The health report - do you need to spray for pulse Botrytis diseases? Ask the new data logger

Take home messages

• New-age data loggers with telemetry transmitted in-canopy environmental data to a server.
• Disease risk alerts based on the environmental data were emailed in near real-time from the server to researchers and agronomists.
• This risk alert system may ultimately remove the uncertainty around timing of and need for fungicide sprays to control foliar disease.
• It is anticipated this system may reduce the number of sprays and hence reduce costs to growers without increasing the risk of disease.

Background

Chocolate spot disease of faba beans and grey mould of lentils, caused by Botrytis fabae and B. cinerea, respectively, can be difficult to control and growers often use multiple prophylactic fungicide sprays in crops. The area sown to pulse crops is expanding into new districts, both higher and lower rainfall than the traditional growing areas. Appropriate disease management strategies need to be identified for these regions to ensure the successful uptake of these crops.

Current disease management strategies based on prophylactic sprays have been devised in medium to high rainfall zones with a relatively high intensity of individual pulse crop types i.e. with the assumption of high risk of disease. The new higher rainfall zones generally have longer and colder growing seasons than traditional regions, which impacts on disease severity and the number of foliar sprays required to control disease. Following the standard practices has led to a high input cost, with multiple sprays applied during the longer growing season. Conversely, lower rainfall areas may have a lower disease risk, thereby requiring fewer sprays, and lower yields which demand fewer costly inputs. This potentially renders the current strategy of applying fungicides at early flowering ineffective and may be an unnecessary expense.

Botrytis spp. need very humid conditions with temperature above 15°C continuously for several hours within the canopy to infect the faba bean or lentil plants (Davidson and Krysinska-Kaczmarek, 2007; Davidson, 2011). Hence, it seems possible to optimise fungicide application using field observations of the in-canopy environmental conditions via near real-time monitoring.

Method

Recent advances in Internet of Things (IoT) connectivity, referred to as Low Power Wide Area Networks (LPWANs), present opportunities to affordably acquire real-time field data. This can be used to help mitigate the risks associated with agricultural pests and pathogens. These networks are based on wireless technology that connects devices/sensors deployed in the field. The LPWAN technology utilised by this project is Narrow Band IoT (NB-IoT) and is currently operated in Australia by Telstra and Vodafone.

Compared to traditional mobile networks (3G/4G), NB-IoT operates at a lower cost, has greater power efficiency (devices with years of battery life), transfer small packets of data (for example, temperature, relative humidity (RH) and soil moisture for this project), and support more devices over a greater area (10km2).

We deployed data loggers (Figure 1) and passive spore traps at seven sites across the south east (SE), Yorke Peninsula (YP) and Mid-North (MN) of South Australia (SA). Slides from passive spore traps were collected fortnightly. These samples were quantified for conidia of pulse Botrytis spp. by DNA assays at the end of the growing season, to compare spore release with environmental data. Field cameras were installed at several of these locations to monitor symptom development in the crops. A foliar fungicide trial in faba beans was sown at one of the monitoring sites, Frances, in SA to investigate whether an alert triggered by the in-crop environmental data can determine the need for foliar fungicides.

The treatments were:

1. Nil
2. Tebuconazole at 145mL/ha (Orius^® 430 SC, 430g/L active ingredient (a.i.)) at grass spray, carbendazim 500mL/ha (Nufarm Spin Flo^® Systemic Fungicide, 500g/L a.i.) at canopy closure and carbendazim 500mL/ha at early podding spray if required
3. Tebuconazole at 145mL/ha at grass spray, carbendazim 500mL/ha at early flower and carbendazim 500mL/ha at early podding spray if required
4. Tebuconazole at 145mL/ha at grass spray, carbendazim 500mL/ha according to in-canopy conditions (RH and temperature)
5. Carbendazim 500mL/ha according to in-canopy conditions (RH and temperature).

