The 2019/2020 bushfire season was one of the most devasting to occur in Australia. Between October 2019 and February 2020, almost 13 million hectares of land were burned in New South Wales and the Australian Capital Territory. The last fires were extinguished in February 2020, however the damage was extensive across the country.

Flora and fauna

Across the entire country, the largest area of land burned was conservation land. The Blue Mountains and the Gondwana world heritage sites suffered widespread damage. Many threatened species were affected by the bush fires. Furthermore, an estimated 1.5 billion wildlife animals, who resided in habitats destroyed by the fires, were killed.

Property damage

As well as the loss of wildlife, the properties of many Australians were destroyed. In New South Wales alone, thousands of buildings were destroyed or damaged. Insurance claims directly in relation to bushfires across the entire country were valued at 1.9 billion Australian dollars as of January 2020. The full financial impact from these fires is yet to bet determined.

As of Januray 2020, 25 people had lost their lives in New South Wales due to the 2019/2020 Australian bushfire season. A total of 34 people had died in the bushfires since October 2019.

In January 2020, the Australian federal government announced a two billion Australian dollar relief package to aid the bushfire recovery after the 2019/2020 fire season. Of this fund, 100 million Australian dollars had been allocated to primary producers in the country. Most of the funds, around 1.6 billion Australian dollars, had not yet been allocated as of that month.

Australia’s Black summer

The 2019/2020 bushfire season was one of the worst to ever occur in Australia. Between October 2019 and February 2020, bush fires ravaged many parts of Australia. The bushfires were worsened by the extreme heat and drought conditions across the country. Additionally, many fires were either deliberately or accidently ignited by people. The effects were devasting on wildlife, people, property, and the environment.

What was the impact?

Due to habitat destruction, many wildlife lost their lives, with some experts estimating that over 1.5 billion animals were lost. A total of 34 people died, many through battling the fires. In New South Wales, thousands of properties were damaged or destroyed. Insurance claims had already reached 740 million Australian dollars in early January 2020. Air pollution from the smoke haze and particulate matter reached record levels in the eastern and southern states. Many Australians were affected in some way by the bush fires, with the extent of damage not yet fully assessed.

The National Indicative Aggregated Fire Extent Dataset has been developed rapidly to support the immediate needs of the Department of Climate Change, Energy, the Environment and Water (DCCEEW, previously DAWE) in:quantifying the potential impacts of the 2019/20 bushfires on wildlife, plants and ecological communities; and,identifying appropriate response and recovery actions.The intent was to derive a reliable, agreed, fit for purpose and repeatable national dataset of burnt areas across Australia for the 2019/20 bushfire season.The NIAFED was first published on 13 February 2020 and was updated several times during 2020 to reflect updates to fire extent datasets from state and territory agencies. Most changes across these versions, after February (end of summer), reflect refinements on previous extent mapping, rather than new burnt areas. Fire analyses and decision making within the department after June 2020 has been based on the GEEBAM dataset. The GEEBAM dataset reports on fire severity within the NIAFED v20200225 extent envelope and includes some areas determined to be unburnt within NIAFED areas.NOTE: previous versions of this dataset are available on request to geospatial@dcceew.gov.auThe dataset takes the national Emergency Management Spatial Information Network Australia (EMSINA) data service, which is the official fire extent currently used by the Commonwealth and adds supplementary data from other sources to form a cumulative national view of fire extent. This EMSINA data service shows the current active fire incidents, and the Department map shows the total fire extent from 1 July 2019 to the 22 June 2020.EMSINA have been instrumental in providing advice on access to data and where to make contact in the early stages of developing the National Indicative Aggregated Fire Extent Dataset.This dataset is released on behalf of the Commonwealth Government and endorsed by the National Burnt Area Dataset Working Group, convened by the National Bushfire Recovery Agency.Known Issues:The dataset has a number of known issues, both in its conceptual design and in the quality of its inputs. These are outlined below and should be taken into account in interpreting the data and developing any derived analyses.The list of known issues below is not comprehensive: it is anticipated that further issues will be identified in the future, and the Department welcomes feedback on this. We will seek as far as possible to continuously improve the dataset in future versions.In addition, the 2019/20 bushfire season is ongoing and it can be expected that the fire extent will increase.Future versions of the dataset will therefore document and distinguish between changes arising from methodological improvement, as distinct from changes to the actual fire extent.The dataset draws data together from multiple different sources, including from state and territory agencies responsible for emergency and natural resource management, and from the Northern Australian Fire Information website. The variety of mapping methods means that conceptually the dataset lacks national coherency. The limitations associated with the input datasets are carried through to this dataset. Users are advised to refer to the input datasets’ documentation to better understand limitations.The dataset is intentionally precautionary and the rulesets for its creation elect to accept the risk of overstating the size of particular burnt areas. If and when there are overlapping polygons for an area, the internal boundaries have been dissolved.The dataset shows only the outline of burnt areas and lacks information on fire severity in these areas, which may often include areas within them that are completely unburnt. For the intended purpose this may limit the usability of the data, particularly informing on local environmental impacts and response. This issue will be given priority, either for future versions of the dataset or for development of a separate, but related, fire severity product.This continental dataset includes large burnt areas, particularly in northern Australia, which can be considered part of the natural landscape dynamics. For the intended purpose of informing on fire of potential environmental impact, some interpretation and filtering may be required. There are a variety of ways to do this, including by limiting the analysis to southern Australia, as was done for recent Wildlife and Threatened Species Bushfire Recovery Expert Panel’s preliminary analysis of 13 January 2020. For that preliminary analysis area, boundaries from the Interim Biogeographic Regionalisation of Australia version 7 were used by the Department to delineate an area of southern Australia encompassing the emergency bushfire areas of the southern summer. The Department will work in consultation with the expert panel and other relevant bodies in the future on alternative approaches to defining, spatially or otherwise, fire of potential environmental impact.The dataset cannot be used to reliably recreate what the national burnt area extent was at a given date prior to the date of release. Reasons for this include that information on the date/time on individual fires may or may not have been provided in the input datasets, and then lost as part of the dissolve process discussed in issue 2 above.With fires still burning extents are not yet refined.Fire extents are downloaded daily, and datasets are aggregated. This results in an overlap of polygon extents and raises the issue that refined extents are disregarded at this early stage.The Northern Australian Fire Information (NAFI) dataset is only current to 19 June 2020.

