Immersive landscapes: modelling ecosystem reference conditions in virtual reality

Context Understanding the variability and dynamics of ecosystems, as well as their responses to climate or land use change, is challenging for policy makers and natural resource managers. Virtual reality (VR) can be used to render virtual landscapes as immersive, visceral experiences and communicate ecosystem dynamics to users in an effective and engaging way. Objectives To illustrate the potential and believability of VR, a team of landscape ecologists and immersive visualisation researchers modelled a reference Australian Box Gum Grassy Woodland landscape, an endangered eucalypt woodland ecosystem that is difficult to observe in its pre-European colonisation form. Methods We document considerations for designing the immersive virtual landscape, including the creation of animated three-dimensional (3D) plants that alternate between the seasons, and soundscapes that change through the course of a simulated day. We used a heuristic evaluation with experts to assess the potential of immersive VR landscape modeling. Results This cross disciplinary collaboration resulted in a VR experience that was evaluated in a series of meetings by 27 ecologists and managers in biodiversity conservation, many of whom were familiar with Box Gum Grassy Woodlands. 88% of participants stated that the simulation was believable and participants thought that virtual landscapes held great potential for education, public engagement and land management. Conclusions Possible future directions include open-source libraries of ecological 3D models, and the visual simulation of historic landscapes and future climate change scenarios.

Automated summary

The use of virtual reality to model endangered ecosystems, such as the Australian Box Gum Grassy Woodland, has shown potential for effectively communicating ecosystem dynamics. The immersive virtual landscapes were well received by ecologists and managers, demonstrating their potential for education, public engagement, and land management. Potential future directions include developing open-source ecological 3D models and simulating historic landscapes and future climate change scenarios.