Research and Development s12

DEPARTMENT for ENVIRONMENT, FOOD and RURAL AFFAIRS CSG 15 Research and Development Final Project Report (Not to be used for LINK projects)

Two hard copies of this form should be returned to: Research Policy and International Division, Final Reports Unit DEFRA, Area 301 Cromwell House, Dean Stanley Street, London, SW1P 3JH. An electronic version should be e-mailed to [email protected]

Project title Oilseeds Diseases and Winter Pests databases

DEFRA project code AR0509

Contractor organisation Central Science Laboratory and location Sand Hutton York

Total DEFRA project costs £ 162,909

Project start date 01/04/02 Project end date 31/03/03

Executive summary (maximum 2 sides A4)

The aim of this project was to gather data on the incidence and severity of oilseed rape pests and diseases in England and Wales, to collect associated data on agronomic practice, pesticide use, and to gather information on site and environmental factors. This survey has generated data covering a one year period from autumn 2001 to summer 2002. In conjunction with previous year’s data, these have been used to identify the importance of individual oilseed rape pests and diseases in England and Wales and to investigate the influence of a wide range of factors such as cultivar, planting date, geographical location and pesticide use on diseases and pests. Full reports of these analyses can be found in the Annual Oilseed Rape Pest and Disease Survey Reports (available from CSL).

The Defra-funded Winter Oilseed Rape Survey has been carried out annually since 1986. Data from the project has highlighted problem areas for pest and disease control and pesticide use. Major changes in pesticide control practices have been identified and attributed to fluctuating pest and disease risk, improved fungicide chemistry, economic pressures and advances in the understanding of epidemiology and control. The survey data have also provided the means for developing, validating and continuously updating forecasting systems, e.g. Decision Support System for Oilseed Rape Insect Pests (DORIS), the light leaf spot forecast, and the current PASSWORD project which aims to integrate the light leaf spot and phoma disease forecasts with the DORIS system into one Decision Support System.

Ninety-seven crops were selected for pest and disease monitoring from farms throughout England & Wales, the number of farms selected being in proportion to the area of rape grown in each county. In addition, 10 untreated crops were also monitored to assess disease levels in the absence of fungicide treatment. The crops were sampled on three occasions; in the autumn at mid-leaf production (early December), in the following spring at early stem extension (March) and again in the summer at pod ripening (July). On each occasion, 25 plants were sampled randomly and the severity of diseases assessed. The autumn and spring samples were also assessed for aphids, rape winter stem weevil and slug damage. Additionally, the autumn

CSG 15 (Rev. 6/02) 1 Project Oilseeds Diseases and Winter Pests databases DEFRA AR0509 title project code samples were assessed for cabbage stem flea beetle. A sub-sample of 10 plants was taken at random from the autumn and spring samples and assayed for the presence of beet western yellows virus, cauliflower mosaic virus and turnip mosaic virus using an enzyme-linked immunosorbent assay technique. Agronomic details including sowing date, cultivar and pesticide use were also recorded for each crop surveyed. The data collected were entered onto an INFORMIX relational database at the CSL for subsequent analysis. Data have supplemented the existing DEFRA Winter Oilseed Rape Pest and Disease Survey database to generate a dataset covering the period 1986 to 2002. Summary information, interactive searches, risk maps and annual reports from this project can be accessed via the web at http://www.crop-disease-surveys.com.

Disease levels have fluctuated considerably during the years 1999 to 2002, with the incidence of some diseases reaching unprecedented levels during this period. Levels of all major stem diseases in 2002 were above the ten-year means, as were levels of light leaf spot and dark pod spot on the pods. High levels of phoma canker and light leaf spot continue the increasing trend in incidence of these diseases over the last four or five years. Phoma canker is now the major threat to the industry and losses in 2001 were estimated to have cost £36 million (Turner et al., 2001). Losses for light leaf spot were £18M whilst dark pod spot and sclerotinia each cost less than £1M in lost yield. Yield losses in 2002 were higher than in 2001 mainly due to an increase in light leaf spot. The need for more reliable forecasts to aid the targeting of sprays for both pests and diseases is clearly highlighted; a requirement which is currently being addressed in the PASSWORD project in the UK which aims to develop a decision support system for pests and diseases of oilseed rape. It is clear from analyses of seasonal and regional fluctuation in disease levels over time that the weather is a key driver in these disease epidemics. Adverse weather has also prevented fungicide applications at key stages during the autumn and winter and therefore influenced disease control. Data from the UK meteorological office show that the average rainfall in 1997 was equal to the 30-year mean, and that rainfall levels in every year since have greatly exceeded this figure. Similarly, average temperatures for central England have exceeded the 30-year mean in nine out of the last ten years. Whilst these weather conditions may be a transient feature, the possibility of climate change must be considered. Data show that if these conditions persist there will be a considerable impact on disease activity on an annual basis.

The current levels of disease activity indicate that the potential to control these diseases using fungicides is still not being fully exploited. This is despite a record level of fungicide applications in 2002. Use of fungicides has continued to increase throughout the last ten years with over 90% of crops treated annually between 1998 and 2001. The total cost of applications peaked in 1998 at £27.50 per hectare and have since declined slowly to a figure of £23.60 per ha in 2000 rising again to £26.00 per ha in 2001. The frequency of use of sprays, the choice of active substances and the timing of the sprays all indicated that growers were aware of the threat posed by diseases and were taking steps to control them. Advances in the understanding of disease epidemiology and control have brought about improvements in the timing of fungicide sprays to control particular pathogens. Data from the surveys support the importance of early autumn sprays for control of light leaf spot and spring sprays for control of phoma canker. Data suggest that fungicide use reduced disease incidence but that the improvement in level of control by changes in timing or number of sprays was, in many cases, minimal. Poor fungicide timing and limited information on variation in epidemic development are contributing factors. Improved chemistry and the availability of new information on timing have not significantly decreased overall fungicide use, nor does it appear to have made a major improvement to disease control on a national scale. There is clearly a requirement for more accurate assessment of disease risk in order to determine the need for and timing of fungicide applications in order to improve gross margins, particularly given the increasing economic pressures on farming. Use of insecticides has also remained high during the survey period with 78% of crops treated in 2002. This is despite levels of pest infestation being well below threshold in the vast majority of fields monitored.

