Dust-storm frequencies, community attitudes, government policy and land management practices during three major droughts in New South Wales, Australia

Abstract

This study assessed whether dust-storm frequency during major droughts in New South Wales (NSW), Australia, has changed and what may have caused any change. The frequency of days with dust storms, i.e. when visibility is <1000 m, is presented for the dust storm year (July to June), with the maximum number of dust storms for three major droughts, namely, 2017/20, Millennial and World War II droughts. Community attitudes, government policy and land management practices have changed since the 1940s, and these factors were reviewed to determine whether they explain changes in dust-storm frequency. Two data sources were used: meteorological weather codes from the Australian Bureau of Meteorology and dust particulate matter <10 µm (PM10) from the DustWatch/Rural Air Quality Monitoring Network. The particulate-matter data were converted to dust-storm days (DSD) to create a yearly time series. The meteorological data records were coded as dust storms and required no modification. Results showed that 1944/45 was the dustiest year, with 4.4 times more DSD than in 2019/20 and 9.9 times more DSD than in 2009/10. One reason for the higher DSD in 2019/20 than in 2009/10 was the area protected from wind erosion by vegetation cover above 50%. In 2019/20, 69% of NSW was protected from wind erosion, compared with 79% in 2009/10. We suggest the primary reasons for lower DSD in 2019/20 and 2009/10 than in 1944/45 were community attitudes, government policy and land management practices; these, in combination, help maintain vegetation cover. Since the 1940s, the focus of land management has changed from ‘taming the land’ to ‘sustainably using the land’. Government policy in the 2000s is focused on supporting farming businesses and communities to manage and prepare so as to successfully manage drought. Land management practices that maintain ground cover are now widely practised.