Predicting the climatic risk of myrtle rust during its first year in New Zealand

Abstract

After the first detection of myrtle rust (Austropuccinia psidii) on mainland New Zealand in May 2017, the Ministry for Primary Industries sought information about how weather conditions would affect regional and seasonal risk of disease establishment to help plan the incursion response. Using internationally published information, a pathogen-process model was developed to predict infection, latent period and sporulation in relation to weather variables (temperature, relative humidity and solar radiation). This Myrtle Rust Process Model (MRPM) was implemented by the National Institute of Water and Atmospheric Research Limited using numerical weather model data to produce weekly maps of potential risk. Predicted risk was greatest in northern North Island and decreased further south, but was still substantial in coastal areas of the north-western South Island during summer and autumn. Risk was low in southern coastal areas of the South Island and the lowest risk occurred in mountainous areas, particularly in the South Island. Retrospective analysis of surveillance data showed that the MRPM accurately predicted geographic risk and it is currently in use for tactical planning of incursion surveillance and organism management.

Symptom development and latent period of Austropuccinia psidii (myrtle rust) in relation to host species, temperature, and ontogenic resistance

“…See Equation for the fitted function and Table for the parameter estimates. (d) Compares latent period responses for each species with the preliminary estimate used in the Myrtle Rust Process Model (MRPM; Beresford et al ., )…”

“… See Equation for the fitted function. Also shown are the preliminary parameter values for the MRPM (Beresford et al , ) and the corrected values derived from Lophomyrtus sp. in this study. …”

“…On highly susceptible species, uredinial epidemics cause death of leaves, stems, flowers, and fruit, and, in severe cases, plant death (Glen et al , ). It has been implicated in localized species extinction in some natural habitats in Australia (Carnegie et al , ; Pegg et al , ) and is a cause for concern in New Zealand, where it arrived in 2017 (Beresford et al , ).…”

“…Some other key species, such as Leptospermum scoparium ( mānuka ), which is important to the honey industry, appeared to be less at risk (Beresford et al , ). Climatic modelling showed that myrtle rust infection risk was greatest in the northern North Island, decreasing southward and at higher altitude, but with significant risk also in northern and western areas of the South Island (Beresford et al , ).…”

“…The latent period for rust fungi, unlike the infection process, is not dependent on external wetness or high relative humidity (Roelfs et al , ). Accurate A. psidii latent period information is also needed to calibrate the Myrtle Rust Process Model that was developed for climatic risk prediction when A. psidii first arrived in New Zealand (Beresford et al , ). The parameters for the latent period submodel were estimated from limited published information available at the time, and definitive data for different hosts over a range of temperatures are required to verify the accuracy of the latent period function.…”