Report from the Sub-committee on Agriculture

10. Partial phasing out of pesticides in agriculture

10.1 Description of intermediate scenarios
10.2 0+scenario (almost total phase-out)
10.3 +scenario (limited use)
10.4 ++scenario
10.5 Evaluation of intermediate scenarios that are economically optimised
10.6 Documentation from field trials and experience from farmers’ groups show that a 50% reduction is possible
10,7 Control of application in different scenarios
10.8 Overall evaluation of the scenarios for total and partial phase-out
10.9 Conclusions concerning the intermediate scenarios

10.1 Description of intermediate scenarios

3 intermediate scenarios

The main committee decided that the following 3 scenarios were to be analysed:

	the 0+scenario (almost total phase-out): A scenario that permits sufficient use of pesticides to enable compliance with the current phytosanitary rules (the Danish Plant Directorate&;s regulations). The scenario is very similar to the 0-scenario.
	the +scenario (limited use): A scenario that describes areas in which it is known from experience that heavy losses occur if pesticides are not used. Basically, the areas included are those in which yield losses have, on average, exceeded 15%. The crop rotations used are similar to those practised in the described 0-scenario.
	the ++scenario (optimised use): In this scenario, the aim is to avoid all serious economic losses from pests. The production is deemed to be close to present production. The scenario assumes use of all available damage thresholds and harrowing, where these methods can compete with the chemical methods.

Agronomic scenarios

The content of the different scenarios is mainly described by means of tables. In the + and ++scenarios, the losses are expressed as crop-percentage losses at national level. The losses are then evaluated agronomically and economically and expressed in relation to the contribution margins of the crops and the types of farm. A total treatment frequency index is calculated for the different scenarios.

Economically optimised scenarios

Besides purely agronomically defined scenarios, the Danish Institute of Agricultural and Fisheries Economics has set up economically optimised intermediate scenarios for the 0-scenario with a view to showing the consequences for the economy of farms and the socioeconomic consequences. The economically optimised scenarios include a large number of agronomic restrictions as substitution for pesticides.

In the following, a description is given of both the purely agronomic scenarios and the scenarios set up on the basis of both agronomic and economic principles. The intermediate scenarios also form the basis for the Sub-committee on Environment and Health’s evaluation of the environmental consequences of changing from the present production to a total or partial phase-out of pesticides.

A list has been prepared of the pesticides that would probably be used in the 0+ and +scenarios. The list for the ++scenario is expected to be identical with the agents approved at present.

10.2 0+scenario (almost total phase-out)

Complicance with plant health rules

The crops that would be unable to meet specific, statutory requirements concerning purity or that are subject to requirements concerning prevention and control of pests covered by the quarantine rules are defined in notices from the Danish Plant Directorate. The list in table 10.1, in which the specific areas covered by the 0+scenario are described, has been checked with the Danish Plant Directorate.

Table 10.1  Look here
Situations in which dispensation would be granted for continued use of pesticides in a 0-scenario

The scenario is generally reckoned to be very close to a 0-scenario. In relation to 0-pesticide production, there would only be consequences for cultivation and yields in the case of seed grass, seed potatoes and pot plants/nursery cultures. Because of the dispensation granted for these crops, it is estimated that the productions in question could be maintained and would meet the requirements concerning purity and quality.

The permitted treatment of cereals with dressing agents corresponds to that proposed in the 0-scenario and is therefore not expected to change the production (table 10.9). Similarly, it is not considered that the permission to control wild oat would have any serious effect on the overall production, except that a serious problem with wild oat in a 0-scenario would require restructuring of the crop rotation and involve heavy costs for weed control if the requirements concerning control were to be met. These changes could be avoided in a 0+scenario.

Very limited use

The treatment frequency index would be very low in a 0+scenario. For most of the types of farm, it would be almost 0, while for potato and seed producers it would be less than 5% of the present level. As the scenario lies very close to the 0-scenario, its economic consequences have not been analysed.