The benefits of cultivar resistance for disease control are also not being exploited to the full. The most severe and widespread attacks of light leaf spot occurred in 1994 and 1995, and during this epidemic 56% of cultivars grown had a NIAB resistance rating <=5 for light leaf spot, indicating that they were susceptible cultivars. Given the current threat from diseases and the need to minimise inputs, growers need to limit potential disease problems by utilising cultivars that have a higher resistance rating to stem canker and light leaf spot. However, despite this, data from 2002 show continued growing of susceptible cultivars. Although overall levels of susceptibility were lower in 2002, there are still significant implications for disease risk in the future. Given the current threat from diseases and the need to minimise inputs, growers need to reduce potential disease problems by utilising cultivars that have a higher resistance rating to stem canker and light leaf spot. Control strategies need to be reassessed to encompass a more targeted and integrated approach using

CSG 15 (Rev. 6/02) 2 Project Oilseeds Diseases and Winter Pests databases DEFRA AR0509 title project code cultural practice and forecasting alongside chemical control. This would result in additional economic and environmental benefits and aid the move towards sustainable crop production.

Overall the disease survey has provided valuable data on the status of individual pests and diseases in 2002, which can be put into context by comparisons with disease levels over a longer period of time. The influence of cultural practice and economic change can be clearly identified and there are indications of major risks for the future. The surveys provide the only independent source of information on actual farm practice and are being utilised to provide new understanding of disease epidemiology. There are benefits to Defra policy from identifying priorities for achieving sustainable arable production and environmental benefits. The continuing use of unnecessary pesticide applications in many crops only serve to highlight the need for more reliable forecasts and risk assessments in order to target treatments more effectively while minimising yield loss.

CSG 15 (Rev. 6/02) 3 Project Oilseeds Diseases and Winter Pests databases DEFRA AR0509 title project code

Scientific report (maximum 20 sides A4) Background

Generation of statistically valid datasets on the status and importance of diseases and influences of current management practices can make a very significant contribution to the movement towards more sustainable crop production. The changing status and importance of oilseed rape pests and diseases has major impacts not only on the UK oilseed rape industry, but also on consumers, particularly with concerns over pesticide residues and impacts on the environment. Data can assist in the development of strategies to improve seed health, pesticide minimisation and regulation, improvement of quality and the overall competitiveness of the UK oilseed rape industry. Oilseed rape is the second most important combinable crop in England & Wales in terms of area grown and pesticide inputs. The potential role of new technologies in maintaining the economic viability of oilseed rape and continued improvements in crop production and strategies such as pesticide optimisation, ICM and use of forecasting systems will need careful monitoring and validation. The survey data provide a means of assessing the impact of these strategies and new technologies.

The DEFRA-funded Winter Oilseed Rape Survey has been carried out annually since 1986. Data from these surveys have been used to quantify changes in pests and diseases and the impacts of crop protection treatments and agronomic practices on their control. The long-term national survey of winter oilseed rape diseases has highlighted major changes in fungicide control practices as a result of fluctuating disease risk, improved fungicide chemistry, economic pressures and advances in the understanding of disease epidemiology and control. The survey data also provide the means for developing, validating and continuously updating forecasting systems, e.g. the Decision Support System for Oilseed Rape Insect Pests (DORIS), the light leaf spot forecast, and the current PASSWORD project which current PASSWORD project which aims to integrate the light leaf spot and phoma disease forecasts with the DORIS system into one Decision Support System

The main scientific objectives were:

1. Monitor the incidence and severity of oilseed rape diseases in 2001/2002 2. Monitor the incidence of oilseed rape pests and virus diseases in 2001/2002 3. To relate and cross-correlate the disease and pest incidence data with agronomic information 4. To collate, interpret and disseminate results

Results from previous surveys have been summarised by Hardwick & Turner (1994) and Turner et al. (2000).

Methodology

The crops were sampled on three occasions; in the autumn of 2001 at mid-leaf production (early December), in spring 2002 at early stem extension (March) and again in the summer at pod ripening (July) and growth stages were recorded using the scheme of Sylvester-Bradley (1985) and Sylvester-Bradley et al. (1984). On each occasion, 25 plants were sampled randomly and the severity of diseases assessed. The autumn and spring samples were also assessed for aphids, rape winter stem weevil and slug damage. Additionally, the autumn samples were assessed for cabbage stem flea beetle. A sub-sample of 10 plants was taken at random from the autumn and spring samples and assayed for the presence of beet western yellows virus, cauliflower mosaic virus and turnip mosaic virus using an enzyme-linked immunosorbent assay technique. Agronomic details such as sowing date, cultivar and pesticide use were also recorded for each crop surveyed.

CSG 15 (Rev. 6/02) 4 Fungal disease assessment. i. Autumn and spring

Leaf diseases were assessed on a whole plant basis. Each plant was assessed for dark leaf spot (Alternaria brassicae), downy mildew (Peronospora parasitica), light leaf spot (Pyrenopeziza brassicae), phoma leaf spot phase of canker (Leptosphaeria maculans), powdery mildew (Erysiphe cruciferarum) and white leaf spot (Mycosphaerella capsellae). The percentage leaf area affected was scored and recorded for each disease. These scores were then used to give a mean percentage leaf area affected for each individual disease. ii. Summer

Pod diseases, principally light leaf spot, white leaf spot, powdery mildew and dark pod spot were assessed on a whole plant basis. The percentage pod area affected was scored and recorded for each disease. These scores were then used to derive a mean percentage pod area affected for each individual disease.

