Fire frequency at Gamba grass sites in NT, Australia
ENVT 4408 GIS Automation and Programming
Introduction
Gamba grass (Andropogon gayanus) was first introduced to northern Australia in 1931 as a pasture species in order to improve cattle production (Department of Agriculture and Fisheries, 2016). It is a dense grass that grows up to four meters high (Figure 1). It adapts well to extreme conditions such as drought, fire and low nutrient soils that characterize the Top End, out-competing the native grass species (Department of Agriculture and Fisheries, 2016).
Fig. 1. Gamba grass in the Northern Territory retrieved from: http://www.theguardian.com
It’s a highly invasive weed currently affecting up to 15,000 km2 of the Northern Territory but having the potential to extend this to a 25 times greater area, 380,000 km2 (Northern Territory Government, 2009). Gamba grass has been listed as a Weed of National Significance since 2012, acknowledging its potential threat to northern Australia (Department of Environment and Energy, 2012).
Map 1. Gamba grass distribution in the northern NT, Australia
As a result of the climate, frequent, low intensity fires are natural part of the landscape in northern Australia, with certain areas having been burnt 10-19 times in a 20 year-long period (Map 1). However, in invaded areas fire regime has completely changed, as Gamba grass grows substantially higher than native weed species, it increases the fuel load from the average 6 tonnes/hectare up to 30 tonnes/hectare, generating much greater amounts of biomass and hence greater fire intensity (Setterfield et al., 2010). Moreover, it not only increases fire intensity but changes fire type from surface fire regimes to passive crown fires, increasing the vertical distribution of radiant heat (Setterfield et al., 2010). These changes to fire regime have devastating consequences. Although plants and animals are well adapted to low intensity fires, increased intensity fires pose threat to the biodiversity of the Australian savannas, having the potential to completely change the landscape from woodlands to grasslands (Rossiter et al., 2003). With crown fires not being part of the regular fire regime of the north because of the scattered occurrence of trees, crown fires have disastrous effects. The new fire regime is more similar to the fire regime in South Australia where instead of frequent, low-intensity fires high-intensity fires occur once or twice in a 28-year long period (Australia State of the Environment, 2016). However, there is a great difference, namely that in northern Australia these intense fires occur every year because Gamba grass provides the appropriate amount of fuel load annually, hence there is no time to recover. Plants and animals have the capacity to survive solely one or two of these intense fire events.
Map 2. The map shows the yearly fire frequency in the Northern Territory from 2000-2018
Aims
It is of interest to assess the distribution and spreading patterns of Gamba grass in the northern part of NT Australia because that has crucial implications to fire management. Gamba grass provides a great amount of fuel load and hence it increases fire frequency. The main aim of this study is to the assess how frequently Gamba grass sites burn, using fire frequency from 2000-2018 and Gamba grass distribution. Moreover, to assess Gamba grass distribution in close proximity to the main roads in the study area.
Methods
Study area
In this project I am going to focus on the tropical north of NT an area close to Darwin where Gamba grass was first introduced. The first paddocks of Gamba grass were planted in the Coomalie region, from where it has started to spread rapidly to adjacent areas (Rossiter et al., 2009).
Northern Australia has tropical monsoon climate, which is characterised by a wet season (November-April) and a dry season (May-October). Air temperature in consistently high through the year with mean maximum temperature of 33 °C.
All annual rainfall occurs in the wet season, hence in the dry season bushfires are extremely frequent: there is no moisture, fuel is provided by the grasses and the ignition source is usually lightning as in Darwin lightning storms are very common. The vegetation mostly consists of continuous grasslands and shrubs, with trees being fairly scattered, hence tree crowns are generally unconnected. The woodland is dominated by Eucalyptus miniate and E. tetrodonta and the grasslands are dominated by native perennial species (Setterfield et al., 2013). In the area, pastoralism is a common form of land use which is a low input management system.
Map 3. Study area: northern part of NT, Australia
Spatial analysis using GIS
Spatial analysis was used to assess the fire frequency at Gamba grass sites in the study area. The fire frequency data was in one data set that contained the number of times fire occurred from 2000-2018 in the Northern Territory. The use of automation was crucial in this project. After data pre-processing, a feature selection iterator model was used to create twenty different shape files each containing the areas of the individual fire frequencies from 0-19. A loop statement was used to clip each fire frequency to the Gamba grass distribution resulting in the fire frequency (0-19) at Gamba grass sites. Lastly, the buffer tool was used with two buffer distances.
The project tool was used to project all data to the same projected coordinate system (GDA 94 Australian Albers). The clip tool was used to clip major roads to the study area. Twenty different shapefiles, one of each fire frequency were created using a feature selection iterator model (Model 1). Automation was crucial as creating twenty shape files manually (Select tool and Export data) would have been ineffective and time-consuming. The input was the fire frequency data, with the fire frequency feature as grouping feature. All the individual fire frequency values were collected and exported to different shapefiles in the same geodatabase.
Model 1. Feature selection iterator
The next step was to use a loop (Python IDLE) and clip the fire frequencies to the Gamba grass distribution. The twenty fire frequency shapefiles were in the same geodatabase.
Fig. 2. Loop used for clip
Lastly the buffer tool was used with buffering distances 1 km and 5 km to buffer the main roads in the study area and assess the characteristics of Gamba grass sites in close proximity to roads.
Fig. 3. Loop used for buffer
Results
The results of the spatial analysis show the fire frequency at all Gamba grass sites in the study area and the high fire frequency Gamba grass sites in relation to the buffered main roads.
The low fire frequency Gamba sites (fire frequency 0-4) are concentrated in close proximity to major towns i.e. Darwin and Katherine.
Map 4. Gamba grass sites that burnt 0-4 times from 2000-2018.
