Data Repository for. Hydrological transformation coincided with megafaunal extinction in central Australia

Transcription

1 GSA Data Repository Data Repository for Hydrological transformation coincided with megafaunal extinction in central Australia Tim J. Cohen, John D. Jansen, Luke A. Gliganic, Joshua R. Larsen, Gerald C. Nanson, Jan-Hendrik May, Brian G.Jones, and David M. Price correspondence to: This PDF file includes: Methods Supplementary Figures DR1 to DR18 Supplementary Tables DR1 to DR8 Supplementary References 1

2 Methods 1. Archaeology We use the Australian Archaeological database (Williams et al. 2008) to plot the distribution of archaeological sites in the Lake Eyre basin (LEB). This dataset utilises radiocarbon and luminescence ages to highlight the dominance of sites < 23 ka within the LEB (Fig DR1). 2. Optically stimulated luminescence chronology Dose rate measurement Lithogenic radionuclide activity concentrations were determined using highresolution gamma spectrometry (Murray et al., 1987), with dose rates calculated using conversion factors (Stokes et al., 2003). β-attenuation factors were taken from (Mejdahl, 1979) whilst cosmic dose rates were calculated from (Prescott and Hutton, 1994). For all samples, dose rates have been calculated using the as measured radionuclide contents. An assumed water content of 2 ± 1% was used to calculate dose rate attenuation by moisture (Aitken, 1985), and an internal dose rate of 0.03 ± 0.01 Gy/ka was assumed based on previous measurements of quartz from Australia (Bowler et al., 2003) and South Africa (Jacobs et al., 2006). Results are shown in Table DR1. Sample preparation and D e measurement Quartz grains from each of the Lake Eyre (n = 15) samples were obtained and prepared in the standard manner (e.g. Aitken, 1998). Sodium polymetaphosphate was used to remove clays, followed by wet then dry sieving to obtain the µm fraction. This was then treated with hydrochloric acid and hydrogen peroxide to remove carbonates and organics. Hydroufluoric acid (40%) etches of 40 and 5 minutes were then undertaken to remove the majority of feldspars and to etch the outer ~10 µm alpha-irradiated rind of each grain, followed by a further wash in hydrochloric acid to remove acid-soluble fluorides. Two density separations using sodium polytungstate solutions with specific gravities of 2.72 and 2.62 were then used to ensure mineralogical purity. Finally, the samples were wet sieved through a 180 µm sieve to remove any grains fractured during the etching process. Individual sand-sized ( µm diameter) quartz grains were mounted on aluminum discs with µm-diameter holes and loaded into a Risø DA20 TL/OSL reader for measurement. Single quartz grains were stimulated with green (532 nm) laser light (Bøtter-Jensen et al., 2003) for 2 s at 125 C. D e values were estimated by summing first 0.22 s of signal and using the final 0.33 s as background. The ultraviolet OSL emissions were measured using an Electron Tubes Ltd 9635Q photomultiplier tube fitted with 7.5 mm of Hoya U-340 filter. Laboratory irradiations were given using calibrated 90 Sr/ 90 Y beta sources mounted on Risø TL/OSL reader. A modified single-aliquot regenerative-dose (SAR) procedure (Murray and Wintle, 2000) was used to obtain D e values. This procedure makes use of the sensitivity-corrected OSL signals from laboratory applied regenerative doses to construct a dose-response curve onto which the sensitivity-corrected natural signal is projected to obtain a D e estimate. Sensitivity changes are monitored and corrected for using the OSL signal resulting from a small test dose, 10 Gy in this study, applied after the optical stimulation of each natural and regenerative dose (Murray and Wintle, 2000). 2

3 The suitability of the SAR procedure for obtaining a robust dose estimate from each grain was assessed using standard tests. Grains were rejected as having an unsuitable OSL signal if they failed one of the following tests: 1. Grains with dim OSL signals were rejected if (i) the test dose of the natural had a background that was less than 3 times its signal, or (ii) no OSL decay curve was observed. 2. Recycling ratio (Murray and Wintle, 2000): a grain was rejected if the ratio of two identical regenerative doses was more than 2σ difference from unity. Two recycling ratios (40 and 178 Gy) were employed in this study. 3. Thermal-transfer/recuperation test (Yoshida et al., 2000): a grain was rejected if the ratio of a sensitivity-corrected OSL signal following a 0 Gy regenerative dose and the natural was greater than OSL-IR depletion ratio test (Duller, 2003): grains whose OSL signal was depleted following infra-red stimulation for 40 s at 50 C were inferred to have feldspar inclusions within the grain, and were thus rejected. 5. Grains for which the sensitivity-corrected natural signal did not intersect the dose-response curve (i.e., class-3 type grains following Yoshida et al., 2000). In addition, grains were rejected if the sensitivity-corrected natural signal was projected onto the saturated portion of the dose-response curve and if the estimated D e value was consistent with zero at 2σ. The latter criterion was employed to remove intrusive modern grains from the dataset. Dose recovery experiments (Roberts et al., 1999; Murray and Wintle, 2003) using single-grains (Gliganic et al., 2012) were used as an empirical means of determining the appropriate regenerative and test dose preheat combinations for a representative sample (TR9-2), as well as demonstrating the SAR procedure could be used to reliably obtain a known dose. Two subsets of grains were bleached in unfiltered sunlight for ~3 days to remove the latent natural OSL signal before being given a beta irradiation of 40 Gy or 120 Gy. These tests were used to assess the suitability of the measurement conditions and sensitivity-corrected OSL signal at doses below and close to the anticipated OSL-saturation limit 48. Results, presented in Table DR2, show that the regenerative and test-dose preheat combination of 260 C for 10 s and 220 C for 5 s, respectively, produced measured/given dose ratios consistent with unity in both tests, suggesting the suitability of this procedure for dose estimation. A modified SAR procedure using a preheat combination of 260/220 C and a 10 Gy test dose was used to measure between 800 and 1000 quartz grains from the Illusion Plains samples. Table DR3 lists the number of grains measured for each sample, as well as the number rejected based on the various criteria described in the previous section. Approximately 70 and 80% of grains were rejected because of the dimness of their OSL signals while 2 and 9% had saturated OSL signals. Between 6 and 11% of grains had OSL signals suitable for D e estimation. D e analysis The D e values obtained from grains with suitable OSL signals are plotted as radial plots in Figures DR3-DR6. The central age model (CAM; Galbraith et al., 1999) was used to calculate the weighted mean D e (displayed as lines in Figure DR3 DR6) and overdispersion values of each dataset (Table DR3). The overdispersion value represents the spread in D e values beyond that accounted for by measured 3