Important: Our technical support team is available to assist you during business hours only. Please keep in mind that we can only address technical difficulties during these hours. When using the product to make decisions, please take this into consideration.

Abstract This spatial product shows consistent ‘near real-time’ bushfire and prescribed burn boundaries for all jurisdictions who have the technical ability or appropriate licence conditions to provide this information. Currency Maintenance of the underlying data is the responsibility of the custodian. Geoscience Australia has automated methods of regularly checking for changes in source data. Once detected the dataset and feeds will be updated as soon as possible. NOTE: The update frequency of the underlying data from the jurisdictions varies and, in most cases, does not line up to this product’s update cycle. Date created: November 2023 Modification frequency: Every 15 Minutes Spatial Extent

West Bounding Longitude: 113° South Bounding Latitude: -44° East Bounding Longitude: 154° North Bounding Latitude: -10°

Source Information The project team initially identified a list of potential source data through jurisdictional websites and the Emergency Management LINK catalogue. These were then confirmed by each jurisdiction through the EMSINA National and EMSINA Developers networks. This Webservice contains authoritative data sourced from:

Australian Capital Territory - Emergency Service Agency (ESA) New South Wales - Rural Fire Service (RFS) Queensland - Queensland Fire and Emergency Service (QFES) South Australia - Country Fire Service (CFS)
Tasmania - Tasmania Fire Service (TFS)
Victoria – Department of Environment, Land, Water and Planning (DELWP)
Western Australia – Department of Fire and Emergency Services (DFES)

The completeness of the data within this webservice is reliant on each jurisdictional source and the information they elect to publish into their Operational Bushfire Boundary webservices. Known Limitations:

This dataset does not contain information from the Northern Territory government. This dataset contains a subset of the Queensland bushfire boundary data. The Queensland ‘Operational’ feed that is consumed within this National Database displays a the last six (6) months of incident boundaries. In order to make this dataset best represent a ‘near-real-time’ or current view of operational bushfire boundaries Geoscience Australia has filtered the Queensland data to only incorporate the last two (2) weeks data. Geoscience Australia is aware of duplicate data (features) may appear within this dataset. This duplicate data is commonly represented in the regions around state borders where it is operationally necessary for one jurisdiction to understand cross border situations. Care must be taken when summing the values to obtain a total area burnt. The data within this aggregated National product is a spatial representation of the input data received from the custodian agencies. Therefore, data quality and data completion will vary. If you wish to assess more information about specific jurisdictional data and/or data feature(s) it is strongly recommended that you contact the appropriate custodian.

The accuracy of the data attributes within this webservice is reliant on each jurisdictional source and the information they elect to publish into their Operational Bushfire Boundary webservices. Note: Geoscience Australia has, where possible, attempted to align the data to the (as of October 2023) draft National Current Incident Extent Feeds Data Dictionary. However, this has not been possible in all cases. Work to progress this alignment will be undertaken after the publication of this dataset, once this project enters a maintenance period. Catalog entry: Bushfire Boundaries – Near Real-Time Lineage Statement Version 1 and 2 (2019/20): This dataset was first built by EMSINA, Geoscience Australia, and Esri Australia staff in early January 2020 in response to the Black Summer Bushfires. The product was aimed at providing a nationally consistent dataset of bushfire boundaries. Version 1 was released publicly on 8 January 2020 through Esri AGOL software.
Version 2 of the product was released in mid-February as EMSINA and Geoscience Australia began automating the product. The release of version 2 exhibited a reformatted attributed table to accommodate these new automation scripts. The product was continuously developed by the three entities above until early May 2020 when both the scripts and data were handed over to the National Bushfire Recovery Agency. The EMSINA Group formally ended their technical involvement with this project on June 30, 2020. Version 3 (2020/21): A 2020/21 version of the National Operational Bushfire Boundaries dataset was agreed to by the Australian Government. It continued to extend upon EMSINA’s 2019/20 Version 2 product. This product was owned and managed by the Australian Government Department of Home Affairs, with Geoscience Australia identified as the technical partners responsible for development and delivery. Work on Version 3 began in August 2020 with delivery of this product occurring on 14 September 2020. Version 4 (2021/22): A 2021/22 version of the National Operational Bushfire Boundaries dataset was produced by Geoscience Australia. This product was owned and managed by Geoscience Australia, who provided both development and delivery. Work on Version 4 began in August 2021 with delivery of this product occurring on 1 September 2021. The dataset was discontinued in May 2022 because of insufficient Government funding. Version 5 (2023/25): A 2023/25 version of the National Near-Real-Time Bushfire Boundaries dataset is produced by Geoscience Australia under funding from the National Bushfire Intelligence Capability (NBIC) - CSIRO. NBIC and Geoscience Australia have also partnered with the EMSINA Group to assist with accessing and delivering this dataset. This dataset is the first time where the jurisdictional attributes are aligned to AFAC’s National Bushfire Schema.
Work on Version 5 began in August 2023 and was released in late 2023 under formal access arrangements with the States and Territories. Data Dictionary Geoscience Australia has not included attributes added automatically by spatial software processes in the table below.

