Why crop disease prediction systems need to be constantly updated, involving experts from various fields
Potato late blight Phytophthora infestans, is one of the most dangerous diseases of the culture, which requires the constant use of pesticides for prevention.
The development of the disease largely depends on weather conditions, and therefore, several forecasting schemes have been developed around the world in order to reduce the costs of farmers to fight the disease.
The Irish Rules, developed in the 1950s and calibrated to weather forecasts, potato production practices and pathogen pressure P. Infestans still serve as the basis for recommendations to farmers.
However, since the advent of the Irish Rules model, numerous changes have occurred in the composition and dynamics of late blight. A group of Irish scientists from the Teagasc Crop Research Center, the University of Maynooth and the Irish Meteorological Service tested the model in modern conditions and proposed a number of amendments.
Late blight develops and becomes more aggressive
Late blight (or late rot of potato) is one of the most destructive diseases of the potato culture due to the fast reproductive cycle of the pathogen and aggressiveness. In the absence of control, late blight can quickly lead to the complete destruction of the crop, both in the field and during storage after harvest.
In Ireland, historical outbreaks of potato late blight have had a significant cultural and economic impact, leading to mass starvation and subsequent migration of a large part of the population during the 1840s.
Now in Ireland alone, about 5 million euros are spent annually on fungicides to combat late blight of potatoes, while worldwide, the cost of controlling the disease and losing crops exceeds 1 billion euros per year.
The rate of development of the epidemic is largely dependent on the weather, with temperature, relative humidity and precipitation being the most important variables, the latter two factors being closely related.
Long periods of wet and cool weather create favorable conditions for sporulation of pathogenic microorganisms that are carried by rain and wind.
The disease causes damage both indirectly and directly: indirectly, by reducing the photosynthetic surface, and directly when zoospores washed off from the leaves infect tubers in the ground.
Since the late 1970s, increasing globalization has led to worldwide migration of pathogen genotypes, which has led to a shift in dominant, older clonal lines or genotypes commonly called US-1, and has contributed to the development and spread of new lines, some of which show increased aggressiveness.
New genotypes have been discovered in Ireland and have been recorded more frequently in recent years. In addition, most of the potato production in Ireland is based on more susceptible potato varieties to new versions of pathogens.
The diversification of late blight pathogens, combined with the impact of climate change, makes control difficult and the risk of epidemics is higher. As a result, potato growers regularly apply intensive fungicide protection - in Western Europe it reaches more than 10 applications per season.
The need to develop models for predicting potato late blight has long been recognized as an important tool for combating the disease, which is motivated by both environmental and economic factors.
In response to environmental problems arising from the increasing use of agrochemicals, European Community Directive 128/2009 on the sustainable use of pesticides contains strict guidelines on the sustainable use of plant protection products in order to reduce risks to human health and the environment.
Reliable disease forecasting makes it possible to reduce crop losses and yield under adverse weather conditions, as well as substantiate the actual rationale for the use of plant protection products in accordance with national and international rules.
Forecasting systems cannot live in the past and other people's data
At its core, agricultural disease prediction systems use algorithms, both fundamental and empirically based, to predict disease cycles.
Fundamental models are developed on the basis of laboratory experiments in chambers, greenhouses or fields with a controlled environment and describe one or more segments of the relationship between the host and the parasite, subject to environmental influences.
Initially, the development of forecast models for crop diseases was focused mainly on the study of weather phenomena to predict the development and onset of epidemics and was mainly empirical, based on the duration of weather events outside the threshold values and the vegetative stage of plants.
Recently, fundamental approaches have increasingly been used to cover the more complex components of epidemics, along with agricultural practices and chemical protection.
Austin Bourke, one of the pioneers in predicting potato late blight, has developed a PLB model called the Irish Rules. This model sought to incorporate knowledge about the life cycle of the disease, as opposed to a purely empirical approach. For example, the selection of suitable weather criteria for the development of the disease was determined based on documented laboratory experiments, and not from a retrospective analysis of historical weather during disease outbreaks.
However, more recently, as part of a pan-European initiative, a theoretical comparison with several European risk prediction models showed that the Irish model provides farmers with the lowest risk assessment due to its strict criteria.
Field evaluations of the effectiveness of the Irish model showed that control according to its data leads to a significant reduction in the use of fungicides, but with poor control of late blight compared to other Negfry (or DSS) farmers decision support systems or the usual fungicide protection practice.
But if earlier it was “convenient” for farmers to base their decisions on DSS recommendations to justify an increase in the number of chemical treatments, now there is another trend - they are trying to increase economic benefits by reducing costs and adhering to the pesticide policies required by supermarket chains.
“Therefore, now is the time to revise the 'Irish Rules' and conduct an assessment of the system in order to clarify the rules in light of recent changes. It is necessary to provide an integrated, systematic and transparent method for the operational application of the system in the context of changes in the epidemiology of the disease and the strengthening of regulation (market / policy), ”the scientists write in their work.
“Contrary to recent reports, we have found that the risk of late blight epidemics remains low below 12 ° C. With more complete outbreak data and a deeper understanding of the pathogen population, we believe that the temperature threshold in the model could potentially be increased from 10 ° C. C to 12 ° C, providing more opportunities to reduce pesticide use, ”they note.
“The risk prediction model is only useful if it provides the same level of protection as standard practice, while reducing the required costs and labor hours ... Currently, the intervals between sprayings vary from 5 to 7 days in the Irish conditions that we took into account in this study.
We suggest that planting begins the day after the average daily soil temperature exceeds 8 ° C for three consecutive days after April 1. This is a common practice in Ireland as recommended by the Teagasc national advisory body. Farmers usually begin treatment with fungicides as soon as germination reaches 50% and continues until the aboveground part completely dies, usually three weeks after drying. Here we assume that the growing season lasts 120 days. However, pesticide protection continues for these three weeks until the potato aboveground is dried.
We have shown that, on average, the use of risk prediction models makes it possible to reduce the consumption of fungicides in comparison with the standard practice of Irish manufacturers. Possible reductions in dose and number of treatments show differences over the study period. This reflects the nature of agricultural production and further reinforces the need for an integrated pest and disease management approach to define treatment intervals.
Plant disease prediction models are often evaluated by researchers who developed them and are used without calibration in agroecosystems other than those for which they were developed.
The results showed that it is necessary to revise the parameters of the “Irish rules” model for various ecosystems and operational capabilities, as well as for the operational use of the model.
We recommend reducing the threshold values for relative humidity from 90% to 88% and the duration of sporulation from 12 to 10 hours; and introduce the adoption of an additional indicator of leaf moisture, including both precipitation (≥0,1 mm) and relative humidity (≥90%), "the authors of the work conclude.
Read more: https://www.agroxxi.ru/