4 uncertainties on the individual D e estimates. Overdispersion values for our samples range between 23% and 78%. To further characterize the D e distributions for each sample, the finite mixture model (FMM; Roberts et al., 2000) was applied to each dataset. Results are presented in Table DR4 and are shown as grey shaded bands in Figure DR3 DR6. All datasets except one were best fitted by two components, while sample TR2-4 was best fitted by four components. Table DR5 shows the ages calculated using the weighted mean D e values calculated using the CAM and the major population identified using the FMM. The choice of which model most likely represents deposition was assessed on statistical and contextual grounds. For some samples the interpretation of the CAM or FMM as depositional is clear but for other samples, contextual and comparative evaluation is required to deduce which ages most likely represent deposition. For example, we interpret the D e data from samples TR2-3, TR5-1 and TR1-3 to be most appropriately described by the CAM because no dominant FMM component was identified, these samples have a relatively low CAM overdispersion (23 27%), and no stratigraphic evidence for mixing was described. Thus, the use of the FMM cannot be justified for these samples and the age calculated using the CAM is interpreted as depositional. The remaining samples are comprised of populations of grains with significantly different D e values and one dominant population that comprises the majority of grains. Shoreline chronology Stratigraphic observations and statistical methods were used to determine the relationship between ages. Samples were grouped into subsets based on their similarity in stratigraphy, elevation and age. The consistency of age estimates in each elevation-subset was assessed using the homogeneity-test (Galbraith, 2003). Gvalues, which will be from a chi-squared distribution with n-1 degrees of freedom if all estimates are consistent, were calculated for each subset and the G-value was, thus, used to obtain P-values for each subset. A significantly small P-value (e.g. <0.05) suggests that the age estimates in a subset are not consistent with a common value. Results are presented in Table DR6. The samples in each shoreline-subset have statistically consistent ages. A weighted mean shoreline age was, thus, calculated for each subset. These results indicate that Lake Eyre was full to the following depths at: 47.9 ±1.8 ka 18.8 m deep (shoreline at ~3.8 m-ahd) 55.9 ± 2.1 ka 17.5 m deep (shoreline at ~2.5 m-ahd) 59.8 ± 1.9 ka 20 m deep (shoreline at ~5 m-ahd) 78.8 ± 4.2 ka 26 m deep (shoreline at ~11 m-ahd) 93.5 ± 3.8 ka 25 m deep (shoreline at ~10 m-ahd) ± 4.6 ka 24 m deep (shoreline at ~9 m-ahd) 3. Lake hypsometry and water balance Lake Hypsometry and level comparison 4

5 Using topographic data (Leon and Cohen, 2012), we examined the hypsometric changes and differences between Lake Eyre and Lake Frome from the lowest measured level until the highest recorded shoreline on each lake (Fig. 2 main text). Given the lake hypsometry may significantly influence the recorded shoreline elevations (Mason et al., 1994), we first evaluate the degree to which a relative change (%) in inflow volume (V ) causes a relative change in lake depth ( D ) as: dd dd dv = (1) D dv V This demonstrates that for the same inflow volume in both lakes, Lake Frome has a change in depth ~30% greater than does Lake Eyre because of differences in hypsometry (Fig. DR7 DR8). This effect is particularly important in the interpretation of smaller lake depths which are recorded at Lake Frome post 48 ka, but not on Lake Eyre (Fig. 2 main text). Equilibrium water balance We construct a lake water balance for Lake Eyre on the basis of equilibrium levels and area. Although there is in reality considerable variation in lake dimensions over short time periods ( years), especially in arid and semi-arid settings, given the paucity of modern data available we consider the hypothetical equilibrium level to be equivalent to the mean lake water balance over these time scales. For closed lake basins, the basis of the annual water balance is relatively simple, and wellstudied (Langbein, 1961; Benson and Paillet, 1989; Mason et al., 1994): EA = PA + Q (2) Where the only outputs are due to evaporation ( E, m/yr), the only inputs due to river inflow ( Q, m 3 /yr) and direct precipitation over the lake ( P, m/yr), and the geometric response of the lake is in terms of lake area ( A, m 2 ). Since Lake Eyre is the lowest point in Australia, some groundwater inflow can also be expected, however this is likely to be minimal in comparison to the other terms in the water balance (groundwater outflow can be ignored for the same reasons). Since Lake Eyre is a closed lake basin, it is necessary that E > P and we can therefore regard the net atmospheric losses from the lake as L N = E P. Equation (2) therefore simplifies (Benson and Paillet, 1989) to: Q A = (3) L N This illustrates that A is the crucial metric by which closed lakes can respond to changes in inflow (Q ) or the net atmospheric losses ( L N ). Since we have a good estimate of the modern extent of A, here taken to be the largest historical filling in 1974 ( km 2 ), and the palaeo extent of A can be estimated based on the palaeoshoreline elevation (Fig. DR2, Table DR7), we can re-arrange equation (3) in terms of Q to determine the relative changes in inflow hydrology and water availability within Lake Eyre for the last ~120 ka. This also requires realistic estimates of palaeo L N, for which we apply four different values derived from the water balance data available from (Bye and Will, 1989) shown in Fig. DR9a. 5

