Summary and conclusions

Beslutningsstøttesystem til reduktion af pesticidforbrug

After having experienced severe attacks of the aphid transmitted disease “barley yellow dwarf virus” (BYDV) in 2007, a desire to develop a decision support system to assist the farmer in deciding whether to spray for aphids or not in the autumn, has emerged. In the autumn of 2007, an early warning system was therefore established in connection with this project. In this system, the agricultural advisors assessed the scale of attack of aphids in 40 winter wheat fields and 20 winter barley fields. Because a relatively high amount of aphids were found, spraying was recommended in early sown fields, located in high risk areas.

By a visual assessment of BYDV in the unsprayed winter wheat and winter barley fields from the early warning system, in the spring of 2008, it was confirmed, that spraying in many cases were necessary, based on an evaluated threshold of 1 % attacked plants in winter barley, and 2 % in winter wheat. The low thresholds imply that it takes relatively small attacks, before spraying against aphids is cost-effective.

Climate data were used to show aphid day degrees, based on an English model. This model assumes that the second generation aphids occur at 340 day degrees (with a basis of 3°C). In England, spraying is recommended at this stage. The number of day degrees were summarized based on the time of germination in the individual field. Since the model is not tested under Danish conditions, the development in aphid day degrees in the autumn of 2007 was shown instead, relative to “normal autumns”, that means the autumns 1961-90 + 0.7 °C (0.7 °C is the temperature increase from 1961-90 to 1995-2004). Based on the time of germination, the development in day degrees was shown.

According to DMI, the autumn of 2007 was not significantly different from the average with respect to temperature. In September the temperature was normal, and there was more rain and sun than normal. In October the temperature was lower than normal, but dry and sunny. November was a little bit warmer than normal. The winter was however significantly warmer than normal, with an average temperature of 4.1 °C, which is 3.6 °C higher than normal for the period 1961-1990. Because the temperatures in the beginning of the autumn (September-October) were fairly normal, the activity of the aphids was low in this period.

The threshold for BYDV was exceeded in 24 % of the winter wheat fields and 41 % of the winter barley fields located in the high risk areas. The early warning made it possible to target the spraying against aphids in the autumn to fields, where the risk of attack was high. A focused warning has therefore helped to ensure, that the consumption of plant protection products has been lower than if all fields systematically was sprayed as a precaution, due to lack of knowledge.

In order to identify the problems and challenges facing the farmers in the process of decision making, an interview study was performed in May 2008. The farmers included in the study either lived in an area with a low potential risk of BYDV or in an area with high risk. The study shows that farmers in general are good at monitoring for aphids in their fields, but also that they use other relevant information as a support, before they decide to spray. Other relevant information include dialog with the advisors, newsletters, farm press, etc. This information is to a large extent dependent on results from the early warning system.

To sum up, this project shows that early warning is a significant part of the decision making process of the farmer. Often the decision to spray is based on information from several sources, own experiences and registrations in the field as well as decision support models such as Plant Protection Online (PVO), required that the system contains models for the pathogen in focus. PVO does, at the moment, not include any model capable of predicting the need for protection against aphids in the autumn in winter cereals.

A focused warning therefore helps to ensure, that the consumption of plant protection products is been lower than if all fields systematically is sprayed as a precaution, due to lack of knowledge.

Another objective of this project has also been to investigate whether SMS is an effective method to target the broadcasting of the results of the early warning in time and space, so that the farmer only receives the information valid for him in relation to e.g. crop and timing. The investigation confirms that there is a potential for early warning via SMS, but also that the potential is not yet fully exploited.

The early warning for control of pathogens can in principle be issued from anyone and anywhere, including the companies selling plant protection products and research institutions. As it will always be necessary with a certain degree of expert judgment, it is crucial that the persons making these judgments has no economical or career-wise interest in the outcome of the early warning, and that they are closely connected to practical agriculture, realistic experiments and know the decision-making behavior of the farmers.