The influence of regional wind patterns on air quality during forest fires near Sydney, Australia.

Particulate pollution from forest fire smoke threatens the health of communities by increasing the occurrence of respiratory illnesses. Wind drives both fire behaviour and smoke dispersal. Understanding regional wind patterns would assist in effectively managing smoke risk. Sydney, Australia is prone to smoke pollution because it has a large population close to fire-prone eucalypt forests. Here we use the self-organising maps (SOM) technique to identify sixteen unique wind classes for the Sydney region from days with active fires, including identifying sea breeze occurrence. We explored differences in PM2.5 levels between classes and between hazard reduction burning (HRB) and wildfire days. For HRB days, classes with the highest PM2.5 mostly had a sea breeze, whereas better air quality days usually had winds aligned across the region (e.g. all westerly). The wind class with the most HRB days had low wind speeds and a sea breeze and was among the worst wind classes for air quality. For wildfire days, days with a sea breeze were also generally of poor air quality but many classes had at least some poor air quality days, most of which were during the 2019-2020 east coast wildfires in New South Wales. Some poor air quality days occurred in wind classes that sent strong plumes directly over air quality stations, spread smoke over a wide area or transported smoke from outside the region. The classes identified may be useful in scheduling HRBs, for example, identifying days with low pollution risk to conduct an HRB, or for assisting in understanding pollution risk and sending health warnings during HRBs and wildfires. Further development of the approach may allow the creation of multi-day classifications for fire managers to plan HRB ignitions over several days to ensure better smoke dispersal. Further research could incorporate other weather predictors or focus on other regions.

Prediction of air quality in Sydney, Australia as a function of forest fire load and weather using Bayesian statistics.

Smoke from Hazard Reduction Burns (HRBs) and wildfires contains pollutants that are harmful to human health. This includes particulate matter less than 2.5 µm in diameter (PM2.5), which affects human cardiovascular and respiratory systems and can lead to increased hospitalisations and premature deaths. Better models are needed to predict PM2.5 levels associated with HRBs so that agencies can properly assess smoke pollution risk and balance smoke risk with the wildfire mitigation benefits of HRBs. Given this need, our aim was to develop a probabilistic model of daily PM2.5 using Bayesian regression. We focused on the region around Sydney, Australia, which regularly has hazard reduction burning, wildfires and associated smoke. We developed two regional models (mean daily and maximum daily) from observed PM2.5, weather reanalysis and satellite fire hotspot data. The models predict that the worst PM2.5 in Sydney occurs when PM2.5 was high the previous day, there is low ventilation index (i.e. the product of wind speed and planetary boundary layer height), low temperature, west to northwest winds in the Blue Mountains, an afternoon sea breeze and large areas of HRBs are being conducted, particularly to the west and north of Sydney. A major benefit of our approach is that models are fast to run, require simple inputs and Bayesian predictions convey both predicted PM2.5 and associated prediction uncertainty. Future research could include the application of similar methods to other regions, collecting more data to improve model precision and developing Bayesian PM2.5 models for wildfires.

Experiments on the influence of spot fire and topography interaction on fire rate of spread.

Spotting is thought to increase wildfire rate of spread (ROS) and in some cases become the main mechanism for spread. The role of spotting in wildfire spread is controlled by many factors including fire intensity, number of and distance between spot fires, weather, fuel characteristics and topography. Through a set of 30 laboratory fire experiments on a 3 m x 4 m fuel bed, subject to air flow, we explored the influence of manually ignited spot fires (0, 1 or 2), the presence or absence of a model hill and their interaction on combined fire ROS (i.e. ROS incorporating main fire and merged spot fires). During experiments conducted on a flat fuel bed, spot fires (whether 1 or 2) had only a small influence on combined ROS. Slowest combined ROS was recorded when a hill was present and no spot fires were ignited, because the fires crept very slowly downslope and downwind of the hill. This was up to, depending on measurement interval, 5 times slower than ROS in the flat fuel bed experiments. However, ignition of 1 or 2 spot fires (with hill present) greatly increased combined ROS to similar levels as those recorded in the flat fuel bed experiments (depending on spread interval). The effect was strongest on the head fire, where spot fires merged directly with the main fire, but significant increases in off-centre ROS were also detected. Our findings suggest that under certain topographic conditions, spot fires can allow a fire to overcome the low spread potential of downslopes. Current models may underestimate wildfire ROS and fire arrival time in hilly terrain if the influence of spot fires on ROS is not incorporated into predictions.