10.3 +scenario (limited use)

Limiting large yield loss

In this intermediate scenario, slightly more use of pesticides is permitted than in the above-mentioned 0+scenario. The sub-committee has evaluated combinations of crops and pests to find the combinations for which it would be most difficult to avoid control with pesticides. The areas included (table 10.2) depend on the yield losses that can be expected as a consequence of attacks by pests. Basically, the sub-committee has included areas where there would be large yield losses or where it is estimated that a profitable production of specific crops could not be maintained. There would have to be a) substantial average losses from pests (>15-20%) or b) the production would be discontinued or would be impossible to fit into the crop rotation.

The basic assumption is a crop rotation in the +scenario that is similar to the 0-scenario, with not more than 40% winter cereals and with mechanical weed control and resistant varieties as important elements. Considerable monitoring would be needed to determine when there was an attack that would result in large yield losses. Where possible, account has been taken of the fact that these serious losses can occur in some years, but the scenario does not generally allow for the fact that losses greater than 15-20% can occur in a crop in individual localities and in individual years. That is because, for most crops, it is not possible to predict how often such a situation will occur.

Application areas in the +scenario

The areas included in the +scenario are described in table 10.2. The scenario also includes the areas already described in the 0+scenario. Appendix 2 shows in greater detail the losses that can be expected if pesticides may not be used and provides a little background for the individual problem areas.

Table 10.2
Crop/pest combinations for which pesticides are included to avoid very serious losses, and which are included in the +scenario

TFI of 0,5 in the +scenario

Table 10.3 shows the pesticides proposed for the various crops in the +scenario, measured in TFI, while table 10.4 shows how much the applications described would reduce the losses in relation to the 0-scenario.

The scope of the pesticide treatment in the +scenario gives a treatment frequency index of slightly less than 0.5, corresponding to an approx. 80% reduction compared with a level of 2.45 in 1997. About half the consumption in the +scenario would be in the form of herbicides. The treatment frequency index would vary between 0.2 for dairy farms on sandy soil and 1.1 for potato producers on sandy soil.

Change in CM II

Economic analyses of contribution margin II for the different types of farm in a +scenario show a total reduction of 0% for dairy farms on sandy soil, 15% and 36%, respectively, for arable farms on sandy and clayey soil, and 13% and 22%, respectively, for arable farms with seed production and sugar beet, while the reduction for potato producers would be 36% (see appendix 2 and table 10.9).

In the economically optimised +scenario, the treatment frequency indices are generally of the same order of magnitude as in the purely agronomic scenarios, but the reductions in contribution margin II are expected to be smaller. The reduction is 14-15% for dairy farms on sandy soil, 6% and 19%, respectively, for arable farms on sandy and clayey soil, and 15% and 23%, respectively, for arable farms with seed production and sugar beet, while the reduction for potato producers would be 15%.

Table 10.3  Look here!
Present pesticide consumption measured in treatment frequency indices. The consumption in the described +scenario is based on the acreages in the 0-scenario, while the ++scenario is based on the acreage figures from present production (1994 production). The figures do not include set-aside.

Table 10.4  Look here!
Percentage losses in the described scenarios – the 0-, + and ++scenarios. The total losses in the 0-scenario correspond to those shown in table 5.8. The losses in the + and ++scenarios are a best estimate in relation to the permitted use of pesticides. The economic analyses have been based on these losses.

Relevant pesticides in the 0+ and + scenarios

Table 10.5 shows a section of the pesticides that would be used in a 0+ and +scenario. The agents cover herbicides, fungicides and insecticides and represent broad section of the agents that are available today.

Table 10.5  Look here!
Chemicals that can be used for the applications mentioned in a 0+scenario.

10.4  ++scenario

Basically, serious economic losses from pests in agriculture are not expected in this scenario, see table 10.4. The production is largely the same as in the present production. The scenario is based on a proposal from DIAS, which, in 1996, estimated what could be a realistic reduction of the treatment frequency index (Anon., 1997) without affecting the present production economy.

Use of damage thresholds and mechanical weed control

In this scenario it is assumed that all available damage thresholds are used, together with mechanical weed control, where these methods can compete with chemical methods. A crop rotation very similar to that practised today can be expected, with economic optimisation but also optimisation with respect to minimising use of pesticides. More time will have to be spent on monitoring and need-based control than in present production – typically a half to a whole day per 100 ha per week in the growing season (14 days in all).

