Australia releases the most dust in the southern hemisphere, with global reaching affects such as (but not limited to) phytoplankton growth, air pollution, and soil enrichment. Australian dust could have very important biogeochemical effects including New Zealand farming and southern oceanic iron supply. The contribution of Australian dust to New Zealand is not well established. Measurements of dust deposition in peat bogs near NZ estimated that 30-90% was Australian (Marx 2009).
Global average yearly dust emission estimates vary widely from X(TODO) to 2150 Tg yr^-1 (Johnson 2012). Yearly dust emission variance can be extreme, for example north African dust emissions range from 400 to 2200 Tg yr^-1 (Huneeus 2011). In Australia around 100 Tg yr^-1 of dust are released anually(TODO CITE), however the estimate is rough and the large confidence intervals of dust emissions present problems when trying to determine direct radiative effect (DRE) of dust globally. The importance of the Eyre basin and recent increases in dust loading over the region has been shown using the Bureau of Meteorology’s (BOM’s) aerosol optical depth (AOD), Angstrom exponent (AE), and scattering coefficient data (Mitchell 2010).
Australian dust emission is more periodic than emissions from the northern hemisphere, with droughts and extreme weather systems combining to form massive dust storms like the 2009 ’red dawn’ dust storm over Sydney (Leys 2011). In 2002 a dust storm was estimated to have shifted almost 96 Tg over one 24 hour period (Shao 2007). This combined with relatively few long term Australian dust studies leads to lower confidence in simulated Australian dust properties.
One estimate of anthropogenic dust emissions in Australia is 75%, based on 30% or more of each grid box being used (Ginoux 2012).
GEOS-Chem modeled dust simulation is largely untested over Australia and New Zealand. We compare GEOS-Chem dust simulations with data from both AERONET and CoDii Dustwatch stations based in Australia. We examine Australian dust dynamics and seasonality as well as examine how well the model represents large dust events. We also look at how much of a role El Nino Southern Oscillation (ENSO) effects have on Australian dust sources and transport.
GEOS-Chem uses a dust entrainment and deposition scheme(DEAD) from (Zender 2003), using four dust size bins centred at 0.7, 1.4, 2.4, 4.5 microns.
We ran the GEOS-Chem model version 9.02 at 2x2.5 resolution, using offline GEOS-5 meteorological fields. DEAD dust mobilisation is based on surface wind speeds to the third power implemented by (Fairlie 2007). We only model emission, deposition, and transport of dust and carbon, which allows for fast runtime. Running the full chemical model to see if dust was affected by other tracers found only negligible differences (in the order of 10^-5 percent). With 2004 as the spin up year we stored monthly average columns of