*note: Tebuconazole can be used on faba beans for Cercospora Leaf Spot and Faba Bean Rust control under APVMA Permit 13752

Results and discussion

Although some remote monitoring sites in this study had none or a very poor mobile (3G/4G) signal reception, our deployed data loggers on NB-IoT successfully transmitted every six hours to the server. Data consists of hourly recordings of temperature and RH and six-hourly records of soil moisture. Hourly photographs were captured from cameras in the paddocks and transmitted to the server. The RH and temperature data transmitted by the loggers were used to set three levels of trigger warning alerts that were emailed by the server if in-canopy conditions were conducive for Botrytis infection. Yellow, amber or red alerts were emailed if RH was at or above 70%, and temperature was at or above 15°C for more than eight, 10 and 12 hours, respectively.

By the end of 2019 cropping season, nine yellow alerts and one amber alert were received, which finally progressed to a red alert from two data loggers in SE. Interestingly, although five yellow alerts for the fungicide trial site were received, those alerts never progressed to red. As a result, no foliar fungicides were sprayed in treatments (d) and (e) at Frances, according to in-canopy conditions. Only one yellow alert was received for 2019 season from data loggers deployed in YP and MN sites.

The trial site and commercial crops at monitoring sites were observed for disease development with minimal, or no symptoms of chocolate spot or botrytis grey mould seen in crops to the end of season. This suggests that in 2019 in SA, for all seven regions that data loggers were deployed, there was only one instance when fungicides may have been required to control chocolate spot in faba beans, or grey mould in lentils. This system could potentially save growers substantial amounts of money and time and fewer sprays reduce the risk of fungicide resistance developing in pulse crops.

Figure 2. A yellow trigger warning alert received via email from the server, based on in-crop, real-time environmental data.

Conclusion

By using NB-IoT data transmission faba bean and lentil canopies were remotely monitored at several sites in SA. Ultimately, this monitoring will provide a better understanding of microclimate elements, and their effects on Botrytis disease initiation and progression during the season. Based on trigger warning alerts received from telemetry data loggers compared to multiple prophylactic fungicide applications, this technology could avoid unnecessary fungicide application.

Using NB-IoT data telemetry to transmit near real-time environmental data, incorporated with our knowledge of Botrytis biology, can lead to precise timing of fungicide application for better control of Botrytis diseases in pulse crops. In the 2019 season, the trigger warning system for foliar fungicide spray did not send any red alerts for the fungicide trial site at Frances, SA. At this site, the level of the Botrytis disease was none or negligible in those treatments without any foliar fungicide spray during the season.

Despite the 2019 cropping season in many regions in SA being relatively drier than an average year, anecdotal data shows that growers sprayed their paddocks a few times for faba bean chocolate spot or lentil grey mould regardless of the weather conditions in 2019 season. This highlights that the trigger warning system can be a powerful tool to use even in a dry year for growers. Understandably this technology needs to be examined during a number of average, or wet growing seasons to validate its applicability; especially after being examined in a moderately dry year. The results will optimistically help growers to reduce the number of fungicide sprays and make these crops more profitable. This technology has the potential to be used for real-time monitoring of in-crop environmental data in many crops and areas and to be utilised for management of several diseases.

Acknowledgements

The research undertaken as part of this project is made possible by the significant contributions of growers through both trial cooperation and the support of the GRDC, the authors would like to thank them for their continued support.

Thanks to Jamus Stonor, Marzena Krysinska-Kaczmarek and Michelle Russ (SARDI) for their invaluable technical help. We thank Andrew Baker and his team from DATA EFFECTS for providing data loggers and technical support during the 2019 cropping season.

Useful resources

Spot the Difference- Identifying faba bean diseases - GRDC Communities

Crop Protection Products- Pulse Australia

References

Davidson, J. and Krysinska-Kaczmarek, M. (2007) Effects of inoculum concentration, temperature, plant age and interrupted wetness on infection of lentil (Lens culinaris) by Botrytis spp. conidia. Australasian Plant Pathology 36, 389-396.

Davidson, J. (2011) Botrytis gray mold of lentil. In ‘Compendium of chickpea and lentil diseases and pests’. (Chen, W., Sharma, H. C. and Muehlbauer, F. J.) pp. 45-48. (The American Phytopathological Society: St. Paul, MN, USA)

Contact details

Dr Mohsen Khani
SARDI, Plant Research Centre, Urrbrae, SA
08 8429 2285
Mohsen.khani@sa.gov.au