CONFERENCE PROCEEDINGS

Australian Bushfire Conference, Albury, July 1999

Buttongrass moorland fire behaviour prediction system

Jon B. Marsden-Smedley¹, Tim Rudman¹, Adrian Pyrke¹ and Wendy R. Catchpole²

¹Parks and Wildlife Service, Department of Primary Industries, Water and Environment, 134 Macquarie Street, Hobart, 7000.

Ph: 03-6233 6767(w), 03-6233 3972 (fax); email: jonms@dpiwe.tas.gov.au

²School of Mathematics and Statistics, University College, University of New South Wales, Northcott Drive, Canberra, 2600.

Introduction

The Tasmanian Parks and Wildlife Service in association with Forestry Tasmania, the University of Tasmania, the Tasmania Fire Service and the Bureau of Meteorology have developed a fire behaviour prediction system for Tasmanian buttongrass moorlands.

Buttongrass (Gymnoschoenus sphaerocephalus) is a component of wet heaths, swamps, sedgelands and moorlands throughout south-eastern Australia. In Tasmania vegetation dominated by this species is normally referred to as buttongrass moorland (Figure 1).

Figure 1. Buttongrass moorland near Melaleuca Inlet in southwest Tasmania.

The buttongrass moorland fire behaviour prediction system consists of a series of components which predict fuel characteristics, fire behaviour and options for fire management:

Fuel characteristics

- total fuel load;

- dead fuel load;

Fire behaviour

- rate of fire spread;

- flame height;

- sustaining versus non-sustaining fires;

- buttongrass moorland fire danger.

Options for fire management

- prescriptions for prescribed burning;

- options for wildfire control.

Fuel characteristics

Time since the last fire and site productivity are used to predict the total fuel load and dead fuel load. Sites underlain by quartzite, granite, conglomerate and/or gravel derived from these geological types are low productivity while all other sites are medium productivity. The effects of site productivity and time since the last fire on total fuel loads and dead fuel loads are shown in Figure 2.

Figure 2. Total fuel load and dead fuel load.

Fire behaviour

The rate of head fire spread is predicted using: precipitation in the previous 48 hours; relative humidity; temperature; wind speed at 1.7 m and time since the last fire. The head fire flame height is predicted from the head fire spread rate, site productivity and time since the last fire. The Moorland Fire Danger Rating is designed to work in a similar manner to the Forest Fire Danger Rating and is predicted from the head fire spread rate. See Figures 2 and 4. Flank and back fire rates of spread and flame heights are also predicted.

Figure 3. Measuring a buttongrass moorland head fire.

Rosebery wildfire December 1992

Examples of the effect of different fuel moisture contents, times since fire and wind speeds on head fire spread rates are shown in Figure 3.

Figure 3. Effect of variation in fuel moisture, time since fire and wind speed on head fire spread rate. Note: Mf = fuel moisture (% ODW); age = time since the last fire (years).

Figure 4. Buttongrass moorland head fire.

North Strahan hazard-reduction burn October 1997

Operational fire management

Prescriptions have been developed for prescribed burning (Table 1). The fire management options for different levels of fire behaviour are also specified (Table 2).

Table 1. Prescriptions for conducting prescribed burning.

	Hazard-reduction	Ecosystem-management
Season	Apr to early May Sep to early Oct	Apr to June Aug to early Oct
Fire frequency, years	5 to 15	depends on objectives
Days since rain	>2	>2
Temperature, °C	10 to 20	5 to 20
Relative humidity, %	45 to 75	45 to 95
Wind speed, km hr^-1	3 to 10	<10
Soil dryness index	<10	<10
Overnight conditions if requiring burns to self extinguish
Synoptic situation	high pressure cell centred over Tasmania with low wind speeds and if possible, fog low productivity sites only
Site productivity	low productivity sites only
Soil dryness index	<5
Temperature, °C	<10
Relative humidity, %	>60
Wind speed, km hr^-1	<5
Dewfall, mm	>0.1

Table 2. Fire management options at different levels of fire behaviour.

MFDR	ROS	F_H	Fire characteristics
0-5 low	0-7	0-5	suitable for prescribed burning;
6-12 moderate	8-18	2-8	too intense for prescribed burning;
13-24 high	18-35	4-10	head fire control very difficult;
25-50 very high	36-70	5-14	head-fire direct attack not possible; very high risk of spot fires; fire-breaks >100m wide required.
51-100 extreme	>70	>8	fire control not possible.
MFDR = Moorland Fire Danger Rating; ROS = rate of fire spread, m min^-1; F_H = flame height, m.

References and further reading

Marsden-Smedley J. B. (1993).: Fuel characteristics and fire behaviour in Tasmanian buttongrass moorlands. Parks and Wildlife Service, Department of Environment and Land Management, Hobart, Tasmania.
Marsden-Smedley J. B. and Catchpole W. R. (1995a).: Fire behaviour modelling in Tasmanian buttongrass moorlands. I. Fuel modelling. International Journal of Wildland Fire 5: 203-214.
Marsden-Smedley J. B. and Catchpole W. R. (1995b).: Fire behaviour modelling in Tasmanian buttongrass moorlands. II. Fire behaviour. International Journal of Wildland Fire 5: 215-228.
Marsden-Smedley J. B. and Catchpole W. R. In press.: Fire behaviour modelling in Tasmanian buttongrass moorlands. III. Fuel moisture. International Journal of Wildland Fire.
Marsden-Smedley J. B., Catchpole W. R. and Pyrke A. In press.: Fire behaviour modelling in Tasmanian buttongrass moorlands. IV. Fire extinguishment. International Journal of Wildland Fire.
Marsden-Smedley J. B., Catchpole W. R. and Pyrke A. (1998).: Unbounded burning in Tasmanian buttongrass moorlands. Unpublished report prepared for the Tasmanian Fire Research Fund. Parks and Wildlife Service, Forestry Tasmania and Tasmania Fire Service.
Marsden-Smedley J. B., Rudman T., Pyrke A. and Catchpole W. R. In prep.: Buttongrass moorland fire behaviour prediction and management. Prepared for TasForests.

Published by School of Environmental & Information Sciences Charles Sturt University