6 Based on these modern estimates of lake precipitation and evaporation, we adopt a range of L N values from wet (1200 mm/yr) to dry (2100 mm/yr) (Figure DR9b). Using these L N ranges, we can then calculate a hypothetical equilibrium inflow ( Q E ) required to maintain the lakes at the estimated aerial extent for all palaeo-shoreline elevations (Table DR7). Given that shoreline development requires a degree of lakelevel stability, each shoreline is assumed to indicate an interval of equilibrium inflow conditions, and thus Q becomes Q in Equation (3). E Response time We analyse the timescale over which the lake can respond and adjust back to an equilibrium depth following instantaneous inflows via the evaporative adjustment of lake area. Since Lake Eyre is a closed lake this equilibrium response time ( τ e ) is straightforward and can be formulated (Mason et al., 1994) as: A τ e = da (4) L dd Where D is lake depth. Equation (4) shows that the primary adjustment mechanism of closed lakes to changes in inflows is net evaporative losses per unit lake area, and is therefore highly dependant on the lake hypsometry (Fig. DR7). We apply Equation (4) to the hypsometric data in Fig. DR7a for every 1m change in lake depth, and for the range of L N values outlined above (to provide a conservative estimate of uncertainty), and show that the geometric variations produce characteristic changes in the response time of the lake (Fig. DR10b). Lake filling and drying In order to conceptually explore the increases and decreases in inflow required to fill and dry Lake Eyre, we construct a time varying water balance that calculates the volume (V ) of the lake at time t as: Vt = ( Vt 1+ Qt) ( EAt 1 PAt 1) (5) Where A= f( V), which we estimate using linear regression for the volume-area range of interest (Fig. DR11), which gives f( V) = V For E and P the wettest (low E, high P ) measured values from 1974 are used ( E = 1767 mm/yr, P = 472 mm/yr, Fig. DR9a). These constant values for E and P would not necessarily be realistic for the large time ranges covered here, thus they must be considered as a minimum constraint on the inflow required, and therefore also on the likely changes in lake volume. For the sake of simplicity, we also use a series of constant inflow volumes, since there is a large range of possible iterations for this parameter. Given these assumptions, our analysis is useful for a number of reasons. The first is that Equation (5) highlights important limits on the capacity of certain inflow magnitudes to fill the lake, since as EAt 1 PAt 1 Qt, then Qt QE (6) which states that as net lake losses necessarily increase due to the expanding lake area, they may approach the inflow volume and therefore become equivalent to an equilibrium inflow (as indicated by the asymptotic curves on the y-axis in Fig. DR12). At our prescribed E and P rates, inflows less than ~33 km 3 /yr are not able to produce the highest lake levels, since at this aerial extent, the net losses become ~32.9 N 6

7 km 3 /yr. This also results in highly non-linear filling times with small changes in inflow volume resulting in large differences in filling times (Fig. DR12). 4. Fluvial bedload analysis Inventory of bedload units Table DR8 presents an inventory of 159 channel bedload units dated with optically stimulated luminescence (OSL) or thermoluminescence (TL) from natural exposures and drill-holes along the main rivers draining Lake Eyre basin (Fig. DR1). Uncertainty distribution & OSL-TL correlation Figure DR13 presents scatter plots of the dated bedload units (<150 kyr, n=130) with their absolute and relative standard-error uncertainty. The good correlation that exists between a set of 20 duplicate TL and OSL ages (Fig. DR14) warrants their amalgamation for the purposes of frequency distribution analysis (Figs DR15 18). Correcting for bias Fig. DR16 presents the frequency distribution of dated bedload units in the form of a normal kernel density estimate (NKDE): a data-smoothing method constructed by representing each age as a Gaussian probability curve, then averaging the array of curves point-wise to create a single composite curve (Galbraith and Roberts, 2012). The inherent incompleteness of the stratigraphic record is the result of several factors that combine to bias the frequency distribution of recorded geological events. The most important of such biases is the Sadler effect : the power-law decrease in erosion or sedimentation rate with measurement interval, thought to reflect the distribution of hiatuses between geological events (Sadler, 1981; Schumer and Jerolmack, 2009). Average deposition rates were calculated from the inventory of dated channel bedload units (Table DR8), and then log-binned before regression analysis (Fig. DR17). The apparent decrease in deposition rate over time is consistent with the Sadler effect bias, as described by a power law: V = C t ϒ, where V is apparent deposition rate, t is time, and C and ϒ are constants. An empirical calibration was applied as a first-order correction to the age distribution via t ϒ (where ϒ = 0.658, Fig. DR17) yielding a corrected distribution of fluvial activity (Fig. DR18, Fig. 2C). 7

8 Data Repository Figures Figure DR1. Distribution of archaeological sites ( 14 C and TL) in the Lake Eyre basin derived from (Williams et al., 2008). F = Finke River, G = Georgina River, D = Diamantina River, C = Cooper Creek. 8

9 Figure DR2. Distribution of transects (IP1 Illusion Plains North and IP2 Illusion Plains South) on Lake Eyre South. Digital elevation derived from 30 m SRTM grid. 9

10 TR9_4 TR9_3 TR9_2 TR9_1 Figure DR3 D e data, displayed as radial plots, obtained from the OSL of single quartz grains for samples from Illusion Plains North Trench 9. The bold lines display the CAM D e estimate and the shaded bands display populations identified using the FMM. 10

11 TR4_1 TR1_4 TR1_3 TR1_2 TR1_1 Figure DR4 D e data, displayed as radial plots, obtained from the OSL of single quartz grains for samples from Illusion Plains North Trench 4 & 1. The bold lines display the CAM D e estimate and the shaded bands display populations identified using the FMM. 11

12 TR2_4 TR2_3 TR2_2 Figure DR5 D e data, displayed as radial plots, obtained from the OSL of single quartz grains for samples from Illusion Plains South Trench 2. The bold lines display the CAM D e estimate and the shaded bands display populations identified using the FMM. 12

13 TR6_2 TR5_2 TR5_1 Figure DR6 D e data, displayed as radial plots, obtained from the OSL of single quartz grains for samples from Illusion Plains South Trench 6 & 1. The bold lines display the CAM D e estimate and the shaded bands display populations identified using the FMM. 13

14 Figure DR7 Hypsometric relationships for Lake Eyre (a & c), and Lake Frome (b & d). (a) and (b) are for lake area vs depth, and (c) and (d) for volume vs depth. 14

15 Figure DR8 Relative (%) change in lake depth due to a relative change in lake volume for Lake Eyre (open circles) and Lake Frome (black circles) using Equation (1). The best fit linear regression for Lake Frome and for Lake Eyre, indicating a ~30% greater response in lake depth for the same inflow volume in Lake Frome than Lake Eyre. 15

16 Figure DR9 Modern Lake Eyre estimates of (a) annual direct lake evaporation and precipitation from , and (b) the same data converted to L N vs precipitation. The dashed lines in (b) correspond to the range of L N estimates used in the palaeo water balance, where L N = 2100, 1800, 1500, 1200 mm/yr for numbers 1-4 respectively. Data from (Bye and Will, 1989). 16