Attribute Name Description

fire_id ID attached to fire (e.g. incident ID, Event ID, Burn ID).

fire_name Incident name. If available.

fire_type Binary variable to describe whether a fire was a bushfire or prescribed burn.

ignition_date The date of the ignition of a fire event. Date and time are local time zone from the State where the fire is located and stored as a string.

capt_date The date of the incident boundary was captured or updated. Date and time are local time zone from the Jurisdiction where the fire is located and stored as a string.

capt_method Categorical variable to describe the source of data used for defining the spatial extent of the fire.

area_ha Burnt area in Hectares. Currently calculated field so that all areas calculations are done in the same map projection. Jurisdiction supply area in appropriate projection to match state incident reporting system.

perim_km ) Burnt perimeter in Kilometres. Calculated field so that all areas calculations are done in the same map projection. Jurisdiction preference is that supplied perimeter calculations are used for consistency with jurisdictional reporting.

state State custodian of the data. NOTE: Currently some states use and have in their feeds cross border data

agency Agency that is responsible for the incident

date_retrieved The date and time that Geoscience Australia retrieved this data from the jurisdictions, stored as UTC. Please note when viewed in ArcGIS Online, the date is converted from UTC to your local time.

Contact Geoscience Australia, clientservices@ga.gov.au

Context

The Australian bushfire has been devastating. It has caused massive loss of wildlife, forest land, and has even to led human casualties. There have been huge undertakings by the fire department on controlling the bushfire and recovery activity. Thanks to NASA the satellite data by MODIS and VIIRS , near real-time data is made available publically. With Fire still active, with the latest one near capital city of Canberra. The purpose of putting this dataset here is to gain useful insights from this data and make it available to everyone.

Content

There are 4 files in the data. Files are classified as Archive or NRT. The archive data is older and well calibrated. Whereas NRT(near real time) data is generated within just 3 hours of satellite detection. It is to support the immediate needs. Archive data is between sept 1,2019 to Dec 31, 2019. Whereas NRT data is between Jan 1,2020 to Jan 31,2020. Both the files can be merged.

The files are also classified based on satellite - MODIS (M6) and VIIRS( V1). In MODIS (Moderate Resolution Imaging Spectroradiometer) each hotspot detection represents center of l km, meaning atleast one fire is located in less than 1km region. VIIRS (Visible Infrared Imaging Radiometer Suite) has improved spatial resolution of 375m.

The measurement FRP(Fire radioactive power) can be used to detect fire.

Acknowledgements

We acknowledge the use of data and imagery from LANCE FIRMS operated by NASA's Earth Science Data and Information System (ESDIS) with funding provided by NASA Headquarters.

• NRT VIIRS 375 m Active Fire product VNP14IMGT. Available on-line [https://earthdata.nasa.gov/firms]. doi: 10.5067/FIRMS/VIIRS/VNP14IMGT.NRT.001.
• MODIS Collection 6 NRT Hotspot / Active Fire Detections MCD14DL. Available on-line [https://earthdata.nasa.gov/firms]. doi: 10.5067/FIRMS/MODIS/MCD14DL.NRT.006

cover photo credits - Photo by Daniel Morton on Unsplash

Inspiration

Some points to start of ->

• Which regions were the most affected?
• Time line- Progress of Bush fire.
• Visualization using of geo data on map - using folium, bokeh or plotly .
• can this dataset be integrated with other datasets to obtain insights?
• Check out the starter-notebook to get started (data has been loaded).

Disclaimer by NASA -

The LANCE system is operated by the NASA/GSFC Earth Science Data and Information System (ESDIS). The information presented through LANCE, Rapid Response, GIBS, Worldview, and FIRMS are provided "as is" and users bear all responsibility and liability for their use of data, and for any loss of business or profits, or for any indirect, incidental or consequential damages arising out of any use of, or inability to use, the data, even if NASA or ESDIS were previously advised of the possibility of such damages, or for any other claim by you or any other person. ESDIS makes no representations or warranties of any kind, express or implied, including implied warranties of fitness for a particular purpose or merchantability, or with respect to the accuracy of or the absence or the presence or defects or errors in data, databases of other information. The designations employed in the data do not imply the expression of any opinion whatsoever on the part of ESDIS concerning the legal or development status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. For more information please contact Earthdata Support.