Australian forests, megafires and the risk of dwindling carbon stocks.

Over the Austral spring and summer of 2019/20 > 7 million ha of Eucalyptus forest and woodland, including some of Australia's most carbon dense ecosystems, were burnt on the east coast of Australia. We estimated bootstrapped mean CO2 emissions of c. 0.67 Pg, with other available estimates ranging from 0.55 to 0.85 Pg. Eucalyptus forests are renowned for their ability to resist and recover from wildfire so it would be expected that emitted CO2 will be reabsorbed. The combination of drought and frequent fires is likely reducing the capacity to recover from the fire so future Australian forests may store less carbon. Broadscale prescribed burning is a widely promoted approach to reduce uncontrolled wildfires, yet the benefits for the management of carbon stores are controversial. Prescribed burning can reduce carbon losses from subsequent wildfire, yet the "carbon costs" of it may equal or outweigh the "carbon benefits" in reduced wildfire emissions. Likewise, mechanical thinning of vegetation to reduce fuel loads also carries heavy carbon costs with uncertain carbon benefits. Research involving empirical measurements, modelling and a mix of large-scale management intervention is urgently required to determine what interventions can maximise carbon storage in the face of climate change-driven fires.

Suppression resource decisions are the dominant influence on containment of Australian forest and grass fires.

Fire agencies aim to contain wildfires before they impact on life, property and infrastructure and to reduce the risk of damage to the environment. Despite the large cost of suppression, there are few data on the success of suppression efforts under varying weather, fuel and resource scenarios. We examined over 2200 forest and 4600 grass fires in New South Wales, Australia to determine the dominant influences on the containment of wildfires. A random forest modelling approach was used to analyse the effect of a range of human and environmental factors. The number of suppression resources per area of fire were the dominant influence on the containment of both forest and grass fires. As fire weather conditions worsened the probability of containment decreased across all fires and as fuel loads and slope increased the probability of containment decreased for forest fires. Environmental controls limit the effectiveness of wildfire management. However, results suggest investment in suppression resources and strategic fuel management will increase the probability of containment.

Mapping and exploring variation in post-fire vegetation recovery following mixed severity wildfire using airborne LiDAR.

There is a public perception that large high-severity wildfires decrease biodiversity and increase fire hazard by homogenizing vegetation composition and increasing the cover of mid-story vegetation. But a growing literature suggests that vegetation responses are nuanced. LiDAR technology provides a promising remote sensing tool to test hypotheses about post-fire vegetation regrowth because vegetation cover can be quantified within different height strata at fine scales over large areas. We assess the usefulness of airborne LiDAR data for measuring post-fire mid-story vegetation regrowth over a range of spatial resolutions (10 × 10 m, 30 × 30 m, 50 × 50 m, 100 × 100 m cell size) and investigate the effect of fire severity on regrowth amount and spatial pattern following a mixed severity wildfire in Warrumbungle National Park, Australia. We predicted that recovery would be more vigorous in areas of high fire severity, because park managers observed dense post-fire regrowth in these areas. Moderate to strong positive associations were observed between LiDAR and field surveys of mid-story vegetation cover between 0.5-3.0 m. Thus our LiDAR survey was an apt representation of on-ground vegetation cover. LiDAR-derived mid-story vegetation cover was 22-40% lower in areas of low and moderate than high fire severity. Linear mixed-effects models showed that fire severity was among the strongest biophysical predictors of mid-story vegetation cover irrespective of spatial resolution. However much of the variance associated with these models was unexplained, presumably because soil seed banks varied at finer scales than our LiDAR maps. Dense patches of mid-story vegetation regrowth were small (median size 0.01 ha) and evenly distributed between areas of low, moderate and high fire severity, demonstrating that high-severity fires do not homogenize vegetation cover. Our results are relevant for ecosystem conservation and fire management because they: indicate that native vegetation are responsive and resilient to high-severity fire, and show the usefulness of remote sensing tools such as LiDAR to monitor post-fire vegetation recovery over large areas in situ.

Acacia shrubs respond positively to high severity wildfire: Implications for conservation and fuel hazard management.