Of the stem diseases, alternaria, light leaf spot, downy mildew and powdery mildew, were assessed as the percentage of stem area affected. Canker (Leptosphaeria maculans), sclerotinia stem rot (Sclerotinia sclerotiorum) and grey mould (Botrytis cinerea) were assessed on a whole plant basis using a 0-4 scale, where:-

0 - no disease 1 - less than half the stem circumference affected 2 - more than half the stem circumference affected 3 - the whole stem girdled by a lesion 4 - plant dead

Results were expressed as a mean disease index for each disease.

Root disease, notably clubroot (Plasmodiophora brassicae), was recorded as the percentage of plants affected.

Pest assessment

Pest assessments were carried out on samples collected in the autumn and spring. i. Aphids were assessed by counting the overall numbers of peach potato (Myzus persicae) or mealy cabbage aphids (Brevicoryne brassicae) and scored as none, few (up to one per plant) or many. ii. Cabbage stem flea beetle (Psylliodes chrysocephala) damage was assessed by removing each leaf close to the main stem and counting the number of visible scars on the mid-rib petiole. Each leaf was dissected and the total number of larvae seen was recorded (Assessed in the autumn only). iii. Rape winter stem weevil (Ceutorhynchus picitaris) were assessed by dissecting each plant stem and crown and recording the total number seen. iv. Slug damage severity was scored as none, little or much.

Comments on other pests such as leaf or blotch miners were included if damage was unusually severe.

Virus disease assessment

In addition to an assessment of plants for fungal diseases, a sub-sample of ten plants per crop was retained from the autumn and spring samples for virus tests. The sub-sample comprised of leaves from 10 randomly selected individual plants. Bulk samples of all leaves from each plant were tested to minimise the risk of missing the presence of virus. Tests were undertaken to identify the presence of beet western yellows virus (BWYV), cauliflower mosaic virus (CaMV) and turnip mosaic virus (TuMV) using an enzyme-linked immunosorbent assay technique (ELISA) (Hill, 1984). Gamma globulin and monoclonal antibodies used for identifying BMYV were supplied by CSL. Anti-rat conjugate was supplied by Sigma Chemical Co Ltd. All were used at dilutions of 1:1000. The gamma globulin and conjugate preparations used for identifying CaMV and

CSG 15 (Rev. 6/02) 5 Project DEFRA title project code

TuMV were supplied by Loewe Biochemical and used at dilutions of 1:200. The dilutions of sap were 1:10 for all three ELISA tests. After testing individual leaves, plants were potted up and grown on under glasshouse conditions so that further testing could be undertaken where positive reactions were obtained.

Data handling

A comprehensive questionnaire was completed for each crop detailing agronomic information such as cultivar, sowing date, previous cropping and pesticide use. All data were entered into an INFORMIX relational database at CSL for subsequent analysis.

The latest results from this project together with annual reports and risk maps can be found on the CSL web site at http://www.crop-disease-surveys.com. An interactive graphics system which allows direct interrogation of the database is also available on the site.

Results

This report covers the period from autumn 2001 to summer 2002, but the tables also contain data from the previous 3 years survey to set the current season’s information into context.

Statistical analysis

Actual means are reported in the tables in this report, although Analysis of Variance has been carried out on a transformed dataset (angular transformation). Values which share the same letter are not significantly different.

Disease levels on the leaves and pods

Incidence of dark leaf spot in the autumn and spring was higher than in previous years. Incidence on the pods was also significantly higher than in recent years and, with 89.7% crops and 40.9% plants affected, was the most common disease on the pods. The highest incidence was recorded in the South West where 81.3% of plants were affected. This is in contrast to 2000 and 2001 when the highest incidence was in the South East, but is consistent with the trends recorded in the high disease years of 1998 and 1999, when highest levels were seen in the South West and South East.

Although incidence was high, severity of dark leaf spot was low and was recorded at trace levels only. Incidence of dark pod spot was significantly higher than in 2001.

Table 1. Incidence and severity of dark leaf and pod spot

1999 2000 2001 2002 p value % plants affected Autumn (leaves) 2.89 bc 1.01 a 2.68 ab 5.53 c <0.01 Spring (leaves) 0.21 a 0.76 a 1.79 a 3.79 b <0.001 Summer (pods) 48.46 c 31.49 b 6.13 a 40.95 c <0.001

% area affected Autumn (leaves) tr tr tr tr NS Spring (leaves) tr tr tr tr NS Summer (pods) 0.69 bc 2.05 c 0.05 a 0.67 b <0.001 NS = Not significant tr < 0.05

Incidence of downy mildew on a national basis was slightly higher on the leaves in 2002 with 77.3% crops affected in the autumn and 82.5% in the spring compared to 60.6% crops affected in the autumn and 78.6% in the spring in 2001. However, incidence within crops was lower with 15.7% plants affected in autumn and

CSG 15 (1/00) 6 Project DEFRA title project code

29.0% in the spring of 2002 compared to 24.9% in the autumn and 33.6% in the spring of 2001. There was no consistent regional trend noted throughout the 2002 season, although at pod development highest pod incidence of 8.0% plants was recorded in the South West which was consistent with the trend recorded in 2001.

Severity of downy mildew was low, severity in the spring and summer being consistent with levels recorded in the previous two years.