10.5  Evaluation of intermediate scenarios that are economically optimised

Cost of monitoring

Intermediate scenarios that are economically optimised are based on the options indicated in the 0+, + and ++scenarios. Table 10.4 shows the expected reduction in loss for each of these options. Costs for monitoring pests have been included (DKK 150/ha per year), together with costs for mechanical weed control and agronomic limitations on, for example, the proportion of winter cereal, crop sequences, etc. (Ørum, 1998).

Special crops and set-aside

As a natural reaction to the options in the 0+ and the +scenarios, special crops would be included in the crop rotations if the way were opened for use of pesticides for controlling the most problematical pests in those crops. Seed grass would return fully in the 0+scenario, while sugar beet would feature in the +scenario. In the case of potatoes, only ware potatoes would be included in the +scenario, whereas starch potatoes could only compete in the ++scenario. The winter-cereal acreage would increase considerable in many crop rotations in the +scenario, while the set-aside acreage would fall compared with the 0-scenario. However, in the case of arable farms on sandy soil, there would still be almost 30% set-aside and still no rape or peas in the crop rotation. Except in the case of potato producers, there would be no difference in the +scenario between the treatment frequency index at production units operated on the basis of agronomic considerations and those that were economically optimised (table 10.8).

TFI in intermediate scenarios

In the economically optimised ++scenario, the distribution of the acreage lies very close to the distribution in present production. There would, however, be a small reduction in the proportion of winter cereals, which would be made up for by more spring cereal. This would be a natural consequence in order to limit the growing problems with grass weed. The acreages with set-aside and special crops would be similar to present production in the optimised ++scenario. Exceptions to this are arable farms on sandy soil and potato producers on sandy soil, where there would still be a substantial proportion of set-aside in the ++scenario because peas and rape would not be able to compete in the price situation in question.

The TFI figures largely correspond to the types of farm described by the sub-committee (Mikkelsen et al., 1998). The figures have been arrived at by using the average consumption figures from the statistics in 1994, followed by a breakdown between the respective crops. Table 10.8 shows the treatment frequency index from two different starting points. In the +scenario, there is a reduction in TFI of between 90 and 80%, while in the ++scenario, there is a reduction of 43-71%.

10.6  Documentation from field trials and experience from farmers’ groups show that a 50% reduction is possible

Trials carried out with PC Plant Protection over a number of years indicate that it would be possible to reduce the treatment frequency index by 30-50% compared with present production (Secher, 1997).

Trials with af TFI of 1,3

Moreover, trials carried out with a low input of pesticides in Køge-Ringsted Farmers’ Association have shown that it would be possible for farms growing mainly cereals to approach a treatment frequency index of 1.3 without any reduction in contribution margin (table 10.6).

However, these low treatment frequency indices depend on intensive monitoring during the entire growing season, and special crops, such as sugar beet, seed grass and potatoes, were not included in the crop rotations (Anon., 1998).

A crucial question when judging these results is whether the results from small plot trials can be transferred to similar success at field level.

Table 10.6
Results of trials with different treatment frequency indices in cereals, from which it appears that even though the grain yield falls with falling pesticide input, the contribution margin (CMI)remains much the same, indicating a considerable potential for reducing pesticide input provided the crop is closely monitored. The reduction in TFI is not due to spraying less but to reduced dosages. Correspondingly, the nitrogen supply is typically reduced by 10 kg N/ha (Køge-Ringsted Farmers’ Association, 1998).

