Anthropogenic land-use change, driven by urban development, agricultural intensification, and resource needs, has led to the deforestation and degradation of ecosystems globally (Foley et al. 2005; Shellberg et al. 2016). Over 2.3 million square kilometers (230M hectares (ha)) of global forest were lost between 2000–2012, mostly in the tropics, while only 0.8 million square kilometers (80M ha) were gained through plantations or regrowth (Hansen et al. 2013). This extreme change has modified natural nutrient cycles, hydrological cycles, and regional climates (Postel et al. 1996; Kalnay and Cai 2003; Foley et al. 2005; Deo et al. 2009; Haddad et al. 2015). Repercussions of these changes have led to global increases in droughts, large wildfires, and floods at unprecedented scales (Westerling et al. 2006; Min et al. 2011; Baranowski et al. 2021; Department of Agriculture Water and the Environment 2021).
Human-driven land conversion has also removed or modified important habitats and resources for wildlife species. This often fragments native habitats (Tulloch et al. 2016; Shackelford et al. 2018) and can introduce exotic species, which create challenges for native species’ survival (Brook et al. 2003; Woinarski et al. 2015). Singapore has lost 95% of its primary forests since 1819 (Corlett et al. 1992) causing substantial extinction of forest species, particularly forest specialists (Brook et al. 2003). Similarly, Borneo has been a hot spot for land cover change, with intense conversion to oil palm plantations and logging causing a loss of at least 30% of native forest since 1973 (Gaveau et al. 2014). As a result, the nation has experienced significant declines in species richness, diversity, and animal abundance (Edwards et al. 2014). Other impacts on wildlife include range contraction (Shackelford et al. 2018), changing species compositions (Crooks 2002), changing species’ ecology and trophic interactions (Hebblewhite et al. 2005), and disease emergence (Patz et al. 2004; Gottdenker et al. 2014; Macdonald and Mordecai 2020).
As humans have invaded the natural world, the frequency of human-wildlife interactions has increased (Patz et al. 2004; Soulsbury and White 2015). These novel interactions carry the threat of zoonotic disease emergence, or emergence of pathogens that can cross-species barriers, representing a serious risk to global public health for humans and animal populations in close contact with humans. Many zoonotic viruses that cause severe diseases in humans, such as Ebola virus, Marburg Virus, Hendra virus, Nipah virus, and SARS, have been the consequence of spillover events from bats (Plowright et al. 2015; Ruiz-Aravena et al. 2021). SARS-CoV-2, which likely spilled over from bats, has caused over 6 million human deaths since its emergence and has had immense impacts on populations globally (Ruiz-Aravena et al. 2021; World Health Organization 2022). The World Health Organization (WHO) has classified several of these bat viruses as priority diseases for research (World Health Organization 2018), demonstrating an urgency to understand the mechanistic drivers of emergence for these and other zoonotic diseases.
Hendra virus is a henipavirus in the Paramyxoviridae family and is a priority disease designated by the WHO (Mahalingam et al. 2012; World Health Organization 2018). It was identified in 1994 in Queensland, Australia when twenty horses and one person died following a spillover event (Murray et al. 1995). Spillover likely occurs when a horse inhales or ingests the excreta of infected bats; all known Hendra virus spillovers have been the result of primary transmission from bat to horse (Plowright et al. 2011; Smith et al. 2014). Infected horses show respiratory and neurological symptoms and can transmit the virus to other horses, humans, and domestic animals (Murray et al. 1995; Plowright et al. 2011). There have been 107 horse cases, all fatal, and seven human cases, of which four were fatal (Queensland Government Business Queensland 2021). Another ecologically and genetically similar henipavirus, called Nipah virus, has caused hundreds of human cases with high mortality rates (range 40–70%) and is now endemic in southeast Asia (Epstein et al. 2020). Understanding the environmental drivers of Hendra virus spillover is a critical step towards characterizing the broader mechanisms of henipavirus spillover from bats.
Hendra virus prevalence in flying foxes has been strongly correlated with the relative abundance of black flying foxes (Pteropus alecto) and spectacled flying foxes (P. consipillactus) and weakly correlated with the abundance of grey-headed flying foxes (P. poliocephalus) and little red flying foxes (P. scapulatus) (Smith et al. 2014; Páez et al. 2017). Flying foxes are typically nomadic, nocturnal nectivores and frugivores that can roost in extremely large groups, some up to 100,000 individuals, in canopy vegetation during the day (Eby 1991; Eby et al. 1999). They commonly forage on small fruits and nectar from members of the Proteaceae and Myrtaceae families, notably Eucalyptus, Melaleuca, Banksia, and Corymbia species (Eby 1991; Bell et al. 2021). The climatic conditions that prompt eucalypt reproduction varies by location, soil type, and season for eucalypt (tribe Eucalypteae within Myrtaceae) species (Law et al. 2000; Eby and Law 2008), creating a patchwork of resources across the landscape. Corresponding to this irregular flowering phenology, flying foxes historically migrated long distances across their range to track the seasonal flowering of eucalypts (Eby 1998; Roberts et al. 2012). This extreme mobility of flying foxes (Welbergen et al. 2020) makes them key pollinators and seed dispersers, important for the maintenance of forest ecosystem health (Marshall 1985; Eby 1998). However, seasonal variation in eucalypt blooming creates a resource bottleneck in winter months. This likely causes nutritional stress in flying fox populations and is hypothesized to play a role in increasing Hendra virus spillovers (Plowright et al. 2008; Edson et al. 2019), as most spillover events have occurred in winter.
Australian flying fox roosting behavior has changed considerably in recent decades. In the mid-2000s, it was observed that small sub-populations of flying foxes were forgoing their historic migratory behavior and forming camps in urban or peri-urban areas (Van Der Ree et al. 2006; Plowright et al. 2011). Recent studies on flying fox roost occupation show black and grey-headed flying foxes most commonly roost in urban and agricultural areas, with few roosts in protected areas (Timmiss et al. 2020). This change in roosting behavior for grey-headed flying foxes is thought to be due to the increased spatiotemporal availability of food in urban environments (Meade et al. 2021; Yabsley et al. 2021) in parallel with extensive habitat loss (Tidemann 1999). However, the impact of anthropogenic land pressure on black flying fox populations has been understudied given the species’ importance in pollinating eucalypt forests and role in Hendra virus spillover.
In bioregions where black flying fox populations were observed from 2012 to 2020, we quantify the changes in native remnant forest extent, native winter habitat extent, human population size, and black flying fox populations. We measure changes by bioregion, which are the primary levels of biodiversity classifications in Queensland (Queensland Government 2014), and within 50km buffers around roosts, which is the nightly maximum foraging area (New South Wales Department of Planning and Environment 2020). We hypothesize that winter habitat loss and remnant forest loss were largest in bioregions with the greatest amount of human population growth. Characterizing these spatiotemporal dynamics will inform our understanding of the impact of land-use change on black flying fox distribution and subsequent Hendra virus spillover in sub-tropical Queensland.