Inter-comparison of cosmogenic in-situ He, Ne and Cl at low latitude along an altitude transect on the SE slope of Kilimanjaro volcano (3°S, Tanzania)

Because the intensity and energy spectrum of the cosmic ray flux are affected by atmospheric depth and geomagnetic-field strength, cosmogenic nuclide production rates increase considerably with altitude and to a lesser degree with latitude. The scaling methods used to account for spatial variability in production rates assume that all cosmogenic nuclides have the same altitude dependence. In this study we evaluate whether the production rates of cosmogenic ³⁶Cl, ³He and ²¹Ne change differently with altitude, which is plausible due to the different threshold energies of their production reactions. If so, nuclide-specific scaling factors would be required.Concentrations of the three cosmogenic nuclides were determined in mafic phenocrysts over an altitude transect between 1000 and 4300 m at Kilimanjaro volcano (3°S). Altitude dependence of relative production rates was assessed in two ways: by determination of concentration ratios and by calculation of apparent exposure age ratios for all nuclide pairs. The latter accounts for characteristics of ³⁶Cl that the stable nuclides ³He and ²¹Ne do not possess (radioactive decay, high sensitivity to mineral composition and significant contributions from production reactions other than spallation). All ratios overlap within error over the entire transect, and altitudinal variation in relative production rates is not therefore evident. This suggests that nuclide-specific scaling factors are not required for the studied nuclides at this low-latitude location. However, because previous studies have documented anomalous altitude-dependent variations in ³He production at mid-latitude sites, the effect of latitude on cross-calibrations should be further evaluated.We determined cosmogenic ²¹Ne/³He concentration ratios of 0.1864 ± 0.0085 in pyroxenes and 0.377 ± 0.018 in olivines, agreeing with those reported in previous studies.Despite the absence of independently determined ages for the studied lava surfaces, the consistency in the dataset should enable progress to be made in the determination of the production rates of all three nuclides as soon as the production rate of one of the nuclides has been accurately defined.To our knowledge this is the first time that ³⁶Cl has been measured in pyroxene. The Cl extraction method was validated by measuring ³⁶Cl in co-existing plagioclase phenocrysts in one of the samples.