Experience from farmer’s groups

In connection with the first pesticide action plan, many farmers’ groups were established to exchange experience on plant protection. The groups consist of 8-10 farmers and an agricultural advisers. At the farms represented in these farmers’ groups, the consumption of herbicides and fungicides is lower than the national average, whereas the consumption of insecticides is slightly higher. Participation in the groups has resulted in an overall fall in pesticide consumption. The groups have had a major influence on the choice of pesticides and the dosages (Danish Environmental Protection Agency Report No. 296, 1995). In 1996-1998, the treatment frequency indices used for crops by the members of the farmers’ groups in Ringkøbing and Videbæk Farmers’ Associations were calculated. The results show that indices in line with the action plan’s goals can perfectly well be achieved (table 10.7). (Ringkøbing County’s Arable Farming Report 1998). Generally, however, cereals suffer fewer attacks from diseases in West Jutland than, say, on the Danish islands.

Table 10.7
Statistics showing the treatment frequency index (TFI) used in practice by members of farmers’ groups in Ringkøbing and Videbæk Farmers’ Associations in the period 1996-98

10.7  Control of application in different scenarios

Use of damage thressholds

It lies outside the sub-committee’s mandate to indicate non-technical instruments for controlling pesticide consumption in the three scenarios. Both the +scenario and the ++scenario incorporate the use of damage thresholds and data from the registration system. Compared with present-day practice, all spraying would be carefully considered before being carried out. In the +scenario, for example, there would have to be a well-founded suspicion of an attack of great economic importance. In the ++scenario, farmers would use damage thresholds together with a combination of alternative and chemical methods of controlling weeds etc.

No damage thresholds for some pests

We do not at present have reliable damage thresholds for all the main pests. The development of the pests depends greatly on the weather, and this, together with the fact that treatment has to take place relatively early in the life-cycle of the pests in order to achieve the optimum effect with reduced dosages, means that it is not possible to predict reliably a fixed percentage loss for a given treatment. The needs and percentage losses indicated have therefore, in several cases, been based on experience from control trials in a given growing season. Within certain crops, this makes it difficult to operationalise scenarios that are based on a partial phase-out of pesticides.

In the 0+scenario, it is assumed that dispensation would be granted for the use of pesticides in accordance with specific rules. The treatment frequency index would thus be very limited in this scenario.

10.8   Overall evaluation of the scenarios for total and partial phase-out

Key figures for the intermediate scenarios

As a basis for evaluating the general consequences of the different scenarios for production, table 10.8 shows the respective treatment frequency indices in the intermediate scenarios in relation to present production. Table 10.9 shows key figures for changes in contribution margin II for 10 different types of farm, while table 10.10 gives the total production figures.

Contribution margin II

Contribution margin II is a good measure of the consequences of the different intermediate scenarios for the different types of farm. The contribution margin expresses the economy/ha since this quantity adjusts for loss of yield and extra yields, changes in the cost of purchasing and applying pesticides, together with changes in the cost of mechanical weed control. The value of the costs saved on application of pesticides and the increased costs for mechanical weed control have been determined by using machine station rates. It will be seen that reductions in CMII of 4-93% have been measured for the 0-scenario, 0-36% for the +scenario and 0-17% for the ++scenario. Looking at the individual types of farm, the consequences would thus be least for dairy farms and greatest for farmers with a substantial production of special crops. This accords with the economic analyses at farm level carried out by the Sub-committee on Production, Economics and Employment (Ørum, 1998). The losses at dairy farms are relatively high in the ++scenario because mangolds and maize have been retained in this scenario, but not in the 0 and +scenarios, which gives a better contribution margin.

Table 10.8  Look here!
Treatment frequency index in 3 scenarios shown for 10 different types of farm on clayey and sandy soil.

Where economic optimisation is used, the reduction in CMII for the 0-scenario is generally smaller and more evenly distributed than in the agronomic scenarios. The model-optimised CMII for present production is improved by DKK 50-400/ha compared with the actual present production, which indicates that there may be a potential for improving the economy of present-day farms.

The reason for the differences between the agronomically and economically optimised farms is that the agronomic 0- and +scenarios include a crop rotation that is based on the described 0-scenario, in which production of special crops is maintained, whereas a distribution based on the present production is assumed in the ++scenario and present production. The economically optimised crop rotations express the optimised land consumption with production with or without permission for limited use of pesticides and with the yield losses and crop-rotation restrictions set by the Sub-committee on Agriculture. In the 0-scenario production of special crops is reduced almost to zero and the set-aside acreage is increased considerably in several of the optimised scenarios, although with a maximum upper limit of 30%.

