School of Earth Sciences - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemCalibration of the cosmogenic 21Ne exposure dating technique for application to Quaternary volcanic chronology( 2012)The use of terrestrial cosmogenic nuclides to investigate surface exposure histories continues to gain popularity due to the applicability of the technique to many different minerals and rock types. However, errors associated with cosmogenic exposure ages are typically either poorly specified or relatively high compared to other more established dating methods. This is dominantly due to large systematic uncertainties. This study focused on the stable cosmogenic nuclides 21Ne and 3He, and investigated their production in olivine and clinopyroxene from young (<150 ka) basalt flows. The primary aims of this study were to (i) improve the cosmogenic 21Ne dating technique for application to these minerals, (ii) develop in-house methods for dating young basalt flows by the 40Ar/39Ar step-heating technique, and (iii) inter-calibrate cosmogenic 21Ne surface exposure ages with corresponding 40Ar/39Ar eruption ages, in order to test available cosmogenic nuclide production rate scaling models. Alkali basalt flows (~30 – 130 ka) from three mid-latitude localities were studied: Mount Porndon (Victoria, Australia), Amboy Crater (Mojave Desert, California, USA), and the Nave flow, Mount Etna (Sicily, Italy). Helium and neon isotopic compositions of olivine (n = 13) and clinopyroxene (n = 7) separates from the above localities were determined from two-step heating experiments (900 ̊C and 1600 ̊C) using a VG5400 mass spectrometer. All samples showed evidence of variable degrees of helium loss and isotope mass fractionation, which compromised the calculation of cosmogenic 3He (3Hec) ages. Although minor neon loss was also apparent in most samples, it did not affect calculation of cosmogenic 21Ne (21Nec) ages. Overall, olivine separates yielded more consistent cosmogenic 21Ne ages than corresponding clinopyoxene separates extracted from the same samples. Crushing experiments on selected samples demonstrated an atmospheric composition for the 21Ne/20Ne ratio of the trapped neon component. Apparent weighted mean 21Nec exposure ages were calculated at: 64 ± 10 (2σ, Mount Porndon), 105 ± 23 ka (2σ, Amboy Crater), 29.0 ± 7.6 ka (2σ, Nave flow), respectively. Prior to analysing the very young (<150 ka) basalts from the selected calibration sites by the 40Ar/39Ar technique, studies of slightly older samples (300-600 ka) collected from the Newer Volcanic Province of southeastern Australia (Mount Rouse basalt (n = 4) and Mount Warrnambool basalt (n = 2)) were carried out on a VG3600 mass spectrometer in order to establish appropriate sample preparation procedures and analytical protocols. Multi-aliquot 40Ar/39Ar furnace step-heating analyses of whole-rock samples from several (n = 4) Mount Rouse basalt flows yielded weighted mean ages of: 309 ± 20 ka (2σ, NVP19), 382 ± 24 ka (2σ, NVP06); 301 ± 27 ka (2σ, NVP20) and 280 ± 19 ka (2σ, NVP21). Samples from Mount Warrnambool (n = 2) yielded weighted mean ages of: 547 ± 23 ka (2σ, NVP03) and 535 ± 27 ka (95% CI. NVP04). The internal concordance of results and agreement with previous K-Ar ages enabled validation of the experimental protocols adopted. Basalt samples younger than 150 ka from Mount Porndon, Amboy Crater and Mount Eccles were analysed on a new generation multi-collector ARGUSVI mass spectrometer via laser step-heating. This system yielded significantly more precise isotopic measurements than was possible using the older VG3600 system. Two to four groundmass aliquots were analysed per sample (n = 4), giving weighted mean 40Ar/39Ar ages of: 105 ± 22 ka (2σ; Mount Porndon, NVP14), 121 ± 8 ka (2σ; Amboy Crater, 08010), 92 ± 8 (2σ; Mount Eccles, NVP18)), 84 ± 13 (2σ; Mount Eccles, NVP12) and 293 ± 7 ka (95% CI; Mount Rouse, NVP21). Although lava surfaces showing apparent primary flow textures were sampled for cosmogenic dating, comparison of the cosmogenic 21Ne and 40Ar/39Ar ages indicates that significant erosion has occurred at the outcrops sampled for exposure dating from the Mount Porndon lava flows. Erosion modelling assuming an eruption age of 105 ± 22 ka (2σ, 40Ar/39Ar) suggests an erosion rate ranging from 2 - 7 mm/ka, depending on the value assumed for the scaled 21Nec production rate. As estimated 21Nec production rates for Mount Porndon are at the lower end of the range predicted by latitudinal production rate scaling models, it is possible that available scaling models may require further refinement for application to Southern Hemisphere localities. For the Amboy Crater flow, an erosion rate of 1 mm/ka is estimated, consistent with previous studies. In the case of the Nave flow, the agreement of the 21Nec exposure age for sample 08025 with the weighted mean K-Ar age (32 ± 8 ka (2σ)) of Blard et al. (2005) is interpreted as evidence for minimal erosion at this outcrop. However, the other two samples collected from adjacent outcrops on the Nave flow yielded slightly younger results, suggesting that erosion cannot be excluded for this flow. In order to refine estimates for 21Nec production rates in olivine and clinopyroxene, studies involving step-heating of large (several grams) mineral separates from young (ca. < 30 ka) basalt flow surfaces that have undergone negligible erosion and have independently constrained eruption ages are recommended. For the purposes of investigating muon-induced production of cosmogenic neon in olivine, 21Nec concentrations were studied in samples spanning an 11 m vertical profile through an Eocene basalt flow located in the Monaro Volcanic Province, NSW, Australia. A 40Ar/39Ar age of 55.52 ± 0.02 Ma (2σ) is calculated for this flow from step-heating (n = 4) experiments on whole-rock samples. It is demonstrated that excess 21Ne in the deepest samples can be accommodated solely by nucleogenic neon production, meaning that muogenic neon production in olivine is insignificant, even over million year time-scales. This justifies the assumption implicit in previous studies that 21Nec production in olivine (and likely clinoproxene) is purely the result of spallation reactions. Preliminary modelling based on 21Nec concentrations in the studied profile suggests an appreciably low average Cainozoic erosion rate of ca. 1 m/Ma.