Current and future risks of drought-induced mortality in Pinus radiata plantations in New South Wales, Australia. Issue 4 (2nd October 2022)

• Record Type: Journal Article
• Authors: Carnegie, A. J.; Kathuria, A.; Nagel, M.; Mitchell, P. J.; Stone, C.; Sutton, M.
• Is Part Of: Australian forestry. Volume 85:Issue 4(2022)
• Page Start: 161
• Page End: 177
• Publication Date: 2022-10-02
• Subjects: climate change -- diseases -- drought -- forest health surveillance -- pests -- spatial modelling; Forests and forestry -- Australia -- Periodicals; Forests and forestry -- Australasia -- Periodicals; 634.90994
• DOI: 10.1080/00049158.2022.2145722
• Languages: English
• ISSNs: 0004-9158

Abstract

ABSTRACT: Drought is a regular feature of Australian landscapes, and its intensity and frequency are likely to increase in a changing climate. Land managers are grappling with managing the impacts of drought, with large-scale die-offs occurring more frequently in forests globally. Drought-induced tree mortality has caused major impacts in Pinus radiata plantations in New South Wales, Australia, with extended drought, heatwaves and pest and disease attacks all recognised as predisposing, inciting or contributing factors. The extent and severity of drought-induced tree mortality has been mapped across the Pinus plantation estate in New South Wales annually since 1996. In this study, we used this long-term empirical data to develop a model of drought risk for P. radiata plantations. Using random forest, we identified site index, annual temperature, annual rainfall, elevation and increasing number of hot days (above 20°C and 35°C) as the influencing variables associated with drought-induced tree mortality. We then used this model to look at the risk of drought-induced tree mortality under climate-change scenarios in 2050 and 2070. Although forest managers already understand the drought risk in their estates, we developed an empirical model and produced GIS layers at high resolution (100 m) to assist in more accurately and effectively managing drought. The accuracy and precision of our model (overall accuracy 89.2%, kappa 0.75) enables forest managers to include it in their decision-making in the management of the potential impacts of drought on the current plantation estate (e.g. via modified silvicultural regimes) as well as in the future (e.g. modified silvicultural regimes or planting drought-tolerant genotypes).