17 Figure DR10 Equilibrium lake inflows for Lake Eyre (A) for the different P and E lake climate conditions in Fig. DR9. The 1974 peak inflow (~33 km 3 ) is shown for comparison as a dashed horizontal line. The grey shaded area indicates that the inflow reduction extent as an estimation based on data from (Magee et al., 1995; Magee et al., 2004). At 3 m AHD (lake depth 12 m) is a deflated desert pavement, with polyginal cracking and ventifact pebbles with bright desert varnish. This is interepreted by (Magee et al., 1995; Magee et al., 2004) as evidence of widespread lake-level lowering and deflation. We have used this and the bounding chronological control from units above and below as the only estimate for previous lake level low-stands. Lower case letters label each lake phase. (B) Change in lake response time for 1m depth increments (using Equation (4) with the shaded grey area covering the range of lake climates in Fig. DR9. The lake area-depth curve (black line) is shown for comparison. The numbers (1, 2, 3) with arrows below correspond to different response time trends discussed in the text: 1 refers to the fast response time, post 48 ka phase, 2 refers to the large increase in response time hump, and 3 refers to the variable but generally high response time range within which the higher shorelines occur. For simplicity in the main text (and Fig. 3 therein), we combine trends 2 and 3 to be a single, faster response time phase. The lower case letters correspond to the lake phases in (A). 17

18 Figure DR11 Lake Eyre volume and area regression used as f( V )( f( V) = V ) in Equation (6). 18

19 Figure DR12 Evolution of Lake Eyre filling volumes for different constant inflow regimes. 36km 3 is equivalent to the estimated total inflows during 1974 (Bye and Will, 1989). Not shown are the asymptotic filling times for and inflow of 33km 3 /yr, which attains the maximum lake volume ( ~343 km 3 ) in ~83 years, and an inflow of 30km 3 /yr which can produce a maximum volume of 300km 3 in ~88 years. The alternating ephemeral flow conditions (36 : 0 km 3 /yr) become asymptotic between lake volumes of ~ km 3 after ~65 years. The small case letters correspond to the lake levels in Fig. DR10. 19

20 Figure DR13. Scatter plots of absolute standard error and relative standard-error uncertainty on the OSL (grey circles) and TL ages (open circles). Note that the two dating methods display no systematic differences; the combined arithmetic mean relative standard error is 10.0 %. Figure DR14. OSL age versus TL age for 20 duplicate samples for which both OSL and TL ages have been measured. OSL results include 13 multi-grain aliquot analyses (blue) and 7 single-grain aliquot analyses (orange), all with ±1σ errors, 95% confidence limits (grey lines), and the 1:1 correlation (dashed grey line). 20

21 Figure DR15. Frequency distribution histograms of dated bedload units using two different bin-widths: 5 kyr and 10 kyr, respectively (n=130). 21

22 Figure DR16. Frequency distribution of dated bedload units (n=130) presented as a normal kernel density estimate (blue line; 1 kyr bandwidth), plus rank-age plot with ±1σ errors (black dots with grey error bars). 22

23 Figure DR17. Deposition rate versus age (n=159), where deposition rate is computed as age divided by depth. All data are log-bin averaged, yielding the regression equation: y = 0.658x , expressed in power-law form as y = x

24 Figure DR18. Frequency distribution of dated bedload units (n=130) normalised to remove the Sadler effect bias, and presented as a normal kernel density estimate (blue line; 1 kyr bandwidth), plus rank-age plot with ±1σ errors (black dots with grey error bars). 24

25 Supplementary Tables Table DR1 Dose rate data derived from high-resolution gamma spectrometry. All radionuclide data are in Bq/kg. The total dose rate includes an internal alpha dose rate of 0.03 ± 0.01 Gy/ka. Sample 238 U 226 Ra 210 Pb 232 Th 40 k Beta dose rate (Gy/ka) Gamma dose rate (Gy/ka) Cosmic dose rate (Gy/ka( Total dose rate (Gy/ka) TR5_ ± ± ± ± ± ± ± ± ± 0.03 TR5_ ± ± ± ± ± ± ± ± ± 0.03 TR6_ ± ± ± ± ± ± ± ± ± 0.03 TR4_1 36 ± ± ± ± ± ± ± ± ± 0.05 TR1_ ± ± ± ± ± ± ± ± ± 0.04 TR1_ ± ± ± ± ± ± ± ± ± 0.03 TR1_2 31 ± ± ± ± ± ± ± ± ± 0.04 TR1_1 32 ± ± ± ± ± ± ± ± ± 0.05 TR9_ ± ± ± ± ± ± ± ± ± 0.04 TR9_3 51 ± ± ± ± ± ± ± ± ± 0.04 TR9_ ± ± ± ± ± ± ± ± ± 0.03 TR9_ ± ± ± ± ± ± ± ± ± 0.03 TR2_ ± ± ± ± ± ± ± ± ± 0.03 TR2_ ± ± ± ± ± ± ± ± ± 0.03 TR2_ ± ± ± ± ± ± ± ± ± 0.03

26 Table DR2 Dose recovery experiment results. R d PH 1 T d PH 2 grains Number of 40 Gy surrogate natural Measured/given dose ratio Overdispersion (%) 220 C 160 C ± ± C 160 C ± ± C 160 C ± ± C 220 C ± ± C 220 C ± ± Gy surrogate natural 260 C 160 C ± ± C 220 C ± ± 0 1 Regenerative dose preheat ( C) for 10 s. 2 Test dose preheat ( C) for 5 s. 26

27 Table DR3 Number of single grains measured, rejected due to poor OSL characteristics (following criteria described in supplementary text), and accepted for D e calculation. Sample TR5-2 TR5-1 TR6-2 TR1-4 TR1-3 TR1-2 TR1-1 TR9-4 TR9-3 TR9-2 TR9-1 TR4-1 TR2-4 TR2-3 TR2-2 number of grains measured Rejection Criteria Dim Inconsistent recycling ratio (40 Gy) Inconsistent recycling ratio (178 Gy) Inconsistent OSL-IR depletion ratio L0/Tx > LN/TN * Saturated Class D e consistent with 0 at 2sigma Number of grains accepted Overdispersion (%) 4 26±3 27±3 67±5 36±3 24±2 25±3 27±4 63±7 27±3 46±5 25±3 24±3 78±7 23±3 30±4 Weighted mean D e (Gy) 4 64±3 62±2 42±3 52±2 59±2 79±3 95±4 72±6 100±4 113±7 79±3 83±3 49±5 64±3 81±4 1 Dim grains include those rejected for having a Tn signal less than three times the Tn background and grains that did not produce a visible Tn decay curve. 2 Includes grains with a horizontal dose response curve and grains for which De<2*D0. 3 Grains statistically consistent with 0 were assumed to be contaminant/intrusive (i.e., not representative of deposition) and were rejected. 4 Overdispersion and weighted mean D e calculated using the central age model (Galbreith et al., 1999). 27