This is the download zipped data for the AUS GEEBAM map service.The Australian Google Earth Engine Burnt Area Map (AUS GEEBAM) is a rapid, national approach to fire severity mapping. It has been developed rapidly to support the immediate needs of the Department of Climate Change, Energy, the Environment and Water (DAWE) in:a) quantifying the potential impacts of the 2019/20 bushfires on wildlife, plants and ecological communities, andb) identifying appropriate response and recovery actions.AUS GEEBAM Fire Severity uses Sentinel 2A satellite imagery from before and after fire to estimate the severity of burn within each 40m grid cell. Fire severity is defined as a metric of the loss or change in organic matter caused by fire.The extent of the 2019/2020 fires was derived from the National Indicative Aggregated Fire Extent Dataset (NIAFED). NIAFED was sourced from the national Emergency Management Spatial Information Network Australia (EMSINA) data service, which is the official fire extent currently used by the Commonwealth and adds supplementary data from other sources to form a cumulative national view of fire extent.AUS GEEBAM relies on a vegetation index (Relativised Normalized Burnt Ratio, RNBR) that is calculated for burnt areas and adjacent unburnt areas, before and after the fire season. The result is a map of four fire severity classes that represent how severely vegetation was burnt during the 2019/2020 fires.To determine a reference unburnt condition, the NIAFED extent was buffered by 2km. For each NVIS broad vegetation type, in each IBRA bioregion a reference unburnt RNBR class was determined. That value was available to calculate a standardised offset or a reference unburnt value.Each IBRA bioregion was systematically assessed to correct for obvious errors. For example, the Very High severity class could be adjusted down by one RNBR Value for a fire where its extent extended into an area of lower severity. Conversely, there were areas of shrublands that had clearly burnt at Very High severity where all of the biomass is likely to have been consumed but low pre-fire biomass had given it a lower RNBR Value.Each pixel of AUS GEEBAM contains the raw RNBR Value, the RNBR Class and the GEEBAM Value. This allows an end user to observe which values have been adjusted during the calibration away from the default global RNBR Value and allows for some transparency in the process.GEEBAMValueGEEBAM ClassDescription1No dataNo data indicates areas outside NIAFED or NVIS categories that do not represent native vegetation (e.g. cleared land, water)2UnburntLittle or no change observed between pre-fire and post-fire imagery.3Low and ModerateSome change or moderate change detected when compared to reference unburnt areas outside the NIAFED extent.4HighVegetation is mostly scorched.5Very highVegetation is clearly consumed.Known Issues:The dataset has a number of known issues, both in its conceptual design and in the quality of its inputs. These are outlined below and should be taken into account when interpreting the data and developing any derived analyses.The list of known issues below is not comprehensive, it is anticipated that further issues will be identified, and the Department welcomes feedback on this. We will seek as far as possible to continuously improve the dataset in future versions.AUS GEEBAM classes are not based on field data and no confidence interval or report on accuracy has been provided.The number of severity classes has been reduced by combining low and moderate severity fires. Single index thresholds are known to feature poor delineation of low fire severity classes.AUS GEEBAM classes are calibrated systematically for each bioregion using visual interpretation of Sentinel 2 false colour composites.The limitations associated with the NIAFED are carried through to this dataset. Users are advised to refer to the NIAFED documentation to better understand limitations.This continental dataset includes large burnt areas, particularly in northern Australia, which can be considered part of the natural landscape dynamics. For the intended purpose of informing on the potential impact of fire on the environmental, some interpretation and filtering may be required.The NIAFED dataset used as the extent layer for AUS GEEBAM Fire Severity is current as of 24 February 2020. More recent versions were available at the time of creation, however, these would have introduced burnt areas from a second fire season in Northern Australia where fire patterns differ greatly to that of southern Australia.NOTE:Report methodology and supporting material is available on request togeospatial@dcceew.gov.au

Our study focuses on the south-eastern region of Australia. In recent years, the south-eastern part has been experiencing increasing frequency of wildfires. However, the 2019-2020 bushfire season was unprecedented in intensity and devastation. It is widely known as ‘Black Summer’.This study combined various biophysical factors, including MODIS MCD64A1 fire product, digital elevation model (DEM), slope, aspect, ECOSTRESS data (i.e., evapotranspiration – ET, evaporative stress index – ESI, land surface temperature – LST, water use efficiency -WUE), NDVI generated from Sentinel-2 data, and rainfall data, for wildfire prediction. To this aim, we designed models that incorporate pre-fire vegetation conditions obtained from ECOSTRESS data to predict the probability of future wildfire occurrence. The predictivity of models and biophysical factors were assessed to understand pre-fire vegetation conditions and wildfire susceptibility.

We used nine variables from four sources as explanatory variables (Table 1). Fire occurrences between the period of September 2019 and March 2020 were obtained from the MODIS MCD64A1 product as a shapefile and mapped. A dataset was created to record the presence and absence of fires, classified as 0 and 1, respectively. Rainfall data were obtained from the Bureau of Meteorology, Australia, for all seven months, which was then compiled and interpolated using the Inverse Distance Weighting (IDW) method. The IDW tool from ArcGIS Spatial Analyst extension is used. DEM derivatives such as slope and aspect were created using Slope and Aspect tools in ArcGIS Pro. Sentinel-2 L2A (16-bit) data was downloaded from the Sentinel Hub EO browser at a resolution of 10 m, and NDVI was mapped using bands 4 and 8. All variable raster images were clipped to extract the study area. ECOSTRESS data products, including Evapotranspiration (ET), Evaporative Stress Index (ESI), Land Surface Temperature (LST), and Water Use Efficiency (WUE), acquired from NASA LPDAAC AppEARS, were used to model wildfire dynamics (Fisher et al., 2020; Zhu et al., 2022). A mosaic dataset in a raster format was created for each variable over the seven months between September 2019 and March 2020.

Table 1. Explanatory variables used in this research and their data sources

Category Explanatory variables Source

ECOSTRESS

Evapotranspiration (ET)

70m resolution ECOSTRESS data from LPDAAC AppEARS https://lpdaacsvc.cr.usgs.gov/appeears/

ECOSTRESS

Evaporative stress index (ESI)

70m resolution ECOSTRESS data from LPDAAC AppEARS https://lpdaacsvc.cr.usgs.gov/appeears/

ECOSTRESS

Land surface temperature (LST)

70m resolution ECOSTRESS data from LPDAAC AppEARS https://lpdaacsvc.cr.usgs.gov/appeears/

ECOSTRESS

Water use efficiency (WUE)

70m resolution ECOSTRESS data from LPDAAC AppEARS https://lpdaacsvc.cr.usgs.gov/appeears/

Vegetation Index

Normalized Difference Vegetation Index (NDVI)

SENTINEL-2 Data (10 m resolution, band 4 and 8 is used) https://scihub.copernicus.eu/dhus/#/home

Climate

Rainfall

Bureau of Meteorology, Australia http://www.bom.gov.au/climate/data/

Topography

Elevation

9 arc-second DEM (~250 m resolution) from Geoscience Australia (Hutchinson et al., 2008)

Topography

Slope

Derived from DEM

Topography

Aspect

Derived from DEM

Two categories of models were developed in this study: general models and monthly models. The general models were specifically constructed to estimate wildfire susceptibility and quantify the significance of input biophysical factors over the entire wildfire period, spanning from September 2019 to March 2020. These models utilized the mean values of explanatory variables throughout this period as independent input variables, with the samples collected from MODIS ground fire points during 2019-2020 serving as the dependent variable. The study integrated a range of explanatory variables, including ECOSTRESS data, vegetation indices, climatic parameters, and topographical factors, to quantitatively assess their respective impacts on the prediction of wildfire.

