The Identification of Local Dust Events from DustWatch PM10 Signals and Their Interaction with Climate and Land Use Variables in Regional New South Wales

Abstract

Wind erosion is a significant cause of land degradation in regional New South Wales. Reports on the state of wind erosion have primarily been focusing on more prominent dust storm events such as the 2009 Sydney “Red Dawn” event. Therefore, the establishment of the DustWatch network provides an unprecedented opportunity for the real-time monitoring of wind erosion in the region of New South Wales. The network provides high resolution dust and climate data which can be analyzed in relation to ground cover, rainfall, and wind patterns. This analysis provides a better understanding of the degree of natural erosion versus accelerated (i.e. anthropogenic induced) erosion. However, this process requires the separation of local dust events from regional dust events. To address this, the DustWatch team has been identifying local and regional dust events using methods of satellite imagery as well as physical surveys. This time-consuming process has accumulated a set of identified events. This research had utilized this set of identified events to derive a model which can predict the nature of dust events based on nine predictors fitted within the model. These predictors are derived from the characterization of the dust events identified by DustWatch, using CoDii’s temporal line graphs which illustrates the characteristics of dust, wind direction, and wind speed associated with each event. The developed logistic regression model was able to predict the nature of 548 events. Including the 183 events that were already identified by DustWatch, this research was able to utilise a total of 731 identified dust events. Spatio-temporal analysis and seasonal analysis were conducted on the set of 731 identified dust events. The analyses showed that the northern and southern regions’ spring-summer dust seasons and summer rainfall patterns, were generally supportive of McTainsh et al.'s (1998) dust storm seasonality. Large-scale climatic drivers such as the El Niño Southern Oscillation (ENSO) was a major cause of increased ground cover loss and higher magnitude of dust entrainment across the study region. The general pattern in wind direction for our study region supports the identification of the Simpson Desert-Channel Country, the Strzelecki Desert, and the Lake Eyre Basin as significant sources of dust to the region of Eastern Australia. Several catchments within New South Wales had been assessed to be vulnerable to wind erosion, which may explain the observation of high local dust entrainment in Moree. Show more