Table 2. Incidence and severity of downy mildew

1999 2000 2001 2002 p value % plants affected Autumn (leaves) 39.36 b 20.08 a 24.85 a 15.71a <0.001 Spring (leaves) 40.68 31.12 33.59 29.03 NS Summer (pods) 5.87 a 16.38 b 3.36 a 2.89 a <0.001

% area affected Autumn (leaves) 0.81 c 0.14 a 0.41 b 0.12 a <0.001 Spring (leaves) 0.68 b 0.29 a 0.21 a 0.27 a <0.001 Summer (pods) 0.02 0.07 0.01 0.02 NS NS = Not significant

In 2002, light leaf spot was recorded at similar levels on the leaves to those recorded in 2001. However, by pod development, incidence was significantly higher in 2002 with 16.0% plants affected compared to 8.0% in 2001. The number of crops affected was similar to levels recorded in 2001. Throughout the season, the highest incidence was recorded in the North with the next highest incidence being recorded in the East. This was similar to the trend seen in previous years.

The severity of light leaf spot was generally low in 2002, and although slightly higher in spring and summer, these differences were not statistically significant from the levels recorded during the previous three seasons.

Table 3. Incidence and severity of light leaf spot

1999 2000 2001 2002 p value % plants affected Autumn (leaves) 0.26 0.42 0.64 0.17 NS Spring (leaves) 7.74 a 11.92 ab 18.00 b 18.06 b <0.01 Summer (pods) 7.53 a 12.30 ab 8.01 a 16.04 b <0.01

% area affected Autumn (leaves) tr tr tr tr NS Spring (leaves) 0.26 0.51 0.55 0.96 NS Summer (pods) 0.71 0.70 0.22 0.64 NS NS = Not significant tr < 0.05

As in previous years, phoma was the most common foliar disease in the autumn and spring affecting 94.8% crops and 40.9% plants in the autumn and 88.7% crops and 32.3% plants in the spring. This incidence is lower than in 2001, when phoma leaf spot was particularly prevalent, but similar to the high levels seen in previous years. Disease incidence on the pods was again low with 10.3% crops and 0.5% plants affected. The highest incidence on the leaves was in the South West. Severity of phoma leaf spot was also low and was not significantly different from levels recorded during the previous two seasons.

Table 4. Incidence and severity of phoma leaf and pod spot

CSG 15 (1/00) 7 Project DEFRA title project code

1999 2000 2001 2002 P value % plants affected Autumn (leaves) 26.81 a 30.78 a 51.40 c 40.95 b <0.001 Spring (leaves) 37.49 a 31.74 a 51.71 b 32.29 a <0.01 Summer (pods) 0.17 a 2.00 b 0.09 a 0.54 a <0.01

% area affected Autumn (leaves) 0.49 0.36 0.51 0.35 NS Spring (leaves) 0.75 b 0.33 a 0.21 a 0.15 a <0.001 Summer (pods) tr tr tr tr NS NS = Not significant tr < 0.05

Disease levels on the stems

As in previous years, phoma canker was the most common disease on the stems with 96.9% crops and 52.9% plants affected. These levels were very similar to those seen in 2001, which, with the exception of 2000, were the highest levels seen since 1993. Light leaf spot was the second most common disease with 81.4% crops and 30.2% plants affected. This was the highest incidence of light leaf spot recorded since 1995. The incidence of alternaria on the stems was significantly higher than in 2001 with 36.1% crops and 7.6% plants affected. Sclerotinia stem rot was higher than in 2001 with 19.6% crops and 2.3% plants affected compared to 12.8% crops and 1.1% plants affected in 2001. Powdery mildew was also markedly higher than in 2001 with 30.9% crops and 14.7% plants affected.

Table 5. Incidence and severity of stem diseases

1999 2000 2001 2002 p value % stems affected Downy mildew 0 0 0 0 NS Light leaf spot 20.24 21.83 27.14 30.23 NS Alternaria 7.77 b 6.00 b 0.26 a 7.59 b <0.001 Phoma canker 40.71 a 56.81 b 54.09 b 52.91 b <0.01 Powdery mildew 1.48 a 1.36 a 2.09 a 14.72 b <0.001 Botrytis 0.61 1.15 1.70 0.37 NS Sclerotinia stem rot 1.74 a 4.09 b 1.06 a 2.31 a <0.001

% area affected Downy mildew 0 0 0 0 NS Light leaf spot 0.76 0.87 1.41 1.34 NS Dark pod spot 0.12 b 0.17 b tr a 0.13 b <0.001 Powdery mildew 0.07 b 0.05 a 0.72 a 1.30 b <0.001 NS = Not significant tr < 0.05

Comparisons with untreated crops

Assessments of untreated crops have been included in the survey since 1998. Although in some years differences between disease levels in conventionally treated and untreated crops were small, in general higher total disease levels were recorded in the untreated crops. This is particularly evident later in the season at the summer assessment. In 2002, there was a marked difference between difference in disease levels in treated and untreated crops (Table 6) particularly in levels of light leaf spot. Levels of light leaf spot on the pods were 3.75 times higher in untreated crops. Similarly levels of phoma canker were almost 20% higher in untreated compared to treated crops. This suggests that fungicides were giving useful control of diseases in treated

CSG 15 (1/00) 8 Project DEFRA title project code

crops, although estimated yield losses remained high, indicating potential for significant improvement to levels of control achieved.

Table 6. Total disease severity on untreated crops compared to conventionally grown crops

% area affected Untreated Conventionally grown

1999 Autumn (leaves) 1.33 1.33 Spring (leaves) 1.57 1.71 Pods 1.76 1.67 2000 Autumn (leaves) 0.59 0.53 Spring (leaves) 1.78 1.28 Pods 6.08 3.24

2001 Autumn (leaves) 1.49 0.94 Spring leaves) 1.40 0.98 Pods 0.44 0.34

2002 Autumn (leaves) 1.26 0.50 Spring leaves) 1.74 1.44 Pods 4.12 2.35

Regional disease incidence and severity (disease risk)

Light leaf spot

A risk forecast has been developed for light leaf spot using survey data from the period 1990-1999 and is available on the Internet at http://www3.res.bbsrc.ac.uk/leafspot/. It has been shown that levels of disease on the pods in the previous season was a significant factor in predicting risks of light leaf spot (Gladders et al., 1995). This latest forecast is based on factors such as disease levels at the end of the previous season and also on weather and crop variables. The regional risk of light leaf spot is clearly evident from Figure 1, generally indicating higher incidence in the North and Midlands & West.