28 Table DR4 Finite mixture model results for each sample from Illusion Plains. The number of components (k) and overdispersion (OD) parameters that best fit the data were identified using the BIC and Llik values. The optimized model results are shown. Sample Number of components (k) OD D e (Gy) Proportion TR ± ± ± ± 0.03 TR ± ± ± ± 0.12 TR ± ± ± ± 0.04 TR ± ± ± ± 0.08 TR ± ± ± ± 0.04 TR ± ± ± ± 0.15 TR ± ± ± ± 0.09 TR ± ± ± ± 0.04 TR ± ± ± ± 0.04 TR ± ± ± ± 0.07 TR ± ± ± ± 0.04 TR ± ± ± ± 0.05 TR ± ± ± ± ± ± ± ± 0.03 TR ± ± ± ± 0.09 TR ± ± ± ± 0.12

29 Table DR5 Dose rate data, D e and age model data, and optical ages for samples from each beach ridge. Note that the bold age is that interpreted to be depositional (see supplementary text). Sample Elevation of sample (m- AHD) Dose rate (Gy/ka) OD CAM D e (gy) CAM age (ka) FMM OD FMM major D e (Gy) Proportion Age of major FMM population (ka) TR5_ ± ± ± ± ± ± 3.3 TR5_ ± ± ± ± TR6_ ± ± ± ± ± ± 3.6 TR4_ ± ± ± ± ± ± 2.7 TR1_ ± ± ± ± ± ± 3.0 TR1_ ± ± ± ± TR1_ ± ± ± ± ± ± 3.5 TR1_ ± ± ± ± ± ± 3.0 TR9_ ± ± ± ± ± ± 4.8 TR9_ ± ± ± ± ± ± 5.8 TR9_ ± ± ± ± ± ± 8.2 TR9_ ± ± ± ± ± ± 6.4 TR2_ ± ± ± ± ± ± 8.8 TR2_ ± ± ± ± TR2_ ± ± ± ± ± ±

30 Table DR6 Shoreline chronology for illusion plains showing results from homogeneity test (Galbreith, 2003) and weighted mean for each shoreline. Values for ū, G and P are from the homogeneity test described by (Galbreith, 2003). Sample Sample Sample age (ka) elevation (m- AHD) ū G P Weighted mean shoreline age (ka) TR4_ ± TR1_ ± ± 1.8 TR1_ ± TR1_ ± ± 2.1 TR5_ ± TR5_ ± TR6_ ± ± 1.9 TR9_ ± TR2_ ± ± 4.2 TR9_ ± TR2_ ± ± 3.8 TR9_ ± TR2_ ± ±

31 Table DR7 Age, depth and area characteristics for the modern and palaeo Lake Eyre along with hypothetical equilibrium inflow under four L N scenarios Age (ka) depth (m) Area (km 2 ) Volume Q1 (km 3 /yr) Q2 (km 3 /yr) Q3 (km 3 /yr) Q4 (km 3 /yr) (km 3 ) * * = 1974 maximum area, depth and volume. Modern and palaeo lake morphometric values taken from (Leon and Cohen, 2012) Q1, Q2, Q3, Q4, correspond to Q E estimates based on L N values of 1200, 1500, 1800, and 2100 (mm/yr) respectively (Figure DR9). This data is used in Figure 3 in the main paper. *Shoreline age and lake empty scenario from (Magee et al., 1995; Magee et al., 2004) 31

32 Table DR8. Compilation of 159 dated bedload units (125 TL, 34 OSL) in the Lake Eyre basin Site i Site ii Site iii Dating method a Lab code b Age b ka Age ±1σ error b ka Sample depth m Reference c Camel Flat site 16 OSL-MGA 16/ Hollands et al. (2006) Camel Flat site 2 OSL-MGA 2/ Hollands et al. (2006) Camel Flat site 2 OSL-MGA 2/ Hollands et al. (2006) Camel Flat site 3 OSL-MGA wm: 3/1.5, 3/ Hollands et al. (2006) Camel Flat site 9 TL W Hollands et al. (2006) Cooper Ck Mt Howitt Trench 1/1 TL wm: W2290, W Maroulis (2000) Cooper Ck Mt Howitt Trench 1/3 TL W Maroulis (2000) Cooper Ck Chookoo CH-2 trench TL wm: W2054, W Maroulis (2000) Cooper Ck Chookoo CH-3 trench TL W Maroulis (2000) Cooper Ck Chookoo CH-3 trench TL W Maroulis (2000) Cooper Ck Chookoo CH-D TL W Maroulis (2000) Cooper Ck Chookoo CH-D TL W Maroulis (2000) Cooper Ck Chookoo CH-G TL W Maroulis (2000) Cooper Ck Chookoo CH-G TL W Maroulis (2000) Cooper Ck Chookoo CH-M TL W Maroulis (2000) Cooper Ck Chookoo CH-S TL W Maroulis (2000) Cooper Ck Chookoo CH-S TL W Maroulis (2000) Cooper Ck Cullyamurra, Transect 1 AH1-1 TL W Coleman (2002) Cooper Ck Cullyamurra, Transect 1 AH2-1 TL W Coleman (2002) Cooper Ck Cullyamurra, Transect 1 AH2-1 TL W Coleman (2002) Cooper Ck Cullyamurra, Transect 1 AH2-1 TL W Coleman (2002) Cooper Ck Cullyamurra, Transect 1 AH3-1 TL W Coleman (2002) Cooper Ck Cullyamurra, Transect 1 AH4-1 TL W Coleman (2002) Cooper Ck Cullyamurra, Transect 2 AH8-3 TL W Bowman (2003) Cooper Ck Cullyamurra, Transect 2 AH9-3 TL W Bowman (2003) Cooper Ck Cullyamurra, Transect 2 AH9-3 TL W Bowman (2003) Cooper Ck Cullyamurra, Transect 3 AH3-2 TL W Bowman (2003) Cooper Ck Cullyamurra, Transect 3 AH3-2 TL W Bowman (2003) Cooper Ck Cullyamurra, Transect 3 AH5-2 TL wm: W3484, W Bowman (2003) Cooper Ck Cullyamurra, Transect 3 AH6-2 TL W Bowman (2003) Cooper Ck Cullyamurra, Transect 3 AH6-2 TL W Bowman (2003) Cooper Ck Durham Downs Rd F/P1 TL W Maroulis (2000) 32