The monthly models were designed to capture pre-fire vegetation conditions and predict wildfire spread one week ahead. We set up a three-week time lag for data collection prior to a wildfire event in the 4th week and predict the probability of wildfire occurrence in the following week (5th week). The mean values of the selected data in three weeks were computed to minimize or eliminate gaps. The model, for example, to predict wildfire occurrence probability in the first week of September (September 1-7), was built using the mean values of explanatory variables during a three-week time from August 1 to August 21. Such a design is to create an effective model to predict wildfire spread and assess the impact of pre-fire plant stress on following wildfire occurrence. The Australian bushfires started to spread in the first week of September 2019 and faded in early April 2020. The fires ceased at the end of October 2019 in south-eastern Australia and reignited in late November 2019. To understand the impact of change in the climate condition of the country after the first fire and to effectively assess the fire influential factor, we built three monthly models to predict (1) the first week of September (the week when the first wildfire started), (2) the last week of November, and (3) the first week of December (the weeks when the second fire started).

Machine Learning is based on algorithms that have the capacity to learn from data and make effective predictions. This learning process involves modeling the hidden relationships between a set of input variables (explanatory variables) and the occurrences of the phenomenon (the dependent variable) (Tonini et al., 2020). we acquired 2037 wildfire occurrence points. Of these, 70% (1426 wildfire occurrence points) were allocated for training, while the remaining 30% (611 wildfire occurrence points) were reserved for validation. Here, we evaluated LR, GWR, and RF algorithms to create models that fit relationships between wildfire events and the explanatory variables. The fit relationships from these models were then used in the susceptibility mapping and assessment of variable influence. Linear Regression (LR), in particular, demands the independence of explanatory variables. To mitigate the impact of the correlation between these variables, we employed a regularization technique using LASSO (L1 regularization). LASSO penalizes the coefficients of correlated variables, prompting the LR model to favor a subset of independent variables and enhance model robustness (Qian et al., 2012). Prior to the application of LR and GWR, we normalized the explanatory variables to a common scale (between 0 and 1) based on their observed maximum and minimum values (Zhu et al., 2022). This normalization ensures equal contributions from all variables. Such scaling facilitates straightforward comparison and interpretation of variable importance.

In addition, to evaluate the accuracy of wildfire susceptibility modeling, pixels were categorized as either fire or non-fire based on a probability threshold value of 0.5. Pixels greater than 0.5 were identified as fire pixels, while those below the threshold value were not considered in the process. A confusion matrix is utilized to evaluate the performance of a classification model that predicts two or more classes. This matrix evaluates the accuracy, sensitivity, and specificity of the model’s outcomes (Parikh et al., 2008).

This dataset covers south-eastern region of Australia during 2019-2020. The dataset includes input explanatory variables of general and monthly models, wildfire susceptibility for each city and fire locations.

This data asset contains observations of individual plants and animals (“occurrences”) sourced from the Atlas of Living Australia. Data on vascular plants are based on the following paper:

Godfree et al. (2021) Implications of the 2019-2020 Megafires for the Biogeography and Conservation of Australian Vegetation. Nature Communications 12: 1023 https://doi.org/10.1038/s41467-021-21266-5

Data on invertebrates are from the following report:

Marsh et al. (2021) Threatened species hub report: Assessment of the impacts of the 2019-20 wildfires on southern and eastern Australia on invertebrate species. NESP Threatened Species Recovery Hub Project 8.3.1 Final report, Brisbane.

Both studies were performed to understand the impacts of the Australian “Black Summer” (2019-20) fires on the taxonomic group in question. Consequently, this aggregated dataset is designed to support off-the-shelf bushfire impact modelling, and to provide useful context for associated biodiversity conservation work. A total of 896 species of vascular plants and 44,146 invertebrate species.

The combined data asset was produced by the Science & Decision Support Team at the Atlas of living Australia (ALA) in collaboration with the authors of the original data sets. This was done as part of an Australian Research Data Commons (ARDC) bushfire data challenges project. The data contained in this deposition has been cleaned by researchers and taxonomists to ensure it is of high quality. Therefore, compared to data returned by the ALA, the unique record ID’s may return a different species or location than is denoted in this data set: these records have potentially been changed by the authors of the data to reflect what they believe is correct.

Lineage: This dataset contains data from the Atlas of Living Australia. Only herbarium data was used for plant data, while for invertebrate data, citizen science data was excluded. Data came from two primary sources: 1) Godfree, Bob; Knerr, Nunzio; Encinas-Viso, Francisco; Albrecht, Dave; Bush, David; Cargill, Christine; Clements, Mark; Gueidan, Cecile; Guja, Lydia; Harwood, Tom; Joseph, Leo; Lepschi, Brendan; Nargar, Katharina; Schmidt-Lebuhn, Alexander; Broadhurst, Linda (2020): Implications of the 2019-2020 Megafires for the Biogeography and Conservation of Australian Vegetation: supporting data and code. v2. CSIRO. Data Collection. https://doi.org/10.25919/sd7h-ff33 2) Bal, Payal, Marsh, Jess, & Woinarski, John. (2021). Data and outputs for NESP Project 8.3.1 [Data set]. In Final report for NESP Threatened Species Recovery Hub Project 8.3.1: 'Fire-affected invertebrates: priority species and management response'. Zenodo. https://doi.org/10.5281/zenodo.5091296.