At the autumn assessment, light leaf spot was recorded in the Northern and Eastern regions only, while at the spring assessment the highest incidence was recorded in the North where 53.6% of plants were affected. The high incidence in the North is consistent with trends seen in previous years as illustrated below. During the summer, the highest incidence of pod disease was in the North where 24.2% of plants were affected. The next highest incidence of 18.7% plants affected was recorded in the East. Highest stem incidence was recorded in the North and Midlands & West where 60.5% and 36.3% stems were affected respectively.

In the autumn, severity of light leaf spot was recorded was recorded at trace levels only. Severity in the spring was highest in the North, Midlands & West and Eastern regions at 2.9%, 0.9% and 0.8% leaf area affected respectively. By the summer the highest severity on the pods was recorded in the North and East

CSG 15 (1/00) 9 Project DEFRA title project code

Figure 1. Regional incidence (% plants affected) of light leaf spot between 1990 and 1999.

Phoma

A risk forecast using survey data from the period 1990-1999 is being developed for phoma canker and is available on the Internet at http://phoma.csl.gov.uk. This forecast is based on the disease levels the previous season, raindays in August and September and the average daily temperature from November to January. The regional risk of phoma canker is clearly evident from Figure 2, generally indicating higher incidence in the East and South East.

Figure 2. Regional incidence (% plants affected) of stem canker between 1990 and 1999.

Incidence of phoma leaf spot was highest in the South West, affecting 61.3% plants in the autumn and 54.7% plants in the spring. The highest incidence of phoma canker was in the North and East where 69.4% and 68.3% stems were affected respectively. The lowest incidences were recorded in Wales (2.0% stems affected) and the Midlands & West (11.3% stems affected).

CSG 15 (1/00) 10 Project DEFRA title project code

The highest severity of phoma leaf spot in autumn was recorded in the Midlands & West (0.8% leaf area affected). In the spring, severity of phoma was low and there was little regional variation.

Incidence of viruses

Testing of bulk samples of all leaves from each plant for virus infection showed that the level of BWYV in the autumn (8.9% plants infected) was higher than in 2001 (5.6% plants infected). Levels of CaMV were similarly low (0.1% plants infected in 2002 compare to 0.2% in 2001). TuMV was again not recorded in the autumn. In the spring, BWYV was found in 11.3% of plants which was higher than the 10.2% plants affected in 2001, but lower than the 19.4% recorded in 2000. Neither CaMV nor TuMV were identified in spring samples in 2002, which is in contrast to the low levels found in 2001 (0.4% plants infected with CaMV and 0.2% with TuMV).

Table 7. Incidence of virus disease on from bulk sample testing

2000 2001 2002 Autumn (% plants) BWYV 8.78 5.64 8.95 CaMV 0.11 0.22 0.11 TuMV 0.21 0 0

spring (% plants) BWYV 19.37 10.21 11.26 CaMV 0.11 0.42 0 TuMV 0 0.21 0

Agronomic information

Previous cropping

The proportion of crops sown following winter wheat decreased from 63.8% of crops in 2001 to 46.3% crops in 2002. This is the lowest proportion since 1998 (Table 8). The proportion of crops sown following other cereals (mainly winter barley) increased to 41.1% in 2002 from 33% in 2001. The proportion of crops sown following set-aside (10.5% crops) was the highest since 1997.

Table 8. Previous cropping

% crops 1999 2000 2001 2002 Winter wheat 58.2 54.7 63.8 46.32 Other cereal 35.2 42.1 33.0 41.05 Set-aside 4.4 2.1 2.1 10.53

Sowing date

As in previous years, the largest proportion of crops was sown in August, although the proportion of late August sowings was higher than in 2001 (Table 9). A similar proportion of crops was sown in early September, but a smaller proportion was sown in late September. As in 2000, with the possible exception of light leaf spot in the spring and on the pods in the summer, which was recorded at higher levels in August- sown crops compared to September-sown crops, there was little evidence that sowing date had any effect of disease levels.

Table 9. Sowing date of crops

CSG 15 (1/00) 11 Project DEFRA title project code

% crops 1999 2000 2001 2002 < 20 August 7.8 6.4 9.4 10.5 20 – 31 August 57.8 57.4 44.8 55.8 1 – 7 September 20.0 26.6 25.0 26.3 8 – 14 September 11.1 7.5 14.6 6.3 > 14 September 3.3 2.1 2.1 1.1

Cultivar selection

Apex was again the most popular cultivar sown for the ninth consecutive year. However, the proportion grown in 2002 decreased to only 15.6% of crops compared to the second most commonly grown cultivar Escort, which accounted for 13.5% crops. The decreasing proportion of Apex continued the downward trend seen over recent years. Fortress was the third most popular cultivar in 2002 accounting for 12.5% crops, and Recital the fourth most popular accounting for 11.5% crops. The use of the restored hybrid Pronto decreased from 9.5% crops in 2001 to 5.2% in 2002.