High severity wildfires pose threats to human assets, but are also perceived to impact vegetation communities because a small number of species may become dominant immediately after fire. However there are considerable gaps in our knowledge about species-specific responses of plants to different fire severities, and how this influences fuel hazard in the short and long-term. Here we conduct a floristic survey at sites before and two years after a wildfire of unprecedented size and severity in the Warrumbungle National Park (Australia) to explore relationships between post-fire growth of a fire responsive shrub genera (Acacia), total mid-story vegetation cover, fire severity and fuel hazard. We then survey 129 plots surrounding the park to assess relationships between mid-story vegetation cover and time-since-fire. Acacia species richness and cover were 2.3 and 4.3 times greater at plots after than before the fire. However the same common dominant species were present throughout the study. Mid-story vegetation cover was 1.5 times greater after than before the wildfire, and Acacia species contribution to mid-story cover increased from 10 to 40%. Acacia species richness was not affected by fire severity, however strong positive associations were observed between Acacia and total mid-story vegetation cover and severity. Our analysis of mid-story vegetation recovery showed that cover was similarly high between 2 and 30years post-fire, then decreased until 52years. Collectively, our results suggest that Acacia species are extremely resilient to high severity wildfire and drive short to mid-term increases in fuel hazard. Our results are discussed in relation to fire regime management from the twin perspectives of conserving biodiversity and mitigating human losses due to wildfire.

Some Wildfire Ignition Causes Pose More Risk of Destroying Houses than Others.

Many houses are at risk of being destroyed by wildfires. While previous studies have improved our understanding of how, when and why houses are destroyed by wildfires, little attention has been given to how these fires started. We compiled a dataset of wildfires that destroyed houses in New South Wales and Victoria and, by comparing against wildfires where no houses were destroyed, investigated the relationship between the distribution of ignition causes for wildfires that did and did not destroy houses. Powerlines, lightning and deliberate ignitions are the main causes of wildfires that destroyed houses. Powerlines were 6 times more common in the wildfires that destroyed houses data than in the wildfires where no houses were destroyed data and lightning was 2 times more common. For deliberate- and powerline-caused wildfires, temperature, wind speed, and forest fire danger index were all significantly higher and relative humidity significantly lower (P < 0.05) on the day of ignition for wildfires that destroyed houses compared with wildfires where no houses were destroyed. For all powerline-caused wildfires the first house destroyed always occurred on the day of ignition. In contrast, the first house destroyed was after the day of ignition for 78% of lightning-caused wildfires. Lightning-caused wildfires that destroyed houses were significantly larger (P < 0.001) in area than human-caused wildfires that destroyed houses. Our results suggest that targeting fire prevention strategies around ignition causes, such as improving powerline safety and targeted arson reduction programmes, and reducing fire spread may decrease the number of wildfires that destroy houses.

The potential for LiDAR technology to map fire fuel hazard over large areas of Australian forest.

Fuel load is a primary determinant of fire spread in Australian forests. In east Australian forests, litter and canopy fuel loads and hence fire hazard are thought to be highest at and beyond steady-state fuel loads 15-20 years post-fire. Current methods used to predict fuel loads often rely on course-scale vegetation maps and simple time-since-fire relationships which mask fine-scale processes influencing fuel loads. Here we use Light Detecting and Remote Sensing technology (LiDAR) and field surveys to quantify post-fire mid-story and crown canopy fuel accumulation and fire hazard in Dry Sclerophyll Forests of the Sydney Basin (Australia) at fine spatial-scales (20 × 20 m cell resolution). Fuel cover was quantified in three strata important for crown fire propagation (0.5-4 m, 4-15 m, >15 m) over a 144 km(2) area subject to varying fire fuel ages. Our results show that 1) LiDAR provided a precise measurement of fuel cover in each strata and a less precise but still useful predictor of surface fuels, 2) cover varied greatly within a mapped vegetation class of the same fuel age, particularly for elevated fuel, 3) time-since-fire was a poor predictor of fuel cover and crown fire hazard because fuel loads important for crown fire propagation were variable over a range of fire fuel ages between 2 and 38 years post-fire, and 4) fuel loads and fire hazard can be high in the years immediately following fire. Our results show the benefits of spatially and temporally specific in situ fuel sampling methods such as LiDAR, and are widely applicable for fire management actions which aim to decrease human and environmental losses due to wildfire.

The drivers of wildfire enlargement do not exhibit scale thresholds in southeastern Australian forests.

