Designing profitable and climate-smart farms using virtual reality

example, the New Zealand Government’s Climate Change Response Amendment Act 2019 will require farmers to reduce all greenhouse gasses (except biogenic methane) to zero, and bio- genic methane emissions by more than 24% (compared to 2017 levels) by 2050. Achieving the target is expected to have a significant impact on livestock farms because approximately half of national carbon emissions originate from animal farming [7]. One option for New Zealand farmers is to plant trees across parts of their farm with the intention of offsetting their carbon emissions and potentially earn credits via New Zealand’s 2008 Emissions Trading Scheme (ETS) [8, 9]. Large scale conversion of farmland to forestry following escalating global carbon credit prices has generated anxiety within rural communities [10]. Some commentators have argued that the replacement of livestock farms with forest will lower community resilience and vitality [11]. To date large-scale land conversion to forestry has been primarily confined to low-pro- ducing farms that were previously generating low financial returns [10], new regulatory set- tings and a high carbon price are expected to encourage the conversion of highly productive land into trees [9]. This paper describes the development of a VR application that is designed to help livestock farmers create an integrated forestry-livestock system. We address the current challenge live- stock farmers face to identify the most suitable locations on the property where trees can be

planted so that they can meet their emissions objective (i.e., a carbon neutral enterprise) with- out compromising the profitability of their businesses, or affect amenity values of the property (such as aesthetic amenities, recreational access or sacred indigenous sites). In practice, the VR application enables farmers to ‘stand in their farm’ (a virtual environment) and virtually plant different tree species in different locations. Different tree species have different carbon seques- tration rates, aesthetic, and establishment costs. Therefore, the impacts of tree planting depend on the selected tree species and the planting location. Farmers can subsequently, visualise the resulting landscape, and assess what this means for farm profitability and their emissions tar- gets. It also allows the comparison of different tree planting strategies. Our application has been built to a to prototype stage. The project brought together a multidisciplinary team with capabilities in VR, remote sens- ing, farm economics, carbon accounting and product design. The team produced a high-reso- lution landscape visualisation that enables the user to identify planting areas based on the spatial variability of pasture production and in doing so, quantify the carbon sequestration and profitability outcomes from different planting scenarios. The VR environment also enables users to visualise the aesthetic changes made to the farm under each scenario. The prototype described in this paper has been developed in Unreal Engine1 and is set on an existing 400 ha

New Zealand beef and sheep farm in Canterbury, New Zealand. Several digital decision support tools are available in New Zealand that can be used to assist farmers to design carbon neutral farm enterprises [e.g., 12, 13]. We also acknowledge the pres- ence of technical guidance documents that are freely available in New Zealand to help farmers understand the implications of converting land to forestry, such as look-up tables and techni- cal guides. However, our new application complements the existing information by adding VR technology, a novel communication style seldomly used to convey scientific data [14]. We take this approach because the ability of humans to understand and assimilate information from pictures is greatly enhanced compared to conveyance through traditional scientific representa- tions such as graphics and numbers [3, 15–17]. Therefore, we believe our approach will sup- port more effective conversion of information into user knowledge and facilitate the implementation of new practices. So far, the use of visualisation in landscape planning has largely been limited to a supporting role for participatory processes or to help policy making [18, 19]. PLOS ONE Designing profitable and climate-smart farms using virtual reality PLOS ONE | https://doi.org/10.1371/journal.pone.0286723 June 2, 2023 2 / 16 Employment through AgResearch’s research programme ‘New Zealand Bioeconomy in the Digital Age’. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.