This data set was generated in collaboration with those involved in the respective datasets referenced above.

This spatial product was built in direct response to the public's, industry's and governments demand/requirement for a National Bushfire Boundary layer during and after the 2020 Black Summer Bushfires.

Emergency Management Spatial information Network Australia (EMSINA) members (National and the Developers) begun developing this consolidated National Bushfire Boundary Webservice on December 30, 2019. With the technical assistance of Esri Australia the first public facing product (v1) became available on January 8, 2020. Geoscience Australia staff were involved in the development and maintenance of the FME scripts.

Version 2 of the product was released in February as EMSINA and Geoscience Australia began automating the product. This release exhibited a reformatted attributed table to accommodate the new automation scripts; unfortunately this release resulted in having to release the data with new access end points.

The product was continuously developed until early May 2020. On May 4, 2020 the EMSINA Committee made the decision to stop running the script i.e. 1 month after the official end of the bushfire season.

The product and the scripting were subsequently archived on both EMSINA and GA infrastructure. The National Bushfire Recovery Agency was provided a copy of the automation scripts on May 7, 2020 and then on June 26, 2020 a complete copy of the data from January 8, 2020 to May 4, 2020 in a geodatabase format.

The EMSINA Group formally ended their technical involvement with this project on June 30, 2020.

The Emergency Management LINK (EM-LINK*) catalogue was the source for identifying the majority of the required data. The consolidated National Bushfire Boundary Webservice contains data from:

 Australian Capital Territory - Emergency Service Agency (ESA)

 New South Wales - Rural Fire Service (RFS)

 Queensland - Queensland Fire and Emergency Service (QFES)

 Queensland - Queensland Department of Environment and Science (DES)

 South Australia - Country Fire Service (CFS)

 Tasmania - Tasmania Fire Service (TFS)

 Victoria - Emergency Management Victoria (EMV)

Where EMSINA and/or Geoscience Australia had difficulty in accessing the required spatial data and/or attributes assistance was provided by the National Bushfire Recovery Agency and Emergency Management Australia.

Known Limitations of the Data:

This database does not contain information from Western Australia. Neither EMSINA nor the Australian Government were able to negotiate appropriate access to the State’s operational boundary data.

EMSINA is aware of duplicate data (features) contained within this database. This duplicate data is commonly represented in the regions around state lines where it became operationally necessary to understand cross border situations. Care must be taken when summing the values to obtain a total area burnt.

EMSINA and Geoscience Australia became aware of the following error in the database in April 2020:

• Upon receipt of the Queensland data it was mistakenly assessed as Operational Boundaries and not as Historic FY Boundaries. This data was included into the database until the mistake was discovered in early April.

The data within this aggregated National product is a representation of the input data received from the custodian agencies. Therefore, to assess the quality or find out more information about an individual feature or features it is recommend that you contact the appropriate custodian.

The Integrated Water Quality and Environmental Monitoring Dataset before, during and after the 2019/2020 Bushfires is a comprehensive repository of data compiled from multiple sources, including the MER (Monitoring, Evaluating, Reporting) program and other relevant sources. This dataset encompasses various types of information essential for assessing the impact of the 2019/2020 bushfires on water quality in selected ICOLLs (Intermittently Closed and Open Lakes and Lagoons) and river estuaries. It incorporates historical water quality data collected during the bushfires and post-bushfire period through the MER program, rainfall data from nearby BOM (Australian Bureau of Meteorology) weather stations, logger data deployed at selected estuaries during or shortly after the bushfire season, dates of fire impact on the monitored systems, and surface and depth water quality monitoring data from smart buoys deployed across various estuaries.

Components:

1. Estuary and fire impact data Information detailing the systems incorporated in this data set and the dates when these systems were affected by the 2019/2020 bushfires.

2. MER Program Data (Lakes, Rivers): Historical water quality data collected during the 2019/2020 bushfires and subsequent periods as part of the MER program. Includes parameters such as pH levels, turbidity, salinity, chlorophyll-a concentrations, dissolved oxygen concentrations and nutrient concentrations. The MER dataset was collected as part of the NSW Government Monitoring, evaluating and reporting on estuaries program. The reported data has met the required QA/QC standards associated with this program. For further information: https://www.environment.nsw.gov.au/topics/water/estuaries/monitoring-and-reporting-estuaries

3. Rainfall Data: Precipitation data obtained from BOM weather stations located near the monitored estuaries.

4. Logger data: Data recorded by loggers deployed at selected estuaries including Lake Conjola, Meroo Lake, Termeil Lake, Wonboyn Lake, and Tuross River during or immediately following the 2019/2020 bushfire season . Includes measurements of environmental parameters such as water temperature, dissolved oxygen and salinity. Compiled from published data available here: https://datasets.seed.nsw.gov.au/dataset/bushfire-impact-water-quality

5. Smart Buoy Monitoring Data: Surface and depth water quality monitoring data collected from smart buoys deployed in various estuaries (Wonboyn Lake, Wallaga Lake, Lake Conjola, and Durras Lake). This data provides continuous or periodic measurements of water quality parameters, offering insights into spatial and temporal variations in water quality after the bushfire period in the longer-term.

