@conference {bnh-8275, title = {Long-range fire weather predictions developed and service established as new capability for Australia}, booktitle = {AFAC21}, year = {2021}, month = {10/2021}, publisher = {AFAC}, organization = {AFAC}, address = {Online}, abstract = {

Factors providing long-range predictability for this system include the pre-existing moisture content of vegetation (using agricultural drought measures), large-scale modes of variability (e.g., El Nino), sudden stratospheric warmings (e.g., polar vortex variations contributed to the severity of the 2019/20 summer) and long-term climate change.

The guidance products were designed with a user-driven approach, based on strong end-user engagement throughout the project. These long-range predictions are part of broader efforts to deliver seamless guidance over a wide range of time scales. Fire weather predictions and data are now available in a consistent form for long-range predictions out to several months ahead, historical records back to 1950 and future climate projections throughout this century. A seamless service across different time scales is intended to enhance planning capabilities from short to long time-scales, leading to enhanced resilience and disaster risk reduction for natural hazards. These fire weather outlooks have been described as a step change in improved capability, developed through user engagement, including for supporting risk reduction in prescribed burning and operational planning requirements.

}, keywords = {Capability, Fire, predictions, weather}, url = {https://www.afac.com.au/events/proceedings/06-10-21/article/long-range-fire-weather-predictions-developed-and-service-established-as-new-capability-for-australia}, author = {Dowdy, Andrew J} } @article {bnh-6801, title = {Improved predictions of severe weather to reduce community impact}, number = {556}, year = {2020}, month = {03/2020}, institution = {Bushfire and Natural Hazards CRC}, address = {Melbourne}, abstract = {

The project has completed the second year of its refresh and is progressing with some challenges. The first 3{\textonehalf} years featured a strong focus on developing underpinning science in some important areas, notably the transport of embers by bushfire plumes, and understanding the origin of the moisture when pyrocumulus clouds form {\textendash} is it from the atmosphere, or from the fuel? We also conducted significant case studies, including an ensemble-based study of a severe east coast low, an analysis of the meteorology of the State Mine Fire of October 2013, and a study into tropical cyclone secondary eyewall formation.

This second phase of the project pivots towards utilization. We will not neglect investigations into the underpinning science, or case studies, but we are looking to turn our previous discoveries and knowledge gained into useful products.

The first of these is a new thermodynamic tool for predicting the potential for pyrocumulonimbus (pyroCb) formation, the Pyrocumulonimbus Firepower Threshold, or PFT. This parameter depends on the atmospheric state, both the wind strength within the boundary layer and the full thermodynamic profile and aims to predict the minimum fire power (in GW) that is necessary to trigger deep convection. In favourable conditions, a fire of a few GW suffices, and at unfavourable times, 1000 GW would be insufficient. We have found that there is substantial variability in the atmospheric favourability in time and space, and that observed pyroCb often form during a narrow window of opportunity. We are presently preparing for a major utilization exercise, a near-real time trial of the tool over the coming summer. We expect that the results from this trial will help to finalize the development of the tool, and ready it for operations.

Our case study of the tornado outbreak in South Australia that triggered events which cut power to the entire state showed that the tornado cluster was predictable by modern high-resolution numerical weather prediction (NWP) {\textendash} indeed, the deterministic model predicted the precise location of one of the observed tornadoes. Encouraged by this success, we ran a 6-member ensemble simulation of the event, which showed that while the probability of tornadoes was well over 50\%, the level and location of the activity was less certain. To assist with analysis of this event, we also developed a novel diagnostic of tornadic precursors in NWP systems. A journal article is almost ready for submission on this work.

A second case study is analyzing the Tathra bushfire on the south coast of New South Wales, which led to the destruction of over 60 houses. Although occurring in strong northwesterly winds in the lee of the ranges, mountain wave activity appears to have been modest at the time of the greatest fire activity and probably not the major cause of the disaster. Rather, the extreme fire behavior appears to have been driven by strong, very unsteady and fluctuating winds, caused by roll-like circulations in the flow off the ranges, with the interaction between this flow and a later wind change also of interest. The weak role of mountain waves in this event illustrates the value of performing detailed case studies, as at first glance, several of the ingredients for a mountain wave event were present, and indeed active earlier in the day.

Unfortunately, our development of a parameterization of long-range ember transport has made little progress this year, due to the unanticipated difficulty of the task and the illness of a staff member. We plan to resume work on that shortly.

In recognition of the delays on the ember transport parameterization, and also of new opportunities that have arisen in pyrocumulus prediction, we have negotiated an altered project schedule with the BNHCRC.

}, keywords = {community impact, predictions, Severe Weather}, issn = {556}, author = {Jeffrey Kepert and KJ Tory and Dragana Zovko-Rajak and David Wilke and Serena Schroeter} } @article {bnh-7030, title = {Model predictions for fuel reduction burning of eucalypt open forest in the greater Blue Mountains region}, number = {587}, year = {2020}, month = {07/2020}, institution = {Bushfire and Natural Hazards CRC}, address = {Melbourne}, abstract = {

Prescribed burns are a land management tool used for reducing fuel loads in terrestrial ecosystems. Under extended drier, hotter weather conditions they might be used increasingly and more widely to help manage risk of wildfire and subsequent damage to life, property and natural assets. They also represent a form of disturbance to ecosystems, including their biodiversity and biogeochemistry. From a biogeochemistry perspective, we apply the FullCAM carbon accounting model to eucalypt open forest sites in the greater Blue Mountains region that underwent prescribed burns and fieldwork campaigns in 2019. Field data were used to derive values and estimates that guided model calibration and helped to explore the suitability of FullCAM for simulating the effect of prescribed burning on this ecosystem type. The diameter at breast height of overstorey and understorey trees, leaf area index and surface litter fractions were key measurements for estimating production, allocation, turnover (litter input to surface debris) and breakdown (output from surface debris) of carbon pools of forest components and hence, for calibrating FullCAM. Measurements for paired burnt/unburnt plots were key to estimating loss of carbon from forest component pools to the atmosphere due to prescribed fire. Simulation of unburnt forest component pools were reasonable as a calibration, although improvements in simulating fractions of surface litter would probably improve simulations of the effect of prescribed fire on forest component pools. Recommendations related to collection of field data and to model structure are made to improve alignment between model-data comparisons.

}, keywords = {Blue Mountains, eucalypt, Fuel reduction burning, modelling, open forest, predictions}, issn = {587}, author = {David Pepper and Bell, Tina and Malcolm Possell and Danica Parnell} }