Climate model gets the measure of myrtle rust

Myrtle rust never sleeps and, for a while it seems, neither did Plant & Food Research scientist Dr Rob Beresford, who spent the month of June poring through research articles, crunching data and creating mathematical formula to better gauge what the disease may mean for New Zealand.

The end result of the sleep deprived month was the Myrtle Rust Risk Model, specifically designed to understand and predict how myrtle rust will behave under New Zealand conditions.

The model is already proving its worth, with the Ministry for Primary Industries using it to help inform response activities, such as targeted surveillance for the disease.

“The model has three key attributes,” says Dr Beresford. “It warns when the weather is suitable for any spores in the air to infect susceptible plants; it predicts the time from when infection occurs to when rust symptoms may appear; and it assess the suitability of conditions for spores to be produced from infected plants that are showing symptoms.”

With no history of myrtle rust in New Zealand prior to its arrival in May, developing the model was not easy due to a large number of unknowns. 

Dr Beresford’s first step was to dig deep into scientific literature and recorded observations from countries where the disease is already established, such as Brazil, the US (Hawaii) and Australia.

“Although the overseas research is tremendously useful, you can’t assume that myrtle rust will behave in New Zealand in ways observed in other countries with similar climates,” says Dr Beresford.

“New Zealand has its own seasonal weather patterns. Moreover, the genetic differences between plant species in the myrtle family could influence susceptibility, just as there can be differences in the strains of the rust pathogen itself. So, it’s very complex.

“All these things have to be calculated and factored in to the model, with mathematical parameters set to represent things such as plant susceptibility, temperature range and humidity.

“Essential to doing this well is having a good understanding of the biology of the disease and host plant species.”

The risk model is particularly unique for the fact that it simulates the biology of the disease at a fine scale of time and space. Additionally, thanks to NIWA’s sophisticated weather analysis and prediction maps, in combination with its climate-data mapping skills, the NIWA data can be factored into the model hourly, allowing for day-to-day measurability and reporting.

This model can work in conjunction with other climate models developed for myrtle rust that take a more general, broad-brush climate matching approach or rely on long-term weather data.

“The next step to further refine the model is to do more in-depth research into host plant susceptibility,” says Dr Beresford. “This means we can tweak the model from reporting relative risk to something even more definitive.”

Funding for the development of the model came from the Ministry for Primary Industries. Plant & Food Research is currently collaborating with NIWA on mapping the risk of myrtle rust infection in different regions.