Induced thermoluminescence as a method for dating recent volcanism: Hawaii County, Hawaii, USA

• Published in: Journal of volcanology and geothermal research Vol. 349; pp. 74 - 82
• Main Authors: Sears, Derek W.G., Sears, Hazel, Sehlke, Alexander, Hughes, Scott S.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.01.2018
• Subjects: Basalts; Crystallization; Hawaii; Thermoluminescence; Volcanoes; Hawaii; Volcanoes; Basalts; Crystallization; Thermoluminescence
• ISSN: 0377-0273; 1872-6097
• DOI: 10.1016/j.jvolgeores.2017.09.022

We have measured the induced thermoluminescence (TL) properties of fifteen samples of basalts collected from the Big Island of Hawaii in order to continue our investigation into the possible utility of this technique as a chronometer. Previous studies of basalts from Idaho have suggested the induced TL of basalts increases with age. Meteorite data suggest two possible explanations for this observation which are that (1) the initial glassy or amorphous phases crystalize with time to produce feldspar, the mineral producing the TL signal, and (2) feldspars lose Fe as they equilibrate and since Fe is a quencher of TL this would cause an increase in TL. The old basalts from Kohala (>100ka), which are mostly alkali basalts, have TL sensitivities 10–100 times higher than the much younger tholeiites from Kilauea and Mauna Loa (<50ka). The thermoluminescence of feldspars is strongly dependent on composition and when this is corrected for, using literature data, the slope of the regression line for the plot of log TL sensitivity against historic or radiometric age for the Hawaii basalts is within 2 sigma of the regression line for the analogous plot for the Idaho basalts, although the Hawaii line is much shallower (0.0015±0.0012 for Hawaii cf. 0.0039±−0.0014 for Idaho, 2σ uncertainties). However, the intercepts are significantly different (0.78±0.18 for Hawaii cf. −0.079±0.28 for Idaho, 2σ uncertainties). These results suggest that TL sensitivity has the potential to be a means of dating volcanism in the 0–800ka range, although the scatter in the data – especially for the <50ka samples – needs to be understood, and a means found for its removal, before the technique has the possibility of being practically useful. •The compositionally-corrected induced thermoluminescence values for a suite of basalts from Big Island of Hawaii show a weak correlation with age although the TL-age slope for the Hawaii samples is shallower than similar data for basalts from Idaho.•This difference is slope is due to the young (<50 ka) basalts which show a very similar range at both sites but while the Hawaii samples are skewed to higher values, the Idaho samples are skewed to lower values.•Understanding the spread of induced TL values for <50 ka basalts is key to developing this technique as a chronometer.