Introduction

This document summarizes the characterization of Australian tropical aerosol carried out using a combination of sun photometer measurements and data obtained from the Total Ozone Mapping Spectrometer (TOMS).

The first section presents sun photometer measurements of aerosol optical depth obtained from two sites, Jabiru, N.T., and Lake Argyle, W.A. The second section examines the correlation between these measurements and the aerosol index (AI) obtained from TOMS. The derived relationship is then used to map aerosol loading over the Australian tropics during the period July 1996 to December 2001.

Part I: Sun photometer measurements

The two sun photometers operated in Northern Australia form part of the CSIRO Aerosol Ground Station Network (AGSNet), which is affiliated with AERONET, a global network of sun photometer stations coordinated by NASA Goddard Space Flight Center. Further details of the rationale of AGSNet, including a technical description of the instruments used, and intercalibration with Australian Bureau of Meteorology (BoM) sun photometers, may be found in Mitchell and Forgan (2002). This study suggests that the intrinsic accuracy of the CE318 and Carter Scott SPO1A sun photometer operated by the BoM is better than 0.007 in total optical depth at all wavelengths considered.

This document presents data from two AGSNet stations. The first, at Lake Argyle (16.1667 S, 128.7151 E) was installed in May 1999. The second, at Jabiru (12.6607 S, 132.8931 E) was installed in June 2000. AGSNet stations are equipped with Cimel CE318 sun photometers, which are supplied in two versions. Both CE318-1 (standard) and CE318-2 (polarized) were used in this study. The CE318-1 has aerosol channels at 340, 380, 440, 500, 670, 870 and 1020 nm, plus a water vapour channel at 936 nm. The CE318-2 substitutes three polarized channels at 870 nm for the 340, 380 and 500 nm channels of the CE318-1.

Lake Argyle
The time series of daily mean aerosol optical depth at Lake Argyle is shown in Figure 1. From May to August 1999 a CE318-1 instrument was deployed at the site, after which it was replaced by a CE318-2. Figure 2 shows the time series of aerosol optical depth at 440 nm and the Angstrom coefficient defined over the wavelength interval 440 to 870 nm as:

Although aerosol optical depth is more usually quoted at 500 or 550 nm, 440 nm is adopted here since it is the nearest channel to 500 nm common to both the CE318-1 and CE318-2. Both aerosol optical depth and Angstrom coefficient show seasonality, with marked periods of low Angstrom coefficient during the wet season (December-March) caused by the dominance of large particles, while periods of high Angstrom coefficient are associated with the peak of the burning season in August-October, when small smoke particles dominate. It is possible that some of the low Angstrom coefficients during the wet season are due to thin, spatially uniform cirrus cloud, for which screening is problematic.

The monthly mean and standard deviation of 440 nm aerosol optical depth appear in Figure 3. More information regarding the seasonal cycle appears in Figure 4, where three year's data is overlaid. The systematic increase in aerosol burden from a minimum in June to a maximum at the height of the burning season in September-October is clearly evident.

Jabiru
The time series of daily mean aerosol optical depth at Jabiru is shown in Figure 5. A CE318-1 instrument was deployed throughout. Figure 6 shows the time series of aerosol optical depth at 440 nm and the Angstrom coefficient, and the seasonal cycle discussed above is clearly evident. Figure 7 shows the monthly mean and standard deviation of the 440 nm aerosol optical depth at Jabiru, both years of which are overlaid in Figure 8. While the general shape of the seasonal cycle is evident in both years, 2001 had substantially higher aerosol loadings than 2000 during the burning season (September to November).

Comparison of the two sites
Although separated by over 800 km, the seasonal cycles of aerosol at Jabiru and Lake Argyle are remarkably coherent. Figure 9 compares monthly mean and standard deviation at the two sites during 2000, while Figure 10 repeats this for 2001. The complete data set for both sites is plotted in Figure 11. From this it can be seen that the aerosol optical depth minimum between May and June is well-characterized in that there is close agreement between all years and sites, and the standard deviations are small and consistent. The peak aerosol loading occurs in September, with a mean optical depth at 440 nm of typically 0.4-0.5 and a standard deviation around 0.1-0.15.

Data tables
Tables of monthly mean aerosol optical depth, Angstrom coefficient, and column water vapour are presented below. The column identifiers are as follows:

Txxxx : Aerosol optical depth at wavelength xxxx nm;
WV936 : Column water vapour in cm derived from the 936 nm channel;
A4487 : Angstrom coefficient derived from the 440 and 870 nm channels (see above).

Lake Argyle:	1999-05 : 1999-08
	1999-10 : 2001-10
	2002-04 : 2002-12
Jabiru:	2000-06 : 2002-12

Part 2: Validation of TOMS data and generation of aerosol maps

References

Hermann, J.R., Bhartia, Torres, 0., Hsu, C., Seftor, C. and Celarier, E.,
1997: Global distribution of UV-absorbing aerosols from Nimbus 7/TOMS data.
J. Geophys. Res., 102, 16911-16922.

Mitchell, R.M. and Forgan, B.W., 2003: Aerosol measurement in the Australian
Outback: Intercomparison of sun photometers. J. Atmos. Ocean. Tech., 20, 54-66.

Torres, 0., Bhartia, P.K., Herman, J.R., Sinyuk, A., Ginoux, P. and Holben, B.,
2002: A Long-term record of aerosol optical depth from TOMS observations and
comparison to AERONET measurements. J. Atmos. Sci., 59, 398-413.