Table 10.9  Look here!
Percentage change in contribution margin II for 10 different types of farm with and without economic optimisation in the case of total and partial phasing out of pesticides

Change in production

Total cereal production would fall by around 30% in both the agronomic and the optimised 0-scenario, which would mean importing grain in order to maintain the present pig production (table 10.10). Production of potatoes and seed would be approximately halved in the agronomic scenario, while there would be an approximately 30% increase of both rape and peas. This rise would reduce the need for bought-in supplementary fodder. In the economically optimised scenario, this production would be largely replaced by set-aside, and both potato production and beet production would be reduced by more than 90% and seed production by 60%.

Unchanged livestock production

In none of the described scenarios is it thought that production would be reduced to an extent that would affect livestock production This accords with the economic analyses at farm level carried out by the Sub-committee on Production, Economics and Employment (Ørum, 1998). The level of coarse fodder per livestock unit (LU) would be similarly maintained without pesticides (total coarse-fodder production would remain constant). The proportion of cereals would thus be reduced by the proportion needed to substitute the reduction in yield in wholecrop and grass. Farms with livestock would thus need to purchase more primary cereal because they would no longer be able to grow enough themselves to maintain their fodder consumption. In the optimised scenarios with a large proportion of set-aside, the sub-committee has not considered the question of whether, with the reduced cereal yields and increased set-aside acreage, it would be possible to sell liquid manure to neighbouring farms. Nor has the sub-committee considered whether there would be a shortfall in straw production in relation to the current requirements.

Need for grain imports

Net grain exports averaged 17,222 hkg in the years 1993-96 (Agriculture in Denmark 1998). Whereas, in the 0-scenario, a net import of just under 10,000 hkg would be needed (83,986-17,222 – 58,398), in the +scenario, the figure would fall to just under 3,000 hkg (83,986-17,222-63,858) to maintain livestock production at the present level.

In the 0+scenario, there would be changes only in the production of seed potatoes and grass seed, whereas, in the +scenario, there would be an increase in the production of almost all crops in relation to the 0-scenario. Specific production figures are not given for the ++scenario because it is estimated that, in this scenario, there would be only a small change in the yield level in relation to present production, according to the definition of this scenario.

The socioeconomic consequences of a restructuring are included in the analyses from the Sub-committee on Production, Economics and Employment.

Table 10.10  Look here!
The main productions in 1000 hkg (CU) for present production in whole DKK. For the scenarios, the percentage change in production is given. The figures are based on figures from the economic analyses at farm level.

10.9  Conclusions concerning the intermediate scenarios

The sub-committee has considered three intermediate scenarios: a 0+scenario, a +scenario and a ++scenario.

Treated acreage in the 0+scenario

The 0+scenario (almost total phase-out) covers a scenario in which the only aim is to comply with the current phytosanitary legislation and regulations. As in the 0-scenario, this includes treatment of cereals with dressing agents in all early generations (up to and including C1) and, where a need assessment shows that treatment is necessary in the C2 generation (10,000-20,000 ha), field-spraying of about 70,000 ha with seed and seed potatoes, and of land infested with wild oat. Spraying of greenhouse and nursery cultures with pesticides is also permitted in order to comply with rules for export and home sale. In a 0-scenario, the treatment frequency index would be very low – almost zero for most types of farm and less than 5% of the present level in potato and seed production. With respect to total production in Denmark, there would be an increase only in production of grass seed and seed potatoes compared with the 0-scenario.

The +scenario (limited use) covers a scenario in which continued use of pesticides is permitted for control of pests of very great economic importance – treatment, for example, to ensure continuation of a profitable production of special crops. Altogether, the treatment frequency index would be around 0.5 in this scenario, which is a reduction of about 80% compared with the present consumption. The index ranges from 0.2 for dairy farms on sandy soil to 1.1 for potato producers on sandy soil.

