The panzootic highly pathogenic avian influenza (HPAI H5N1) has been detected on the Australian mainland, with incursions from the sub-Antarctic region increasing the threat to domestic wildlife and poultry. The authors aimed to produce a spatially explicit prediction of HPAI H5N1 poultry outbreaks at the local government area (LGA) level across Australia, to inform targeted surveillance and biosecurity planning.
The article is a preprint posted July 28, 2026, and has not been certified by peer review.
The core environmental suitability for HPAI H5N1 occurrence was estimated using a Maximum Entropy (MaxEnt) species distribution modelling approach. The MaxEnt output formed a continuous suitability layer across Australia that represents areas environmentally favorable for HPAI H5N1 occurrence according to the model inputs.
This suitability layer was treated as one of six predictor inputs for the final risk mapping at LGA resolution.
Five additional spatial predictor layers were integrated with the MaxEnt suitability surface:
Each layer reflects a different pathway or driver potentially influencing the risk of poultry infection: environmental suitability for the virus, presence and movements of wild bird reservoirs or vectors, and domestic poultry distribution and density.
The six predictor layers were aggregated and averaged to generate a composite risk estimate for each LGA. This produced an HPAI H5N1 risk map intended to indicate relative risk levels for poultry outbreaks across Australian jurisdictions at the local government scale.
The aggregation method combined environmental, ecological, and agricultural predictors into a single spatial risk product to support decision-making at substate scale.
Model results identified New South Wales (NSW) and Victoria (VIC) as having the highest predicted risk of HPAI H5N1 poultry outbreaks, despite observed incursions to date being concentrated in Western Australia (WA) and South Australia (SA). Additional areas of high predicted risk were identified in WA, SA, and Tasmania (TAS).
By contrast, the Northern Territory (NT) and extensive areas of Queensland (QLD), WA, and SA were predicted to be at low risk according to the composite LGA-level map.
These geographic distinctions reflect the combined influence of environmental suitability, wild bird presence and flyways, and poultry population metrics as represented in the six-layer model.
The spatially explicit risk map is proposed as a framework to support targeted surveillance and preparedness activities. Practical uses include prioritising LGAs for monitoring of wild birds and poultry, directing allocation of biosecurity resources, and informing contingency planning for potential outbreaks.
By highlighting higher-risk LGAs in NSW, VIC, WA, SA, and TAS, the model suggests where intensified surveillance and preventive measures may be most valuable to mitigate the impact of future poultry outbreaks.
The article is a preprint and has not undergone peer review. Details on model performance metrics, specific environmental variables used in the MaxEnt model, temporal considerations of bird movements, and the exact aggregation weighting for the six layers were not reported in the provided source abstract. Users of the risk map should interpret results in the context of these reporting limitations and the preprint status.
The authors declared funding from the NHMRC, with C Raina MacIntyre funded by an NHMRC Investigator Grant. A competing interest statement notes that C Raina MacIntyre is funded by NHMRC and the Medical Research Futures Fund and is Founding Director of EPIWATCH Global Pty Ltd.
The study presents a composite six-layer spatial model that combines a MaxEnt-derived environmental suitability surface with wild bird abundance and movement layers and poultry density metrics to predict HPAI H5N1 poultry outbreak risk at the LGA level across Australia. The model highlights NSW and VIC as the highest predicted risk jurisdictions and identifies other high- and low-risk regions across the country. The authors propose the map as a tool to guide targeted surveillance, preparedness, and biosecurity actions to reduce the impact of future outbreaks.
Note: the source is a preprint and additional methodological and performance details were not reported in the abstract provided.