There was little evidence of a relationship between resistance rating and disease levels, particularly at the autumn and spring assessments. However, at pod development, the highest severities of light leaf spot on the pods were recorded on Royal (2.0% pod area affected) and Recital (1.9% pod area affected) compared to the national mean of 0.6% pod area affected. This is in contrast to the NIAB resistance ratings of 6 and 7 respectively. The highest severities of light leaf spot on the stems were recorded on Apex (2.5% stem area affected) and Pronto (3.4% stem area affected) compared to the national mean of 1.3% stem area affected. Powdery mildew was markedly higher on Fortress (3.8% pod area affected and 2.6% stem area affected) and Royal (4.3% pod area affected and 5.5% stem area affected) compared to national means of 1.0% pod area affected and 1.3% stem area affected. Dark pod spot on the pods was highest on Apex (1.5% pod area affected) and Royal (1.4% pod area affected) compared to a national mean of 0.7% pod area affected. Phoma canker was slightly higher on Royal (1.3% stem index) and Apex (1.4% stem index) compared to a national mean of 0.9% stem index, possibly reflecting Royal’s low resistance rating of 4 for resistance to stem canker.

There has been a decreasing trend in the use of cultivars with high levels of resistance to phoma and a major switch to cultivars considered susceptible to light leaf spot during the last few years (Table 10). This has significant implications for disease control, and indications are that cultivar resistance is not being taken into consideration when determining suitable fungicide control programmes. Data for 2002 show a slight improvement, with a decrease in the proportion of crops with low resistance to phoma and light leaf spot. However, the overall level of susceptibility to phoma is still high and indicates a significant risk factor for this disease.

Table 10. Per cent crops sown with NIAB resistance rating of 5 or less, i.e. ‘susceptible’ cultivars.

% crops 1999 2000 2001 2002 Light leaf spot 0 43.4 33.3 17.4 Stem canker 25.9 65.1 51.1 37.2

Pest incidence

In previous years, rape winter stem weevils (Ceutorhynchus picitarsis) were found only in the Northern and Eastern Regions in the autumn samples, except in 1996 when a small number were recorded in the South Eastern Region. In the autumn samples taken in 2001, very low numbers of rape winter stem weevils were recorded in both the Northern and Eastern regions, reflecting the low or zero counts noted in previous years. Numbers in spring were lower than in autumn, and rape winter stem weevil larvae were only detected in the South Eastern region (mirroring the findings in 1996).

A mean of almost 0.15 cabbage stem flea beetle (Psylliodes chrysocephala) larvae per plant was found in the autumn 2001 samples, lower than in 2000, and confirming the break of the previous trend towards increasing CSG 15 (1/00) 12 Project DEFRA title project code

incidence recorded during 1998-2000 (Figure 3). However, despite this, greater numbers were found in the crops than in 1997. The larval infestation resulted in a mean of 6.1% of the leaves sampled bearing characteristic leaf scars caused by this pest. The range of infestation levels in individual fields in autumn 2001 was 0 - 3.1 larvae per plant, slightly smaller than in the previous year. Thus none of the fields monitored exceeded the current treatment threshold of five larvae per plant or 50% leaves scarred (Walters et al., 2001).

0.8 m e t 0.7 s t

e n a g

l 0.6 a p b r b

e 0.5 a p c

s f e

o 0.4 l

t r e e e

b 0.3 b

m a u e

l 0.2 n f

a 0.1 e

M 0 1996 1997 1998 1999 2000 2001 2002

Figure 3. Mean number of cabbage stem flea beetles recorded in autumn samples

Pesticide Use

Fungicide Use

The proportion of crops receiving a fungicide treatment in 2002 was 94.6%, which was a slight increase on 2001( 90.9% treated). Eighty-three per cent received a treatment before the spring, which was an increase compared to 2001 (65.9%), but was similar to levels recorded in 2000 (80.0%) (Table 11). The proportion of crops treated in the spring increased markedly from 54.5% in 2001 to 67.7% in 2002 reflecting increasing use of fungicides as growth regulators. The proportion treated at flowering declined to 26.9% in 2002 compared to 43.2% in 2001, which was the lowest proportion of crops treated at flowering since the early 1990s. The proportion of crops receiving a post-flowering treatment remained similar to 2000 and 2001.

Table 11. Number of sprays applied

% crops 1999 2000 2001 2002 Untreated 6 8 9 5 One spray 35 23 22 22 Two sprays 32 40 47 53 Three sprays 25 18 23 17

Mean no. of sprays 1.9 1.7 2.0 2.0

The cost of fungicide use per hectare has been estimated at Figure 4. The total estimated cost per hectare was £26, the majority of which was spent on autumn/winter treatments (£11.60 per ha) and spring treatments (£9.50 per ha). The proportion of estimated costs attributed to autumn/winter fungicide applications increased CSG 15 (1/00) 13 Project DEFRA title project code

from £9.20 per ha in 2001 to £11.60 in 2002 and the proportion of estimated costs attributed to spring applications rose from £7.60 per ha in 2001 to £9.50 in 2002. This was in accordance with the increase in proportion of crops treated during autumn/winter and spring. The total estimated costs of fungicide treatments was higher than 2000 and 2001 and similar to 1999. ) a

h 30

r e p 25 £ (

s t s 20 o c

e d i 15 c i g n

u 10 f

d e t

a 5 m i t s

E 0 1999 2000 2001 2002

Autumn/winter Spring Flowering Post-flowering Figure 4. Estimated fungicide costs

Insecticide Use

The proportion of crops receiving an insecticide spray increased slightly from 72.7% in 2001 to 79.6% in 2002, which was in accordance with levels observed between 1998 and 2000. This is due to the rise in the proportion of crops receiving an insecticide treatment in the autumn/winter, which has risen from 62.5% crops in 2001 to 75.3% in 2002. As in all previous years, the majority of insecticide treatments were applied in the autumn/winter. A similar proportion of crops received treatments in the spring and at flowering in 2002 compared with 2001.

Seed Treatment Use

Seed treatments were used on 67.8% of crops in 2002, a slight increase from the 60.2% in 2001, but lower than in 2000 (71.3% crops) and 1999 (81.2% crops). Products containing thiram or iprodione were equally popular. Also popular was beta-cyfluthrin/imidacloprid which was used on 26.9% of crops.