33 Site i Site ii Site iii Dating method a Lab code b Age b ka Age ±1σ error b ka Sample depth m Reference c Cooper Ck Durham Downs Rd F/P6 TL W Maroulis (2000) Cooper Ck Durham dune DD1 TL W Maroulis (2000) Cooper Ck Durham dune DD1 TL W Maroulis (2000) Cooper Ck Durham dune DD3 TL W Maroulis (2000) Cooper Ck Durham dune DD3 TL W Maroulis (2000) Cooper Ck Gidgealpa fp transect Gidge2H1 TL W Cohen et al. (2010) Cooper Ck Gidgealpa fp transect Trench 1 OSL-SG GT1, re-w Cohen et al. (2010) Cooper Ck Gidgealpa fp transect Trench 2 OSL-SG GT Cohen et al. (2010) Cooper Ck Gidgealpa fp transect Trench 3 OSL-SG GT Cohen et al. (2010) Cooper Ck Gidgealpa fp transect Trench 5 OSL-SG GT Cohen et al. (2010) Cooper Ck Gidgealpa fp transect Trench 5 OSL-SG Cohen (unpub.) Cooper Ck Gidgealpa fp transect Trench 6 OSL-SG GT Cohen et al. (2010) Cooper Ck Goonbabinna wh T1 TL W Maroulis (2000) Cooper Ck Maapoo wh M/H2 TL W Nanson (unpub.) Cooper Ck Maapoo wh M/H3 TL W Nanson (unpub.) Cooper Ck Maapoo wh MW/H1 TL W Nanson (unpub.) Cooper Ck Maapoo wh MW/H1 TL W Nanson (unpub.) Cooper Ck Maapoo wh MW/H1 TL W Nanson (unpub.) Cooper Ck Naccowlah wh Exc. 1 TL wm: W551, W Rust & Nanson (1986) Cooper Ck Naccowlah wh Exc. 2 (TL8) TL W Rust & Nanson (1986) Cooper Ck Naccowlah wh Exc. 4 (TL7) TL W Rust & Nanson (1986) Cooper Ck Narberry wh NH1 TL W Maroulis (2000) Cooper Ck Narberry wh NH2 TL W Maroulis (2000) Cooper Ck Shire Rd SR1 TL W Maroulis (2000) Cooper Ck Shire Rd SR1 TL W Maroulis (2000) Cooper Ck Shire Rd SR2 TL W Maroulis (2000) Cooper Ck Shire Rd SR2 TL W Maroulis (2000) Cooper Ck Shire Rd SR2 TL W Maroulis (2000) Cooper Ck Shire Rd SR2 TL W Maroulis (2000) Cooper Ck Shire Rd SR2 TL W Maroulis (2000) Cooper Ck Shire Rd SR3 TL W Maroulis (2000) Cooper Ck Shire Rd SR3 TL W Maroulis (2000) Cooper Ck Shire Rd SR3 TL W Maroulis (2000) 33

34 Site i Site ii Site iii Dating method a Lab code b Age b ka Age ±1σ error b ka Sample depth m Reference c Cooper Ck Shire Rd SR3-T TL W Maroulis (2000) Cooper Ck Shire Rd SR4 TL W Maroulis (2000) Cooper Ck Shire Rd SR5 TL W Maroulis (2000) Cooper Ck Shire Rd SR5 TL W Maroulis (2000) Cooper Ck Shire Rd SR5-T TL W Maroulis (2000) Cooper Ck Shire Rd SR6 TL W Maroulis (2000) Cooper Ck Shire Rd SR8 TL W Maroulis (2000) Cooper Ck Scrubby Camp wh TL W Coleman (2002) Cooper Ck Aarpoo wh TL W Coleman (2002) Cooper Ck Marpoo wh TL W Coleman (2002) Cooper Ck N Dogbite L NDBL-1 TL W Coleman (2002) Cooper Ck E Dogbite L EDBL1-1 TL W Coleman (2002) Cooper Ck Turra fp transect DH2, TurraPC2-1 TL W Coleman (2002) Cooper Ck Turra fp transect DH2, TurraPC2-2 TL W Coleman (2002) Cooper Ck Turra fp transect DH3, TurraPC3-3 TL W Coleman (2002) Cooper Ck Turra fp transect DH3, TurraPC3-6 TL W Coleman (2002) Cooper Ck Turra fp transect DH3, TurraPC3-7 TL W Coleman (2002) Cooper Ck Turra fp transect DH4, TurraPC4-1 TL W Coleman (2002) Cooper Ck Wills grave AH1 TL W Coleman (2002) Cooper Ck Wills grave AH2 TL W Coleman (2002) Cooper Ck Wills grave AH3 TL W Coleman (2002) Cooper Ck Wills grave AH3 TL W Coleman (2002) Cooper Ck Wills grave AH3 TL W Coleman (2002) Cooper Ck Wills grave AH4 TL W Coleman (2002) Cooper Ck Tilcha wh right bank OSL-SG wm: rew2239, rew2467, rew2240, Nanson et al. (2008) rew2468, rew2469 Cooper Ck Malgoona bench mark right bank OSL-MGA Mal2_ Cohen (unpub.) Cooper Ck Malgoona bench mark right bank OSL-MGA Mal2_ Cohen (unpub.) Cooper Ck Malgoona wh right bank OSL-SG MalWH_Up Cohen (unpub.) Cooper Ck Malgoona wh right bank OSL-SG MalWH_Low Cohen (unpub.) Cooper Ck Cuttupirra wh gully OSL-MGA CG Cohen (unpub.) Cooper Ck Cuttupirra wh gully OSL-SG CG Cohen (unpub.) Cooper Ck Cuttupirra wh basal Katipiri sands OSL-SG CBK Cohen (unpub.) Cooper Ck Eli Hartig Soak Katapiri sands OSL-SG KCS Cohen (unpub.) 34