CM II in agronomic +scenario

Economic analyses of contribution margin II for the different types of farm in a +scenario show a total reduction of 0% for dairy farms on sandy soil, 15% and 36%, respectively, for arable farms on sandy and clayey soil, and 13% and 22%, respectively, for arable farms with seed production and sugar beet, while the loss for potato producers would be 36%.

This reduction is based on an assumption of largely the same restructuring of production as described in the 0-scenario. The chosen input is deemed sufficient to retain the present production of sugar beet, seed-producing crops and potatoes. The scenario permits the use of pesticides where pests result in an average yield loss of more than 15%. The scenario thus does not calculate the consequences in individual localities and at individual farms in some years because, for most crops, it is not possible to predict how often such a situation will arise. The scenario also permits the use of pesticides in outdoor vegetables, fruit and berries and ornamental greenery to the extent needed to maintain production (approx. 20,000 ha with market-garden crops and 35,000 ha with ornamental greenery).

CM II in economically optimised +sceanario

In the economically optimised +scenario, the treatment frequency indices would generally be of the same order of magnitude as in the agronomic scenarios, except in the case of potato producers. The reduction in contribution margin is 14-15% for dairy farms on sandy soil, 8% and 19%, respectively, for arable farms on sandy and clayey soil, and 15% and 23%, respectively, for arable farms with seed production and sugar beet, while the reduction for potato producers is 15%.

In the ++scenario (optimised use), pesticides may still be used to the extent needed to avoid economic losses. The scenario assumes the use of all available damage thresholds, together with harrowing and other mechanical weed-control methods where these can compete with the chemical methods. Crop rotations are expected to be as at the present time, with economic optimisation, but also optimisation with respect to using as little pesticide as possible. Compared with present production, more man-hours would have to be spent on monitoring pests and using damage threshold programmes.

Treatment frequency of 1,5-1,7 in the ++scenario

The total treatment frequency index (TFI) in the pure agronomic scenario would be approx. 1.7 if set-aside were omitted. This corresponds to a 31% reduction compared with the treatment frequency index in 1997 and 36% compared with the treatment frequency index in the reference period 1981-85. This covers a range from 0.7 for dairy farms on sandy soil to 2.6 for potato producers on sandy soil. In the corresponding economically optimised scenario, TFI ranges from 0.2 at intensive dairy farms on sandy soil to 2.6 at farms with a large production of potatoes. The average TFI for the optimised ++scenario is 1.45. Contribution margin II for all farms would not differ significantly from present production, although in some of the optimised crop rotations there are indications that it would be possible to improve on the present contribution margins for certain types of farm. The average reduction in contribution margin II would be 2%.

Need for better warning systems

In all, the intermediate scenarios would reduce considerably the losses expected in the 0-scenario. In the +scenario, the yield losses would typically be reduced by 25-50%, while there would be practically no losses at all in the ++scenario. In the case of diseases and pests, damage thresholds and warning systems would have to be used in order to reduce the percentage losses. However, we do not have reliable damage thresholds for all areas at present and, for reliability, long-term weather forecasts would be needed. Therefore, there is considerable uncertainty, particularly with respect to the amount of spraying needed to ensure against yield losses of more than 15%. Compared with present production, many more man-hours would have to be spent on monitoring pests, and to avoid attacks of diseases, a lot of breeding work would be needed in step with the breakdown of "resistance".

Experience from trials with intermediate scenarios

Experience from trials and farmers’ groups with intensive input from agricultural advisers indicates that a treatment frequency index of around 1.3 for vegetable production is realistic. Similarly, experience with a TFI of 0.5, corresponding to the +scenario, is not available.

References

Anon. (1997): SP Report No. 11, 1997

Anon. (1995): Environmental Report from the Danish Environmental Protection Agency, 296. Farmers’ Groups’ project

Anon. (1998): Arable Farming Report from Køge-Ringsted Farmers’ Association

Anon. (1998): Arable Farming Report from Ringkøbing County Farmers’ Association

Secher, B.J.M. (1997): Effect of research and advice on development of consumption and pattern of consumption. SP Report No. 11, 1997

Ørum, J.E. (1998): Consequences for production economy of phasing out pesticides. Report prepared for the Pesticide Committee, 1998