Data for 26,062 native Australian plant species assessed against ten post-fire recovery criteria. Details of criteria and methods available in Gallagher, R. V. (2020) National prioritisation of Australian plants affected by the 2019–2020 bushfire season. Report to the Commonwealth Dartement of Agriculture, Water and Environment. https://www.environment.gov.au/system/files/pages/289205b6-83c5-480c-9a7d-3fdf3cde2f68/files/final-national-prioritisation-australian-plants-affected-2019-2020-bushfire-season.pdf

The TROPESS CrIS-SNPP L2 Peroxyacetyl Nitrate for Australian Fires, Standard Product contains the vertical distribution of the retrieved atmospheric state of peroxyacetyl nitrate (PAN), formal uncertainties, and diagnostic information measured by the CrIS instrument on the Suomi-NPP satellite. This product focuses on the Australia region (60S-0S; 100E-177.5E) for the time period from 2019-11-01 to 2020-01-31, during the outbreak of Austrailan wildfires. The NASA TRopospheric Ozone and Precursors from Earth System Sounding (TROPESS) project, uses an optimal estimation algorithm, known as the MUlti-SpEctra, MUlti-SpEcies, Multi-SEnsors (MUSES).The data files are written in the netCDF version 4 file format, and each file contains one day of data. The data have a spatial resolution of 14 km (CrIS nadir FOV), and are reported at 16 vertical levels from the surface to 0.1 hPa. The principal investigator for the TROPESS project is Kevin W. Bowman.

This dataset details the proportion of the geographic range of 26,062 Australian plant species burnt in the 2019-2020 megafire; threatened listing status on state and Commonwealth threatened species legislation; species endemic status in each state/territory according to the Australian Plant Census; and risk ranking for exposure to high fire frequency (short intervals between fires) and cumulative impacts of fire (populations dominated by immature individuals). Further details are provided in the users should consult and cite the associated paper:

Gallagher, R.V., Allen, S., MacKenzie, B.D.E., Yates, C.D., Gosper C.R, Keith, D.A., 29 Merow, C., White, M., Wenk, E., Maitner, B.S., He, K., Adams, V.M. & Auld, T.D. (2021) High fire frequency and the impact of the 2019-2020 megafires on Australian plant diversity. Diversity & Distributions.

Usage Notes Species names were listed as accepted in the Australian Plant Census as of July 2020. Range data was sourced from three lines of evidence: (1) cleaned occurrence data (latitude-longitude point locations) associated with digitised herbarium specimens accessed from the Australasian Virtual Herbarium (https://avh.ala.org.au/) via the Atlas of Living Australia Application Programming Interface (https://api.ala.org.au/) in July 2020; (2) range mapping built from Poisson Point Process models, range bagging and area of occurrence (AOO) calculations; and (3) maps for Species of National Environmental Significance (SNES) for species listed on the Commonwealth Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act) available from the SPRAT database (http://www.environment.gov.au/cgi-bin/sprat/public/sprat.pl). Details of the building of range models are available in the paper associated with this dataset.

The spatial extent of the 2019-2020 fires was quantified using the National Indicative Aggregated Fire Extent Dataset (NIAFED; https://data.gov.au/dataset/ds-environment-9ACDCB09-0364-4FE8-9459-2A56C792C743/details?q=). Geographic ranges were intersected with the NIAFED dataset and proportion of burnt range is reported in the columns: "Proportion of range map burnt", "Proportion of SNES range map burnt (EPBC Act species only) ", and "Proportion of point locations burnt".

Exposure to high fire frequency and the cumulative fire risk rankings were created by intersecting ranges with fire history data for the last 5 years (non-woody species), 15 years (woody species) and 50 years (rainforest trees) and trait data on fire response. Species level data on growth form and fire response traits (resprouter, obligate seeder) were sourced from the AusTraits database (https://www.biorxiv.org/content/10.1101/2021.01.04.425314v1).

The annual spatial extent of fires between September-March between 1969-2018 was quantified by combining data from remote sensing and state agency fire history databases. Remotely sensed data on fire extent in each season between 2003 and 2016 was accessed from the Global Fire Atlas https://www.globalfiredata.org/fireatlas.html and – using the same methods – fire extent data was created for the 2017 and 2018 seasons using imagery from the MODIS product (MCD64A1). Alternate data on annual fire history were accessed under license from environment agency databases in three Australian states – New South Wales (NSW National Parks and Wildlife Service Fire History – Wildfire and Prescribed Burns dataset https://data.nsw.gov.au/data/dataset/1f694774-49d5-47b8-8dd0-77ca8376eb04), Western Australia (Western Australian Department of Biodiversity, Conservation and Attractions Fire History dataset (1969-2020)), and Victoria (Victorian Department of Environment, Land, Water and Planning Fire History dataset). Methods for assigning species ranks are provided in Gallagher (2020) https://www.environment.gov.au/system/files/pages/289205b6-83c5-480c-9a7d-3fdf3cde2f68/files/final-national-prioritisation-australian-plants-affected-2019-2020-bushfire-season.pdf

All correspondence about the dataset should be directed to rachael.gallagher@mq.edu.au. Additional data about fire impacts and threat interactions, as well as code for anlayses, are also available.

Geospatial layer of predicted concentration of red dust generated by bushfires in Australia in late 2019. Layer resolution is 10 m and generated from Sentinel-2 imagery from 1 November 2019 to 31 March 2020 over areas of permanent snow and ice from GLIMS (see geopackages).