Molluscicide Use

Fifty-six per cent of fields were treated with molluscicide, a lower proportion than in 2001, but similar to previous years. Of the fields treated, only 8% received more than one application, a significant decrease from the proportion treated in 2001 (43.3%).

Herbicide, Growth Regulator and Desiccant Use

The proportion of crops treated with a herbicide has remained consistently high since their use was first monitored in 1998. In 2002, 98.9% of crops received a herbicide treatment. The majority of crops (93.5%) were treated in the autumn/winter. Thirty per cent of crops were treated pre-emergence, a similar proportion to that recorded in 2000 and 2001. Twenty per cent were treated in the spring, which was markedly fewer than in 2001 when 42% of crops were treated. Few crops were treated with a herbicide after the spring.

The most popular active ingredient used pre-emergence was metazachlor, used either alone or in co- formulation.

CSG 15 (1/00) 14 Project DEFRA title project code

Only 1.1% of crops received a growth regulator. However, growth regulation is also achieved through the use of some triazole fungicides, and 51.6% of crops were treated with either tebuconazole or metconazole, both of which are noted for their properties as growth regulators (particularly when applied in the spring).

Desiccants were used on 32.3% of crops which was very similar to 2001. Glyphosate and diquat were the only active ingredients recorded in 2002, with glyphosate being by far the most popular

Conclusions

Disease levels have fluctuated considerably during the years 1999 to 2002, with the incidence of some diseases reaching unprecedented levels during this period. Levels of all major stem diseases in 2002 were above the ten-year means, as were levels of light leaf spot and dark pod spot on the pods. High levels of phoma canker and light leaf spot continue the increasing trend in incidence of these diseases over the last four or five years. Phoma canker is now the major threat to the industry and losses in 2001 were estimated to have cost £36 million (Turner et al., 2001). Losses for light leaf spot were £18M whilst dark pod spot and sclerotinia each cost less than £1M in lost yield. Yield losses in 2002 were higher than in 2001 mainly due to an increase in light leaf spot. The need for more reliable forecasts to aid the targeting of sprays for both pests and diseases is clearly highlighted; a requirement which is currently being addressed in the PASSWORD project in the UK which aims to develop a decision support system for pests and diseases of oilseed rape. It is clear from analyses of seasonal and regional fluctuation in disease levels over time that the weather is a key driver in these disease epidemics. Data from the UK meteorological office show that the average rainfall in 1997 was equal to the 30-year mean, and that rainfall levels in every year since have greatly exceeded this figure. Similarly, average temperatures for central England have exceeded the 30-year mean in nine out of the last ten years. Whilst these weather conditions may be a transient feature, the possibility of climate change must be considered. Data show that if these conditions persist there will be a considerable impact on disease activity on an annual basis.

The current levels of disease activity indicate that the potential to control these diseases using chemical methods is still not being fully exploited. This is despite a record level of fungicide applications in 2002. Use of fungicides has continued to increase throughout the last ten years with over 90% of crops treated annually between 1998 and 2001. The total cost of applications peaked in 1998 at £27.50 per hectare and have since reduced slowly to a figure of £23.60 per ha in 2000 rising again to £26.00 per ha in 2001. The frequency of use of sprays, the choice of active substances and the timing of the sprays all indicated that growers were aware of the threat posed by diseases and were taking steps to control them. Advances in the understanding of disease epidemiology and control have brought about improvements in the timing of fungicide sprays to control particular pathogens. Data from the surveys support the importance of early autumn sprays for control of light leaf spot and spring sprays for control of phoma canker. Data suggest that fungicide use reduced disease incidence but that the improvement in level of control by changes in timing or number of sprays was, in many cases, minimal. Poor fungicide timing and limited information on variation in epidemic development are contributing factors. Improved chemistry and the availability of new information on timing have not significantly decreased overall fungicide use nor does it appear to have made a major improvement to disease control on a national scale. There is clearly a requirement for more accurate assessment disease risk and early warning of disease development in order to determine the need for and timing of fungicide applications in order to improve gross margins, particularly given the increasing economic pressures on farming. Use of insecticides has also remained high during the survey period with 78% of crops treated in 2002. This is despite levels of pest infestation being well below threshold in the vast majority of fields monitored.

The benefits of cultivar resistance for disease control are also not being exploited to the full. The most severe and widespread attacks of light leaf spot occurred in 1994 and 1995, and during this epidemic 56% of cultivars grown had a NIAB resistance rating <=5 for light leaf spot indicating that they were susceptible cultivars. Given the current threat from diseases and the need to minimise inputs, growers need to limit potential disease problems by utilising cultivars that have a higher resistance rating to stem canker and light leaf spot. However, despite this, data from 2002 show continued use of susceptible cultivars. Although overall levels of susceptibility were lower in 2002, there is still significant implications for disease risk in the future. Given the current threat from diseases and the need to minimise inputs, growers need to reduce potential disease problems by utilising cultivars that have a higher resistance rating to stem canker and light leaf spot. Control strategies need to be reassessed to encompass a more targeted and integrated approach using cultural

CSG 15 (1/00) 15 Project DEFRA title project code

practice and forecasting alongside chemical control. This would result in additional economic and environmental benefits and aid the move towards sustainable crop production.

Overall the disease survey has provided reliable data on the status of individual pests and diseases in 2002, which can be put into context by comparisons with disease levels over a longer period of time. The influence of cultural practice and economic change can be clearly seen in the information gathered and there are indications of major risks for the future. The surveys provide the only independent source of information on actual farm practice and can be utilised to provide a clearer understanding of disease epidemiology and a view of where research and government policy should be targeted to promote a more sustainable approach to crop production and reduce the risk of pollution resulting from farming practice. The recorded use of routine pesticide applications only serve to highlight the need for more reliable forecasts and risk assessments in order to target treatments more effectively while minimising yield loss. Development of such forecasting systems and decision support systems should remain a priority, and have been successfully developed and validated using the oilseed survey data. e.g., light leaf spot forecast and phoma forecast (as part of the PASSWORD decision support system).

