Influence of antecedent soil moisture conditions on the synoptic meteorology of the Black Saturday bushfire event in southeast Australia.

Kala, J., J.P. Evans and A.J. Pitman
Quarterly Journal of the Royal Meteorological Society, 141(693), 3118-3129, doi: 10.1002/qj.2596, 2015.


The dynamics and large-scale drivers of heat wave (HW) events in Australia are well documented. However, the influence of soil moisture in modulating HWs is largely unexplored. We focus here on a recent significant HW event in southeast Australia that preceded the Black Saturday bushfires (3–7 February 2009). During this period, the southeast of Australia experienced unprecedented warm conditions, which, in conjunction with high fuel load and mesoscale weather conditions, led to devastating bushfires. We examine how different initial soil moisture conditions with lead times of 5, 10, and 15 days prior to the event would have altered its overall dynamics at the continental scale. We show that at short lead times (5 days), the influence of perturbing soil moisture is mostly linear. Decreasing (increasing) soil moisture increases (decreases) maximum temperatures, associated with an intensification of the upper-level anticyclone. The effect of increasing soil moisture is more nonlinear than decreasing soil moisture with increasing lead time; namely, increasing soil moisture can also lead to an increase in maximum temperature over some parts of the domain, rather than a decrease everywhere. At lead times of up to 15 days, the imposed perturbation in soil moisture, mostly confined to the Tropics, is essentially lost such that the impact on maximum temperatures on the day of the event cannot be related to the sign of the imposed perturbation in soil moisture. Our results highlight the importance of accurate soil moisture estimates in capturing the intensity and spatial extent of HW events in southeast Australia, but only at relatively short lead times.

Key Figure

Figure 7. Difference in maximum 2 m temperature (TMAX) on 7 February 2009, between each experiment and the control (experiment − CNTL) started at (a) 5 days, (b) 10 days, and (c) 15 days lead time. Positive values indicate that the experiment was warmer than the control (CNTL) simulation.

UNSW    This page is maintaind by Jason Evans | Last updated 29 November 2013