Wildfires are complex adaptive systems, and have been hypothesized to exhibit scale-dependent transitions in the drivers of fire spread. Among other things, this makes the prediction of final fire size from conditions at the ignition difficult. We test this hypothesis by conducting a multi-scale statistical modelling of the factors determining whether fires reached 10 ha, then 100 ha then 1000 ha and the final size of fires >1000 ha. At each stage, the predictors were measures of weather, fuels, topography and fire suppression. The objectives were to identify differences among the models indicative of scale transitions, assess the accuracy of the multi-step method for predicting fire size (compared to predicting final size from initial conditions) and to quantify the importance of the predictors. The data were 1116 fires that occurred in the eucalypt forests of New South Wales between 1985 and 2010. The models were similar at the different scales, though there were subtle differences. For example, the presence of roads affected whether fires reached 10 ha but not larger scales. Weather was the most important predictor overall, though fuel load, topography and ease of suppression all showed effects. Overall, there was no evidence that fires have scale-dependent transitions in behaviour. The models had a predictive accuracy of 73%, 66%, 72% and 53% accuracy at 10 ha, 100 ha, 1000 ha and final size scales. When these steps were combined, the overall accuracy for predicting the size of fires was 62%, while the accuracy of the one step model was only 20%. Thus, the multi-scale approach was an improvement on the single scale approach, even though the predictive accuracy was probably insufficient for use as an operational tool. The analysis has also provided further evidence of the important role of weather, compared to fuel, suppression and topography in driving fire behaviour.

Resolving future fire management conflicts using multicriteria decision making.

Management strategies to reduce the risks to human life and property from wildfire commonly involve burning native vegetation. However, planned burning can conflict with other societal objectives such as human health and biodiversity conservation. These conflicts are likely to intensify as fire regimes change under future climates and as growing human populations encroach farther into fire-prone ecosystems. Decisions about managing fire risks are therefore complex and warrant more sophisticated approaches than are typically used. We applied a multicriteria decision making approach (MCDA) with the potential to improve fire management outcomes to the case of a highly populated, biodiverse, and flammable wildland-urban interface. We considered the effects of 22 planned burning options on 8 objectives: house protection, maximizing water quality, minimizing carbon emissions and impacts on human health, and minimizing declines of 5 distinct species types. The MCDA identified a small number of management options (burning forest adjacent to houses) that performed well for most objectives, but not for one species type (arboreal mammal) or for water quality. Although MCDA made the conflict between objectives explicit, resolution of the problem depended on the weighting assigned to each objective. Additive weighting of criteria traded off the arboreal mammal and water quality objectives for other objectives. Multiplicative weighting identified scenarios that avoided poor outcomes for any objective, which is important for avoiding potentially irreversible biodiversity losses. To distinguish reliably among management options, future work should focus on reducing uncertainty in outcomes across a range of objectives. Considering management actions that have more predictable outcomes than landscape fuel management will be important. We found that, where data were adequate, an MCDA can support decision making in the complex and often conflicted area of fire management.

The relationship between particulate pollution levels in Australian cities, meteorology, and landscape fire activity detected from MODIS hotspots.

Smoke from bushfires is an emerging issue for fire managers because of increasing evidence for its public health effects. Development of forecasting models to predict future pollution levels based on the relationship between bushfire activity and current pollution levels would be a useful management tool. As a first step, we use daily thermal anomalies detected by the MODIS Active Fire Product (referred to as "hotspots"), pollution concentrations, and meteorological data for the years 2002 to 2008, to examine the statistical relationship between fire activity in the landscapes and pollution levels around Perth and Sydney, two large Australian cities. Resultant models were statistically significant, but differed in their goodness of fit and the distance at which the strength of the relationship was strongest. For Sydney, a univariate model for hotspot activity within 100 km explained 24% of variation in pollution levels, and the best model including atmospheric variables explained 56% of variation. For Perth, the best radius was 400 km, explaining only 7% of variation, while the model including atmospheric variables explained 31% of the variation. Pollution was higher when the atmosphere was more stable and in the presence of on-shore winds, whereas there was no effect of wind blowing from the fires toward the pollution monitors. Our analysis shows there is a good prospect for developing region-specific forecasting tools combining hotspot fire activity with meteorological data.

The efficacy of fuel treatment in mitigating property loss during wildfires: Insights from analysis of the severity of the catastrophic fires in 2009 in Victoria, Australia.