Publications

Results from the project have been used widely for publication and presentation at farmer meetings and Conferences, including industry-sponsored events. There have press features in the agricultural press and strong exploitation through links with the PASSWORD project. The change to autumn fungicide application reflects promotional activity based on Defra-funded disease surveys, epidemiology and disease management.

Papers

Evans N, Steed JM, Welham SJ, Antoniw JF, Turner JA, Gladders P, Fitt BDL, 2002. Interactive forecasting on the internet of light leaf spot (Pyrenopeziza brassicae) risk for winter oilseed rape. International Organisation for Biological Control Bulletin, 25, 103-107. Sutherland KG, Evans N, Fitt BDL, Gladders P, Morgan D, Northing P, Turner JA, Walters KFA, 2002. Development of a DSS for diseases and pests in winter oilseed rape. Proceedings Crop Protection in Northern Britain 2002. Turner JA, Gladders P, Elcock S, Walters K, Wright DM, 2002. Winter Oilseed rape: survey of pests and diseases 2001/2002. Internal report, Central Science Laboratory and ADAS. Turner JA, Elcock SJ, Gladders P, 2002. A review of pest and disease problems in winter oilseed rape in England and Wales. Proceedings 2002 British Crop Protection Conference – Pests and Diseases 2, 555- 562. Hardwick NV, Turner JA, Slough JE, Elcock SJ, 2002. Oilseed rape and cereal diseases – how are farmers responding to their control? Proceedings 2002 British Crop Protection Conference – Pests and Diseases 2, 903-910. West JS, Huang Y-J, Steed JM, Leech PK, Fitt BDL, Gladders P, 2002. New perspectives on the epidemiology and management of phoma stem canker of winter oilseed rape in England. Proceedings of the BCPC Conference – Pests and Diseases 2002, 2,563-568.

Press articles

Watch out ….. phoma. Farmers Weekly 6 September 2002 p.51 OSR autumn disease threat. Arable Farming 7 September 2002, p.41 Phosphate can help OSR beat autumn threat. Farmers Weekly 20 September 2002 p. 54. Targeted phoma fungicides. Arable Farming 21 September 2002. Race to stop infection, Crops 21 September 2002 p.8 Spray early and spray twice. Arable Farming 22 October 2002 p. 8 &10 Let autumn downpours dictate phoma tactics. Farmers Weekly 25 October 2002 p.56. Tackle leaf spot before you see it. Farmers Weekly 25 October 2002 p.52.

Demonstrations, workshops and seminars

CSG 15 (1/00) 16 Project DEFRA title project code

Gladders, P. (2003) ADAS Crop Centre Conferences at Lincoln, York and Wyboston. Disease control in break crops and cereals. Gladders, P. (2003) HGCA Disease Management Roadshows at Cambridge (25 Feb) and Lincoln (26 Feb). Assessing the risk of yield loss in winter oilseed rape. Discussion forum of cost-effective strategies. Gladders, P. Baierl, A. and Northing, P. (2002) Sprays and sprayers Event, Syngenta, Whittlesford, Cambs 25-26 June

Web pages

Summary information, interactive searches, risk maps and annual reports from this project can be accessed via the web at www.crop-disease-surveys.com.

References

Fitt BDL, Gladders P, Sutherland KG, Turner JA, Welham SJ, Biddulph JE, Hardwick NV, Wale SJ, 1998. Epidemiology and forecasting of light leaf spot (Pyrenopeziza brassicae) on winter oilseed rape in the UK. Integrate control in Oilseed Crops, IOBC Bulletin Vol 21 (5), 15-22. Gladders P, Fitt BDL, Welham SJ, 1995. Forecasting developments of light leaf spot (Pyrenopeziza brassicae) epidemics on winter oilseed rape. Proceedings Ninth International Rapeseed Congress 1995 Cambridge: GCIRC, 1001-1003. Hardwick NV, Turner JA, 1994. Fungicide use on winter oilseed rape in England and Wales, 1986-1993. Quatrieme Conference Internationale sue les Maladies des Plantes 1994 Bordeaux: ANPP, 1163-1170. Hill SA, 1984. Methods in Plant Virology. Oxford: Blackwell Scientific Publications. Sansford CE, Fitt BDL, Gladders P, Lockley KD, Sutherland KG, 1996. Oilseed rape: disease development, forecasting and yield loss relationships. Home-Grown Cereals Authority Project Report OS17. 185pp. Sylvester-Bradley R, 1985. Revision of a code for stages of development in oilseed rape (Brassica napus L.). Aspects of Applied Biology 10, Field trials methods and data handling 1985. Cambridge, UK: ABB, 395- 400. Sylvester-Bradley R, Makepeace RJ, Broad H, 1984. A code for stages of development in oilseed rape (Brassica napus L.). Aspects of Applied Biology 6, The agronomy, physiology, plant breeding and crop protection of oilseed rape 1984 Cambridge, UK: AAB, 399-419. Turner JA, Elcock SJ, Hardwick NV, Gladders P, 2000. Changes in fungicide use on winter oilseed rape in England and Wales during the 1990s. Proceedings Brighton Crop Protection Conference 2000 – Pest and Diseases, Turner J A, Elcock S J, Walters K F A, Wright D M, Gladders P, 2001. Winter oilseed rape: survey of pests and diseases (2000/2001). CSL report. 51pp. Walters KFA, Lane A, Cooper DA, Morgan D, 2001. A commercially acceptable assessment technique for improved control of cabbage stem flea beetle feeding on winter oilseed rape. Crop Protection 20, 907-912.

CSG 15 (1/00) 17