This map describes Australian fuel type classes and their estimated extents using the Bushfire Fuel Classification (BFC) framework of Hollis et al. (2015) and Cruz et al. (2018). The BFC is a hierarchical, structure-based classification system for fuel complexes, enabling distinct fuel extents and characterisations to be mapped directly onto fire behaviour models. The map has been generated using nationally consistent, open-source datasets, predominantly derived from remotely sensed data, to quantitatively characterise vegetation life forms, height and foliage cover. The automated production of this map allows for rapid versioning as data inputs are enhanced. This version of the BFC fuel types map integrates multiple spatial and temporal resolution datasets, ranging from 10 to 500 metres. For a detailed methodology, refer to Joshi et al. (2025). Lineage: Fuel types have been created by combining height and foliage cover data using a rule-based method according to the BFC framework of Hollis et al. (2015) and Cruz et al. (2018). Cover data were taken from the woody and grass foliage cover data derived from Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery averaged from 2000 to 2021, as described by Donohue and Renzullo (2025). Vegetation height and cover profile data came from Scarth et al. (2019) and Lang et al. (2023), which incorporate observations from multiple optical and radar satellite-based sensors. Plantation data were included from the National Plantation Inventory (NPI) (ABARES, 2022). Croplands, Horticulture, and Wetlands were included from the Australian Land Use Mapping (ALUM) (ABARES, 2021). Sedgelands are from the National Vegetation Information System (NVIS) (DCCEEW, 2020). Spinifex is derived from multiple MODIS products; detail method is given in Joshi et al. (2025). Built-up is from the polygon footprints provided by Australian Housing Data Analytics Platform (AHDAP, 2022). Bare ground and Water are extracted from the MODIS time-series from 2000-2021, as given in Guerschman et al. (2018) and Donohue et al. (2022) respectively.

Important Disclaimer: CSIRO advises that the information contained in this dataset comprises general statements and information based on scientific research. The user is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. No reliance or actions must therefore be made on that information without seeking prior expert professional, scientific and technical advice. To the extent permitted by law, CSIRO (including its employees and consultants) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this publication (in part or in whole) and any information or material contained in it.

This paper focuses on the performance of the far-right community in the Australian Twittersphere during two information crises: the 2019-20 Australian bushfires and the early months of the 2020 COVID-19 pandemic. Using a mixed method approach to analysing the performance of far-right accounts active in both crises, and using an information disorder index to estimate the quality of information being shared on Twitter during the two events, we found that far-right accounts moved from the periphery of these disaster-driven conversations during the Australian bushfires to assume a more central location during the COVID-19 pandemic. We argue that an increase in information disorder and overperformance of far-right accounts during COVID-19 is suggestive of an association between the two, which warrants further investigation.

The TROPESS CrIS-SNPP L2 Ozone for Australian Fires, Standard Product contains the vertical distribution of the retrieved atmospheric state of ozone (O3), formal uncertainties, and diagnostic information measured by the CrIS instrument on the Suomi-NPP satellite. This product focuses on the Australia region (60S-0S; 100E-177.5E) for the time period from 2019-11-01 to 2020-01-31, during the outbreak of Austrailan wildfires. The NASA TRopospheric Ozone and Precursors from Earth System Sounding (TROPESS) project, uses an optimal estimation algorithm, known as the MUlti-SpEctra, MUlti-SpEcies, Multi-SEnsors (MUSES).The data files are written in the netCDF version 4 file format, and each file contains one day of data. The data have a spatial resolution of 14 km (CrIS nadir FOV), and are reported at 26 vertical levels from the surface to 0.1 hPa. The principal investigator for the TROPESS project is Kevin W. Bowman.

Paper Abstract:

The 2019–2020 Kangaroo Island bushfires in South Australia burned almost half of the island. To understand how to avoid future severe ‘mega-fires’ and how vegetation may recover from 2019–2020, we can utilize information from the bulk of historical fires in an area. Landsat time-series of vegetation change provide this opportunity, but there has been little analysis of large numbers of fires to build a landscape-level understanding and quantify drivers in an Australian context. In this study, we built a yearly cloud-free surface reflectance normalized burn ratio (NBR) time-series (1988–2020) using all available summer Landsat images over Kangaroo Island. Data were collected in Google Earth Engine and fitted with LandTrendr. Burn severity and post-fire recovery were quantified for 47 fires, with a new recovery metric facilitating comparison where fire frequency is high. Variables representing the current burn, fire history, vegetation structure, and topography were related to severity and yearly recovery with random forest and bivariate analysis. Results show that the 2019–2020 bushfires were the most widespread and severe, followed by 2007–2008. Vegetation recovers quickly, with NBR stabilizing ten years post-fire on average. Severity is most influenced by fire frequency, vegetation capacity and land use with more severe burns in nature conservation areas with dense vegetation and a history of frequent fires. Influence on recovery varied with time since fire, with initial (year 1–3) faster recovery observed in areas with less surviving vegetation. Later (year 6–10) recovery was most influenced by a variable representing burn year and further investigation indicates that precipitation increases in later post-fire years likely facilitated faster recovery. The relative abundance of eucalypt woodlands also has a positive influence on recovery in middle and later years. These results provide valuable information to land managers on Kangaroo Island and in similar environments, who should consider adjusting practices to limit future mega-fire risk and potential ecosystem shifts if severe fires become more frequent with climate change.

Data details:

See paper: Remote Sensing | Free Full-Text | Contextualizing the 2019–2020 Kangaroo Island Bushfires: Quantifying Landscape-Level Influences on Past Severity and Recovery with Landsat and Google Earth Engine (mdpi.com)

See code on GitHub: ZZMitch/KangarooIslandFireHistory_1988to2020: Code from "Contextualizing the 2019–2020 Kangaroo Island Bushfires: Quantifying Landscape-Level Influences on Past Severity and Recovery with Landsat and Google Earth Engine" (RS, 2020) (github.com)

If you use these data, please reference:

Bonney, M.T., He, Y., Myint, S.W., 2020. Contextualizing the 2019–2020 Kangaroo Island bushfires: Quantifying landscape-level influences on past severity and recovery with Landsat and Google Earth Engine. Remote Sensing 12(23), https://doi.org/10.3390/rs12233942.