1.54 36 132.46 189.44 41.00 17.353 0.65 28.0 1.0 05WFS15a8 9.23 58 23.05 65.88 56.86 4.429 0.78 27.3 0.9 05WFS15a9 1.36 35 71.31 138.92 242.00 17.308 0.63 48.5 1.7

39

05WFS15a10 2.61 44 72.87 90.35 143.43 12.229 0.70 34.1 1.2 05WFS15a11 0.84 29 50.41 82.47 187.94 9.228 0.58 42.2 1.7 05WFS15a12 1.01 29 48.70 98.55 58.88 10.123 0.58 44.6 1.7 05WFS15a13 2.48 46 41.91 110.00 64.37 14.396 0.70 55.8 1.9 05WFS15a14 1.24 32 20.19 21.46 150.32 6.311 0.62 74.0 3.8 05WFS15a15 1.46 35 52.20 129.67 74.17 13.654 0.63 48.1 1.7 05WFS15a16 0.54 26 30.53 31.54 11.50 4.122 0.54 37.5 2.9 05WFS15a17 0.65 28 41.66 58.41 37.50 5.047 0.55 30.4 1.5 05WFS15a18 1.38 36 59.61 56.52 19.46 17.860 0.65 70.1 2.7 05WFS15a19 4.60 55 26.64 55.01 34.61 4.834 0.75 30.0 1.0 05WFS15a20 0.47 24 38.93 66.59 170.53 6.703 0.50 45.4 2.6 05WFS15a21 0.59 31 37.00 73.66 32.52 8.672 0.56 52.2 2.7 05WFS15a22 0.60 30 49.10 78.54 127.91 7.630 0.56 36.9 1.7 05WFS15a23 2.24 43 52.54 94.26 113.36 10.247 0.69 36.5 1.3 05WFS15a24 1.73 37 70.79 155.79 46.54 20.059 0.65 52.9 1.8 05WFS15a25 0.82 32 97.31 111.88 104.30 21.929 0.60 54.6 2.1 05WFS15a26 0.39 23 68.53 96.54 56.95 12.785 0.48 54.2 2.6 04WFS15a53§ 2.44 39 14.90 43.45 39.21 0.013 0.66 0.144 0.1 04WFS15a54 3.54 45 29.61 44.60 34.35 6.754 0.72 43.4 2.0 04WFS15a55 2.48 44 4.23 8.41 72.51 1.205 0.70 50.5 10.8 04WFS15a56 1.28 34 26.06 58.34 49.78 4.820 0.60 37.0 2.7 04WFS15a57 1.95 39 35.30 63.82 165.65 8.372 0.65 46.9 2.4 04WFS15a58 2.82 44 68.26 123.77 66.71 17.887 0.69 49.3 2.0 04WFS15a59 4.99 63 27.03 65.52 54.20 7.545 0.75 43.7 1.8 04WFS15a60§ 2.21 37 56.14 134.65 199.16 0.810 0.65 2.63 0.1 04WFS15a61 2.64 42 24.44 62.55 52.24 8.516 0.69 58.2 2.8 04WFS15a62 1.32 34 8.39 30.65 81.93 1.548 0.62 29.4 4.8 04WFS15a63 3.73 46 22.84 51.71 31.09 5.599 0.72 41.2 1.9 04WFS15a64 2.34 39 17.98 56.11 39.23 4.435 0.65 40.1 2.3 04WFS15a65 2.23 45 57.58 66.15 113.55 17.174 0.70 61.6 2.8 04WFS15a66 2.35 44 18.21 54.29 31.45 5.673 0.69 48.8 2.8 04WFS15a67 2.72 45 16.62 50.76 30.58 4.586 0.71 41.5 2.3 04WFS15a68 6.35 71 39.32 57.43 137.01 14.834 0.79 65.2 2.6 04WFS15a69 1.65 36 35.07 38.80 160.03 7.709 0.65 49.1 2.9 04WFS15a70 1.33 34 31.51 83.58 80.25 8.603 0.60 51.3 3.1 04WFS15a71 2.48 40 13.12 0.12 23.19 1.333 0.69 27.0 2.9 04WFS15a72 1.82 36 20.31 57.24 46.73 6.591 0.64 55.8 3.6 04WFS15a73 4.94 56 19.97 66.27 47.18 7.031 0.75 48.3 2.0 04WFS15a74 0.97 31 89.86 86.31 72.99 21.703 0.60 60.7 3.1 04WFS15a75 1.64 34 35.48 82.65 26.37 8.470 0.63 45.2 2.4 04WFS15a76 1.32 31 42.13 81.50 214.65 14.472 0.61 71.3 3.9 04WFS15aA 2.77 41 23.80 67.40 nd 7.367 0.69 50.0 1.7 04WFS15aB 1.57 39 13.27 25.62 nd 3.350 0.66 48.5 2.2