The Gamba grass sites with fire frequencies 5-9 are more spread out in the study area and are not as concentrated around major towns as the lowest fire frequency Gamba grass sites. There are less Gamba grass sites that burnt 5-9 times than 0-4 times from 2000-2019.
Map 5. Gamba grass sites that burnt 5-9 times from 2000-2018.
The Gamba grass sites with fire frequencies 10-14 seem to be concentrated around the major roads. The spatial extent of Gamba grass sites that burnt 10-14 times seem to be greater than the spatial extent of Gamba grass sites that burnt 5-9 times from 2000-2018.
Map 6. Gamba grass sites that burnt 10-14 times from 2000-2018.
The Gamba grass sites with fire frequencies 15-19 seem to be located in four greater clusters in the study area. These Gamba grass sites are not in close proximity to the major towns i.e. Darwin and Katherine. The spatial extent of Gamba grass sites that burnt 10-14 and Gamba grass sites that burnt 15-19 times is similar, however, their spatial pattern is different.
Map 7. Gamba grass sites that burnt 15-19 times from 2000-2018.
The map with the road buffers show that the higher fire frequency Gamba grass sites are concentrated around the main roads in the study site. The majority of high fire frequency Gamba grass sites are in the 5 km buffer of the main roads. Gamba grass sites with fire frequencies 10-14 are frequent in the 1 km buffer of main roads.
Map 8. 1 km and 5 km buffers of main roads and Gamba grass sites with fire frequencies (10-19)
Discussion
The Gamba grass sites in the study area had high fire frequencies and several Gamba grass sites burnt 19 times from 2000-2018. From the results it can be deduced that Gamba grass provides a great amount of fuel load and in general Gamba grass sites burn very frequently. The low fire frequency Gamba grass sites are located close to the major towns i.e. Darwin and Katherine hence the reason for their low frequencies is that areas with greater populations are priority for fire prevention actions. The highest fire frequency Gamba grass sites are located in clusters which is probably the reason why they burn so often.
High fire frequency (10-19) Gamba grass sites were close to the major roads of the study sites, with majority of the sites being maximum 5 km away from roads. This can be explained by the fact that weeds usually spread along roadsides as cars and people are carrying their seeds along the roads.
Limitations
The limitation of this project was that systematically recorded Gamba grass data only covers a small area of the northern part of the Northern Territory, hence the results of this study are not representative of the whole northern NT.
Recommedations
It would be of interest to collect Gamba grass data in the whole northern area of NT, because Gamba grass sites burn frequently which has implications to fire management. As Gamba grass increases fuel load, it would be of interest to assess how fast it spreads and in what direction but for that Gamba grass data would have to be collected for the whole northern area of NT. That data could be used to predict fire frequencies for the Northern Territory, as from the results of this study Gamba grass sites burn frequently.
References
Australian states. (n.p.). Retrieved from: http://www.map.igismap.com
Australia State of the Environment. (2016). Regional and landscape-scale pressures: Bushfire. Retrieved from https://soe.environment.gov.au/theme/land/topic/2016/regional-and-landscape-scale-pressures-bushfire?fbclid=IwAR10d7J_LA1LXX0sInJLLzSN--pAEylAS_vZu9gulJ7PDsQ3XXQ-zkocY-U
Department of Agriculture and Fisheries. (2016). Gamba grass. State of Queensland.
Department of Environment and Energy. (2012). Weeds of National Significance (WONS). Retrieved from https://www.environment.gov.au/biodiversity/invasive/weeds/weeds/lists/wons.html
Gamba grass data provided by Professor Samantha Setterfield, University of Western Australia
Major roads in Australia. (2011). Retrieved from http://www.eatlas.org.au
North Australia & Rangelands Fire Information. (2018). Fire Frequency 2000-2018. Retrieved from http://www.firenorth.org.au
Northern Territory Government. (2009). Weed Risk Assessment Report: Andropogon gayanus (Gamba grass). Palmerston: Northern Territory Government.
Petty, A., Setterfield, S., Ferdinands, K., & Barrow, P. (2012). Inferring habitat suitability and spread patterns from large‐scale distributions of an exotic invasive pasture grass in north Australia. Journal of Applied Ecology, 49(3), 742–752.
Rossiter - Rachor, N., Setterfield, S., Douglas, M., Hutley, L., Cook, G., & Schmidt, S. (2009). Invasive Andropogon gayanus (gamba grass) is an ecosystem transformer of nitrogen relations in Australian savanna. Ecological Applications, 19(6), 1546–1560. https://doi.org/10.1890/08-0265.1
Rossiter, N., Setterfield, S., Douglas, M., & Hutley, L. (2003). Testing the grass‐fire cycle: alien grass invasion in the tropical savannas of northern Australia. Diversity and Distributions, 9(3), 169–176. https://doi.org/10.1046/j.1472-4642.2003.00020.x
Setterfield, S., Rossiter-Rachor, N., Douglas, M., Wainger, L., Petty, A., Barrow, P., … Setterfield, S. (2013). Adding Fuel to the Fire: The Impacts of Non-Native Grass Invasion on Fire Management at a Regional Scale. e59144. PLoS ONE, 8(5), e59144. https://doi.org/10.1371/journal.pone.0059144
Setterfield, S., Rossiter‐Rachor, N., Hutley, L., Douglas, M., & Williams, R. (2010). BIODIVERSITY RESEARCH: Turning up the heat: the impacts of Andropogon gayanus (gamba grass) invasion on fire behaviour in northern Australian savannas. Diversity and Distributions, 16(5), 854–861.
van Klinken, R., & Friedel, M. (2017). Unassisted invasions: understanding and responding to Australia’s high-impact environmental grass weeds. Australian Journal of Botany, 65(8), 678–690.