35 Site i Site ii Site iii Dating method a Lab code b Age b ka Age ±1σ error b ka Sample depth m Reference c Cooper Ck Cuttupirra wh TL W Callen & Nanson (1992) Cooper Ck Cuttupirra wh TL W Callen & Nanson (1992) Cooper Ck Eli Hartig Soak TL W Callen & Nanson (1992) Cooper Ck Pirranna wh TL wm: W813, W Callen & Nanson (1992) Cooper Ck S of Tilla Tilla wh TL W Callen & Nanson (1992) Cooper Ck S of Tilla Tilla wh TL W Callen & Nanson (1992) Cooper Ck Kutjitara W Bluff Gully section OSL-MGAMA wm: KWB 10, KWB Magee (1997) Cooper Ck Kutjitara W Bluff Lower Katipiri ch. OSL-MGAMA KWB Magee (1997) Cooper Ck Kutjitara W Bluff Main section, unit E OSL-MGAMA wm: KWB 6, KWB Magee (1997) Cooper Ck Kutjitara W Bluff Main section, unit I OSL-MGAMA KWB Magee (1997) Diamantina R Boulia Rd Crossing Exc-4 TL W Nanson et al. (1991) Diamantina R Boulia Rd Crossing Exc-5 TL W Nanson et al. (1991) Diamantina R Brighton Downs TL W Nanson et al. (1988) Finke R Finke channel bed TL wm: W1151, W Nanson et al. (1995) Finke R Finke FP 2 TL W Nanson et al. (1995) Finke R Finke FP 2 TL W Nanson et al. (1995) Finke R Finke FP 3 TL W Nanson et al. (1995) Neales R B TL W Croke et al. (1996) Neales R Fitchet overflow channel auger hole 2 TL W Croke et al. (1996) Neales R over Etadunna silcrete TL W Croke et al. (1996) Neales R Section A Warmakidyaboo wh TL W Croke et al. (1996) Neales R Section A Warmakidyaboo wh TL W Croke et al. (1996) Neales R Section B VB wh, profile 1 TL W Croke et al. (1996) Neales R Section B VB wh, profile 1 TL W Croke et al. (1996) Neales R Section C channel fill TL wm: W1498, W1499, W Croke et al. (1996) Neales R Section C TL W Croke et al. (1996) Neales R Section D TL W Croke et al. (1996) Neales R Section E Price wh TL W Croke et al. (1996) Neales R Section F TL wm: W1503, W Croke et al. (1996) Sandover R Ammaroo TL W Tooth (1997) Sandover R Ooratippra TL W Tooth (1997) Sandover R Ooratippra TL W Tooth (1997) Strzelecki Ck Merty Merty AH1 TL W Coleman (2002) 35

36 Site i Site ii Site iii Dating method a Lab code b Age b ka Age ±1σ error b ka Sample depth m Reference c Strzelecki Ck Merty Merty AH1 TL wm: W2251, W Coleman (2002) Strzelecki Ck Merty Merty AH2 TL wm: W3985, W Coleman (2002) Strzelecki Ck Merty Merty AH3 TL wm: W3987, W Coleman (2002) Strzelecki Ck Merty Merty AH4 TL W Coleman (2002) Strzelecki Ck Mudlalee AH1 TL W Coleman (2002) Strzelecki Ck Mudlalee AH1 TL W Coleman (2002) Strzelecki Ck Strzelecki Crossing Pit 2, 5.2 OSL-MGA Pit 2, Larsen (2012) Strzelecki Ck Strzelecki Crossing Pit 3, 4.2 OSL-MGA Pit 3, Larsen (2012) Strzelecki Ck Strzelecki Crossing Pit 4, 5.0 OSL-MGA Pit 4, Larsen (2012) Strzelecki Ck Strzelecki Crossing Pit 6, 3.7, 4.4 OSL-MGA wm: n= Larsen (2012) Strzelecki Ck Strzelecki Crossing Pit 7, 4.5 OSL-MGA Pit 7, Larsen (2012) Strzelecki Ck Strzelecki Crossing Pit 8a, 5.4 OSL-MGA Pit 8a, Larsen (2012) Thomson R Longreach Rb fp, H3 TL W Nanson et al. (1988) Thomson R Longreach Rb fp, H3 TL W Nanson et al. (1988) Thomson R Longreach Rb fp, H5 TL W Nanson et al. (1988) Warburton Ck Lookout Diprotodon section OSL-MGAMA LO Magee (1997) Warburton Ck Lookout Main section TL W Nanson et al. (1991) Warburton Ck Lookout Main section TL W Nanson et al. (1991) Warburton Ck Punkrakaderinna, wood site channel fill OSL-MGAMA Punkra Magee (1997) Warburton Ck Punkrakaderinna, wood site channel fill TL W Nanson (unpub.) Warburton Ck Punkrakaderinna, wood site cliff face section OSL-MGAMA Punkra Magee (1997) Warburton Ck Punkrakaderinna, wood site cliff face section TL W Nanson (unpub.) Warburton Ck Punkrakaderinna, wood site gully TL wm: W965, W Nanson (unpub.) Warburton Ck Toolapinna TL wm: W971, W Nanson (unpub.) Warburton Ck Whip well trench 1 OSL-MGAMA WW Magee (1997) Warburton Ck Whip well trench 2 OSL-MGAMA WW Magee (1997) Western R H3 TL W Nanson et al. (1988) Western R H4 TL W Nanson et al. (1988) a b c TL thermoluminescence, OSL-MGA optically stimulated luminescence, multiple-grain aliquot; OSL-MGAMA optically stimulated luminescence, multiple-grain aliquot, multiple-aliquot; OSL-SG optically stimulated luminescence, single-grain. wm denotes weighted mean age ± weighted mean error, where multiple samples are dated from a single sedimentary unit. reference in which the age is first cited. 36

