Origins and sources of atmospheric precipitation from australia: chlorine-36 and major-element chemistry

Abstract

Temporal and spatial variations of major-element and 36Cl chemistry in rainfall across Australia have been assessed. Bulk precipitation samples were collected from two arrays over two years at three-monthly intervals: the WE array (10 sites) extended in a west to east direction from the coast of Western Australia south of Geraldton, inland to Warburton in Central Australia, and the SN array (8 sites), extended in a south to north direction from Port Lincoln in South Australia to Kakadu in the Northern Territory. The major-element chemistry shows that the main influence en the composition of precipitation in remote areas of Australia is mixing between seawater and continental sources. At most sites along the two arrays it is difficult to distinguish between the separate end-members of this source, except at coastal localities where seawater dominates the chemistry of precipitation. However, the influence of seawater is also evident at non-coastal sites in association with favourable synoptic conditions, such as cold frontal activity in south and western Australia during winter, and monsoonal activity in northern Australia during summer. The continentally-derived end-member is most likely composed of resuspended soil/dust material, including salt-lake and calcareous dune components. In the south of the SN array where agriculture is intense this continental source variably includes a fertiliser component. The chemistry of precipitation across Australia is also affected by an acid-base balance factor, the components of which are derived from natural sources such as biogenic emissions, biomass burning and lightning flash production. The nature of the collection program (i.e. samples are exposed to the atmosphere from the time of deposition to the time of sample retrieval) means biodegradation is also evident in the collected sample chemistry. Chlorine-36 is a cosmogenic isotope with a half-life of 301,000 years. This time frame, combined with the hydrophilic nature of Cl, makes 36Cl useful as a hydrological tracer. The use of 36Cl as a hydrological tracer however, relies on predicted models of 36Cl and stable Cl fallout to calculate 36CJ/CJ ratios for recharge to hydrological systems. The results from this investigation agree with the general shape of the latitude-dependent theoretical 36Cl fallout curve of Lal and Peters (1967), but suggests that the curve underestimates the rate of fallout. A revised mean fallout for the southern hemisphere of 15.4 36CI atomsfm2/s is suggested, and long-term average predictions of 36CJ fallout rates used to predict the input ratios of iv 36CIJC1 in hydrological investigations should be increased by a factor of 1.4 for the southern hemisphere. Further, while stable Cl concentrations in precipitation display a general exponential decrease with distance from the coast, the nature of this relationship is geographically variable, and Cl concentrations in precipitation should be investigated for each study by local direct measurements, a process that is simple and inexpensive. The mean 36Cl fallout for the southern hemisphere, calculated from this work is three times lower than has been measured for precipitation in the northern hemisphere. The lower southern hemisphere fallout rates reflect the lower rates of transfer of stratospheric air to the troposphere in the southern hemisphere, which results from the less dynamic nature of the lower stratosphere in the southern hemisphere. The mean global 36Cl fallout that incorporates measurements from the northern hemisphere with the results of this work is calculated to be 25-35 atomsfm2fs, 2-3 times greater than predicted by Lal and Peters (1967). This suggests that the cross-section for the cosmic-ray production of 36Cl may be underestimated in their paper. This work supports the use of 36Cl as a tracer of atmospheric processes. Is production primarily in the stratosphere suggests that it may trace stratospheric-tropospheric exchange. Seasonal variations in 36Cl fallouts and 36ClJCl show high ratios and fallouts during spring, and at some localities, during summer (i.e. the north of the SN array). The increased spring 36Cl fallouts are attributed to increased transfer of stratospheric 36Cl to the troposphere that occurs as the tropopause height increases during the warmer months. High fallouts during summer in the north of the SN array may be attributed to the direct entrainment of stratospheric air into cumulus clouds during the monsoonal convection. Chlorine-36 exists in the stratosphere predominantly as HCl gas (Wahlen et al 1991). The correlation between 36CJ and N03 and the lack of any relationship between 36Cl, stable Cl and Na concentrations (the latter being entrained as aerosols), suggest that 36Cl is scavenged from the atmosphere as a gas rather than an aerosol phase.