04WFS14A1 1.91 41 48.86 54.83 204.85 15.090 0.68 65.8 2.5 04WFS14A2 3.66 46 20.54 23.18 115.24 6.664 0.72 65.3 2.5 04WFS14A3 2.65 37 37.63 32.97 125.06 10.961 0.67 66.3 2.6 04WFS14A4 1.98 43 33.38 28.26 96.04 8.262 0.69 55.1 2.2 04WFS14A5 2.00 43 25.55 27.84 106.91 7.078 0.69 58.6 2.3 04WFS14A6 3.73 51 40.85 27.91 183.97 13.357 0.74 69.8 2.7 04WFS14A7 1.44 37 46.71 30.74 177.26 12.868 0.65 67.2 3.3

40

04WFS14A8 2.55 43 38.23 27.01 139.78 10.460 0.70 61.5 2.4 04WFS14A9 0.57 28 39.23 19.93 138.54 8.534 0.56 64.2 3.6

04WFS14A10 2.61 43 29.30 27.79 113.57 8.458 0.70 61.7 2.4 04WFS14A11 3.95 50 67.57 35.12 215.63 18.368 0.74 60.1 2.3 04WFS14A12 0.67 27 41.57 26.35 179.58 9.705 0.55 67.5 3.5 04WFS14A13 1.32 33 40.37 27.58 92.04 9.345 0.63 58.8 2.4 04WFS14A14 0.72 29 30.63 31.08 133.75 6.840 0.57 58.2 3.1 04WFS14a15 2.72 38 70.37 45.75 248.65 20.355 0.68 68.0 2.6 04WFS14a16 1.15 33 23.60 25.22 123.31 6.232 0.62 62.8 2.9 04WFS14a17 1.22 32 42.97 32.68 183.10 11.432 0.62 67.1 2.7 04WFS14a18 0.48 25 82.52 69.28 236.25 16.266 0.51 59.0 2.5 04WFS14a19 1.68 49 21.98 33.90 144.20 8.579 0.69 75.8 3.0 04WFS14a20 0.74 32 19.90 20.06 210.96 4.588 0.59 57.6 3.9 04WFS14a21 0.61 26 38.52 51.67 180.01 10.705 0.53 72.7 3.1 04WFS14a22 0.34 26 32.66 42.87 160.11 9.505 0.50 81.4 4.7 04WFS14a23 0.54 24 53.44 42.32 138.37 9.521 0.51 54.4 2.3 04WFS14a24 0.57 29 24.56 25.05 74.51 5.778 0.56 62.3 3.7 04WFS14a25 0.84 32 51.78 44.14 141.03 11.795 0.60 58.3 2.3 04WFS14a26 1.21 33 29.41 22.28 133.26 8.276 0.63 70.2 2.8 04WFS14a27 1.20 33 59.56 41.28 174.68 15.295 0.63 65.0 2.5 04WFS14a28 1.16 32 117.63 92.42 265.51 26.527 0.62 56.8 2.1 04WFS14a29 1.38 44 70.82 72.94 157.76 20.492 0.67 63.8 2.3 04WFS14a30 0.85 29 22.86 30.80 161.16 5.872 0.58 61.8 2.8 04WFS14a31 0.84 31 47.72 27.65 172.66 12.543 0.60 71.4 3.0 04WFS14a32 0.78 30 37.04 32.47 111.22 8.963 0.58 63.6 2.7 04WFS14a33 1.19 35 22.33 18.07 202.59 5.750 0.64 62.3 2.7 04WFS14a34 0.55 25 55.47 29.69 185.38 12.207 0.52 68.7 2.9 04WFS14a35 0.70 27 83.23 56.75 213.67 20.934 0.56 71.6 2.8 04WFS14a36 0.65 25 43.46 24.75 152.89 9.686 0.53 67.9 2.9 04WFS14a37 0.71 26 50.35 36.01 193.29 12.078 0.55 69.0 2.9 04WFS14a38 0.78 29 62.22 32.90 212.40 14.666 0.58 66.7 2.7 04WFS14a39 1.14 35 40.01 32.24 109.67 9.476 0.64 57.8 2.3 04WFS14a40 1.42 34 29.51 28.60 94.99 7.083 0.64 56.5 2.2 04WFS14a41 0.99 31 41.38 32.57 183.58 11.561 0.61 71.6 2.8 04WFS14a42 0.76 30 38.28 25.74 166.16 9.742 0.58 69.3 2.9 04WFS14a43 0.60 27 45.24 30.48 159.06 10.114 0.55 65.1 3.1 04WFS14a44 0.68 30 45.70 30.28 146.98 11.508 0.58 69.2 2.9 04WFS14aA 10.60 65 43.56 28.21 nd 12.244 0.80 56.3 2.1 04WFS14aB 2.60 45 18.30 25.42 nd 6.269 0.71 67.5 2.5

§The He ages of these grains are probably partially reset by wildfire, but this is not conclusive because of the lack of AFT ages on the same grains.

41

Figure 1

42

Figure 2.

43

Figure 3.

44

Figure 4.

45

Figure 5.

46

Figure 6.

47

Figure 7.

48

Figure 8.

49

Figure 9.

50

Figure 10.

51

Figure 11.

52

Figure 12.

53

Figure 13.

54

Figure 14.

55

Figure 15.

56

Figure 16.

57

Figure 17.

58

Figure 18.

59

Figure 19.

60

Figure 20.

61

Figure 21.

62

Figure 22.