37 References: Aitken, M.J., Thermoluminescence Dating. Academic Press, London. Aitken, M.J., An Introduction to Optical Dating: The Dating of Quaternary Sediments by the Use of Photonstimulated Luminescence. Oxford University Press, Oxford. Benson, L.V., Paillet, F.L., The use of total lake-surface area as an indicator of climatic change: Examples from the Lahontan basin. Quat. Res. 32(3) Bowler, J.M., et al., New ages for human occupation and climatic change at Lake Mungo, Australia. Nature 421, Bøtter-Jensen, L., Andersen, C.E., Duller, G.A.T., Murray, A.S., Developments in radiation, stimulation and observation facilities in luminescence measurements. Radiation Measurements 37, Bowman, H.H., The flow hydraulics of Cooper Creek through the Innamincka Dome. Unpublished BSc (Hons) thesis, School of Geosciences, University of Wollongong, Australia. Bye, J.A.T., Will, G. D., The hydrology of the filling in The Great Filling of Lake Eyre in C. W. Bonython, A.S.Fraser. Royal Geographical Society of Australasia, Adelaide, South Australian Branch Callen, R.A., Nanson, G.C., Discussion formation and age of dunes in the Lake Eyre depocentres. Geologische Rundschau 81, Cohen, T.J., Nanson G.C., Larsen, J.R., Jones, B.G., Price D.M., Coleman, M., Pietsch, T.J., Late Quaternary aeolian and fluvial interactions on the Cooper Creek fan and the association between linear and source-bordering dunes, Strzelecki Desert, Australia. Quaternary Science Reviews 29, Coleman, M., Alluvial, aeolian and lacustrine evidence for climatic and flow regime changes over the past 250 ka, Cooper Creek near Innamincka, South Australia. Unpublished PhD thesis, School of Geosciences, University of Wollongong. Croke, J., Magee, J., Price, D., Major episodes of Quaternary activity in the lower Neales River, northwest of Lake Eyre, central Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 124, Duller, G.A.T., Distinguishing quartz and feldspar in single grain luminescence measurements. Rad.Meas. 37, Galbraith, R.F., Roberts, R.G., Laslett, G.M., Yoshida, H. Olley, J.M., Optical dating of single and multiple grains of quartz from Jinmium rock shelter, northern Australia: Part I, experimental design and statistical models. Archaeometry 41, Galbraith, R.F., A simple homogeneity test for estimates of dose obtained using OSL. Ancient TL 21, Gliganic, L.A., Jacobs, Z., Roberts, R.G., Luminescence characteristics and dose distributions for quartz and feldspar grains from Mumba rockshelter, Tanzania. Archaeo. Anthro. Res. 4, Hollands, C.B. Nanson, G.C., Jones, B.G., Bristow, C.S., Price, D.M., Pietsch, T.J., Aeolian-fluvial interaction: evidence for Late Quaternary channel change and wind-rift linear dune formation in the northwestern Simpson Desert, Australia. Quaternary Science Reviews 25, Jacobs, Z., Duller, G.A.T., Wintle, A.G., Henshilwood, C.S., Extending the chronology of deposits at Blombos Cave, South Africa, back to 140 ka using optical dating of single and multiple grains of quartz. Journal of Hum. Evol. 51, Langbein, W.B., Salinity and hydrology of closed lakes. US Geol. Surv. Prof. Paper Series 412, 20. Larsen, J.R., Aspects of the contemporary and Quaternary hydrology of the Lake Eyre Basin, central Australia. Unpublished PhD Thesis, University of Wollongong, Australia. Leon, J.X., Cohen, T.J., An improved bathymetric model for the modern and palaeo Lake Eyre. Geomorph , Magee, J.W., Late Quaternary environments and palaeohydrology of Lake Eyre, arid central Australia, Unpublished PhD Thesis, Australian National University, Canberra. Magee, J.W., Bowler, J.M., Miller, G.H., Williams, D.L.G., Stratigraphy, sedimentology, chronology and palaeohydrology of Quaternary lacustrine deposits at Madigan Gulf, Lake Eyre, South Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 113,

38 Magee, J.W. Miller, G.H., Spooner, N.A. & Questiaux, D., Continuous 150 k.y. monsoon record from Lake Eyre, Australia: Insolation-forcing implications and unexpected Holocene failure. Geology 32, Maroulis, J.C., Stratigraphy and mid to late Quaternary chronology of the Cooper Creek floodplain, southwest Queensland, Australia. Unpublished PhD Thesis, University of Wollongong, Wollongong, 198 pp. Mason, I.M., Guzkowska, M. A. J., Rapley, C.G., Street-Perrott, F. A., The response of lake levels and areas to climatic change. Climatic Change 27(2), Mejdahl, V., Thermoluminescence dating: beta-dose attenuation in quartz grains. Archaeo. 21, Murray, A.S., R. Marten, R., Johnston, A., Martin, P., Analysis for naturally occurring radionuclides at environmental concentrations by gamma spectrometry. Journal of Radioanalytical and Nuclear Chemistry 115, Murray, A.S., Wintle, A.G., Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Rad. Meas. 32, Murray, A.S., Wintle, A.G., The single-aliquot regenerative-dose protocol: potential for improvements in reliability. Rad. Meas. 37, Nanson, G.C., Chen, X.Y., Price, D.M., Aeolian and fluvial evidence of changing climate and wind patterns during the past 100 ka in the western Simpson Desert, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 113, Nanson, G.C., Price, D.M., Jones, B.G., Maroulis, J.C., Coleman, M., Bowman, H., Cohen, T.J., Pietsch, T.J., Larsen, J.R., Alluvial evidence for major climate and flow regime changes during the middle and late Quaternary in eastern central Australia. Geomorphology 101, Nanson, G.C., Price, D.M., Short, S.A., Page, K.J., Nott, J.F., Major episodes of climate change in Australia over the last 300,000 years. In: R. Gillespie (Ed.), Quaternary Dating Workshop. Department of Biogeography and Geomorphology, Australian National University, Canberra, pp Nanson, G.C., Young, R.W., Price, D.M., Rust, B.R., Stratigraphy, sedimentology and late-quaternary chronology of the Channel Country of western Queensland. In: R. F. Warner (Ed.), Fluvial Geomorphology of Australia. Academic Press, Sydney, pp Prescott, J.R., Hutton, J.T., Cosmic ray contributions to dose rates for luminescence and ESR dating: large depths and long-term time variations. Rad.Meas. 23, Roberts, R.G., Galbraith, R.F., Olley, J.M., Yoshida, H., Laslett, G.M., Optical dating of single and multiple grains of quartz from Jinmium rock shelter, northern Australia: Part II, Results and implications. Archaeometry 41, Roberts, R.G., Galbraith, R.F., Yoshida, H., Laslett, G.M., Olley, J.M., Distinguishing dose populations in sediment mixtures: a test of single-grain optical dating procedures using mixtures of laboratory-dosed quartz. Rad. Meas. 32, Rust, B.R., Nanson, G.C., Contemporary and palaeochannel patterns and the late Quaternary stratigraphy of Cooper Creek southwest Queensland, Australia. Earth Surface Processes and Landforms 11, Sadler, P.M., Sediment accumulation rates and the completeness of stratigraphic sections. J. Geol. 89, Schumer, R., Jerolmack, D.J., Real and apparent changes in sediment deposition rates through time. J.Geophys. Res. Earth Surface 114, Stokes, S., Ingram, S., Aitken, M.J., Sirocko, F., Anderson, R., Leuschner, D., Alternative chronologies for late Quaternary (Last Interglacial-Holocene) deep sea sediment via optical dating of silt-size quartz. Quat. Sci. Rev. 22, Tooth S., The morphology, dynamics and Late Quaternary sedimentary history of ephemeral drainage systems on the northern plains of central Australia. Unpublished PhD Thesis, School of Geosciences, University of Wollongong, Australia. Williams, A.N., Smith, M.A., Turney, C.S.M., Cupper, M.L., AustArch1: A database of 14 C and luminescence ages from archaeological sites in the Australian arid zone. Yoshida, H., Roberts, R.G., Olley, J.M., Laslett, G.M., Galbraith, R.F., Extending the age range of optical dating using single supergrains of quartz. Rad. Meas. 32,