Treatment of fuel (e.g. prescribed fire, logging) in fire-prone ecosystems is done to reduce risks to people and their property but effects require quantification, particularly under severe weather conditions when the destructive potential of fires on human infrastructure is maximised. We analysed the relative effects of fuel age (i.e. indicative of the effectiveness of prescribed fire) and logging on remotely sensed (SPOT imagery) severity of fires which occurred in eucalypt forests in Victoria, Australia in 2009. These fires burned under the most severe weather conditions recorded in Australia and caused large losses of life and property. Statistical models of the probability of contrasting extremes of severity (crown fire versus fire confined to the understorey) were developed based on effects of fuel age, logging, weather, topography and forest type. Weather was the primary influence on severity, though it was reduced at low fuel ages in Moderate but not Catastrophic, Very High or Low fire-weather conditions. Probability of crown fires was higher in recently logged areas than in areas logged decades before, indicating likely ineffectiveness as a fuel treatment. The results suggest that recently burnt areas (up to 5-10 years) may reduce the intensity of the fire but not sufficiently to increase the chance of effective suppression under severe weather conditions. Since house loss was most likely under these conditions (67%), effects of prescribed burning across landscapes on house loss are likely to be small when weather conditions are severe. Fuel treatments need to be located close to houses in order to effectively mitigate risk of loss.

The impact of antecedent fire area on burned area in southern California coastal ecosystems.

Frequent wildfire disasters in southern California highlight the need for risk reduction strategies for the region, of which fuel reduction via prescribed burning is one option. However, there is no consensus about the effectiveness of prescribed fire in reducing the area of wildfire. Here, we use 29 years of historical fire mapping to quantify the relationship between annual wildfire area and antecedent fire area in predominantly shrub and grassland fuels in seven southern California counties, controlling for annual variation in weather patterns. This method has been used elsewhere to measure leverage: the reduction in wildfire area resulting from one unit of prescribed fire treatment. We found little evidence for a leverage effect (leverage = zero). Specifically our results showed no evidence that wildfire area was negatively influenced by previous fires, and only weak relationships with weather variables rainfall and Santa Ana wind occurrences, which were variables included to control for inter-annual variation. We conclude that this is because only 2% of the vegetation burns each year and so wildfires rarely encounter burned patches and chaparral shrublands can carry a fire within 1 or 2 years after previous fire. Prescribed burning is unlikely to have much influence on fire regimes in this area, though targeted treatment at the urban interface may be effective at providing defensible space for protecting assets. These results fit an emerging global model of fire leverage which position California at the bottom end of a continuum, with tropical savannas at the top (leverage = 1: direct replacement of wildfire by prescribed fire) and Australian eucalypt forests in the middle (leverage ~ 0.25).

Managing the matrix: decadal responses of eucalypt-dominated savanna to ambient fire regimes.

Much of our understanding of the response of savanna systems to fire disturbance relies on observations derived from manipulative fire plot studies. Equivocal findings from both recent Australian and African savanna fire plot assessments have significant implications for informing conservation management and reliable estimation of biomass stocks and dynamics. Influential northern Australian replicated fire plot studies include the 24-year plot-scale Munmarlary and the five-year catchment-scale Kapalga, mesic savanna (> 1000 mm/yr of rainfall) experiments in present-day Kakadu National Park. At Munmarlary, under low-to-moderate-intensity fire treatments, woody vegetation dominated by mature eucalypts was found to be structurally stable. At Kapalga, substantial declines in woody biomass were observed under more intense fire treatments, and modeling assessments implicate early-season fires as having adverse effects on longer-term tree recruitment. Given these contrasting perspectives, here we take advantage of a landscape-scale fire response monitoring program established on three major northern Australian conservation reserves (Kakadu, Litchfield, and Nitmiluk National Parks). Using statistical modeling we assess the decadal effects of ambient fire regime parameters (fire frequency, severity, seasonality, time since fire) on 32 vegetation structure components and abundance of 21 tree and 16 grass species from 122 monitoring plots. Over the study period the mean annual frequency of burning of plots was 0.53, comprising mostly early-dry-season, low-severity fires. Structural and species responses were variable but often substantial, notably resulting in stem recruitment and declines in juveniles, but only weakly explained by fire regime and habitat variables. Modeling of these observations under three realistic scenarios (increased fire severity under projected worsening climate change; modest and significant reductions in fire frequency to meet conservation criteria) indicates that all scenarios have positive and negative structural implications. Effecting significant regional fire regime change (e.g., reduction in frequency and size of severe fires) is demonstrably feasible, but it incurs risks and potentially some undesirable structural consequences. Given recent Australian and African experience, the generality and application of landscape-scale implications derived from manipulative fire assessments (including variable grazing and browsing regimes) in savanna require more critical assessment.