Project
title / IMPROVED MANAGEMENT OF GRASS WEEDS IN CEREALS.
????? / MAFF
project code / CE0612

ministry of agriculture, fisheries and foodCSG 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
MAFF, Area 6/01
1A Page Street, London SW1P 4PQ
An electronic version should be e-mailed to
Project title / IMPROVED MANAGEMENT OF GRASS WEEDS IN CEREALS.
?????
MAFF project code / CE0612
Contractor organisation and location / ADAS Boxworth
Cambridge
CB3 8NN
Total MAFF project costs / £ 310,897
Project start date / 01/04/1997 / Project end date / 30/09/01
Executive summary (maximum 2 sides A4)
To tab in this section press the tab key and the Control key together
Press the DOWN arrow once to move to the next question.

CSG 15 (1/00)1

Project
title / IMPROVED MANAGEMENT OF GRASS WEEDS IN CEREALS.
????? / MAFF
project code / CE0612

Grass weeds, notably black-grass (Alopecurus myosuroides), bromes (Bromus spp.) and wild-oats (Avena spp.) are threatening the sustainability of intensive cereal production on heavy land by, for example, black-grass and wild-oats developing resistance to herbicides and in surface water contamination from the use of residual herbicides to control these weeds.

DEFRA policy is to develop strategies for both reducing pesticide inputs and alleviating resistance. Other work is looking at resistance, this project concentrates more on ways of minimising herbicide use.

This report covers four different areas of work relating to the potential to adjust herbicide dose. These are; a) the interaction of weather factors with the activity on black-grass of the herbicides isoproturon and clodinafop-propargyl b) the use of fluorescence to identify the timescale of herbicide entry into plants and hence whether there was a need or not to re-spray c) the use of leaf growth rate measurements to indicate whether plants are growing actively or under stress allowing herbicide dose to be adjusted and d) the interaction of cultivations with time of drilling and seed rates on black-grass populations and the levels of control.

a)The interaction of weather factors with the activity of herbicides isoproturon and clodinafop-propargyl for the control of black-grass.

In each of three seasons (1997/98, 1998/99, 1999/00) field trials were undertaken on a heavy clay soil of Hanslope series, in a different field at ADAS Boxworth, known to be relatively free of natural populations of grass-weeds. Black-grass seed (of known non-resistant source) was broadcast by hand at a rate of 400 seeds m-2 prior to drilling winter wheat (cv. Equinox) on the same day, on 2 October 1997, 19 October 1998 and 12 October 1999 respectively.

The herbicide clodinafop-propargyl (CP) was applied at a rate of 15, 30, 60 (full rate) g a.i. ha-1, and isoproturon (IPU) at 625, 1250, 2500 (full rate) g a.i. ha-1, with an untreated control for each date. Sprays were applied on eight occasions between 29 October 1997 to 25 February 1998, on ten occasions between 23 November to 14 April 1999 and on 15 occasions between 12 November 1999 and 6 April 2000 respectively (Figure 1).

Black-grass growth stage at treatment ranged between 1-2 leaves at the first timing up to 2-3 tillers at the last timing in each season.

The mean percentage reduction in black-grass panicle numbers from the untreated control plots were calculated for each herbicide at all doses and timings and the dose required to kill half the black-grass plants (ED50) was calculated. A large range of meteorological factors were summed into periods of up to 14 days before and after spraying and were included in a stepwise multiple linear regression analysis. The factors explaining most of the variation in ED50 for each of the two herbicides were identified.

Weather factors which accounted for a high percentage (between 86-97%) of the observed variation in that season were identified for each herbicide but the factors were not the same in each season. It may be that more sophisticated statistical procedures for dealing with multi-variate analyses could be explored in the future using the data set collected from these experiments in which the influence of factors other than weather (e.g. site and soil type) on activity have been minimised.

b)The use of fluorescence to identify the timescale of herbicide entry into plants.

Isoproturon (and some other herbicides such as cyanazine) acts by interfering with photosynthesis in the plant and photosynthetic rates can be rapidly and non-destructively estimated using equipment that detects changes in the fluorescence from photosystem II. Traditionally plants have required to be kept in the dark for a period of time before taking fluorescence readings. This has made the technique less easy to routinely use in the field but for the instruments we have used in this project this is unnecessary. Of the herbicides tested on black-grass, barren brome and wild-oats in the glasshouse, isoproturon and cyanazine activity could clearly be detected within 1 day of treatment and several days before visible symptoms were obvious. This technique could also differentiate rates of entry of isoproturon into barren brome growing under different moisture regimes and between Peldon (resistant) and Rothamsted (susceptible) black-grass plants in an isoproturon solution.

An ability to predict the need for re-spraying in cases where symptoms were slow to develop would minimise unnecessary herbicide use. This technique is limited to herbicides which interfere with photosynthesis. The importance of isoproturon, which is the major photosynthetic inhibitor herbicide used in cereals, has decreased over the duration of this project but still has a role in resistance management. Further studies to test this under a wider range of field conditions would still be required to progress the use of this.

c)The use of leaf growth rate to indicate whether plants are under stress.

As a generalisation more herbicide is required to kill stressed compared to unstressed plants. Therefore some means of identifying when plants are stressed at the time of treatment could help in deciding the appropriate dose. We have explored the possibility of using the plant itself as the indicator and have measured leaf growth rates. We established by using plants in pots in a glasshouse that the rate of growth of wheat could be used as a good indicator of that of the black-grass. Leaf lengths and areas were measured on wheat plants in the field at Boxworth over winter (2000-2001) and measurements fed in to the computer programme developed by Jonathan Storkey at IACR-Rothamsted under funding within Roame CE0616. Agreement between the measurements and the predictions in one season was good and hence there may be the potential to use models for the growth rates of young plants as indirect indicators of plant stress. This would need to be confirmed for other seasons and could be used in three main ways to indicate;

1) pre-spray ‘stress’ when the full rate or even no herbicide should be used or ‘active growth’ when a reduced rate could be contemplated. This would need further calibration against herbicide activity.

2) the rate of herbicide entry for those herbicides which stop growth and are not appropriate to be monitored by fluorescence.

3) when active weed growth starts in the spring in order to target herbicides which work best on actively growing plants.

d) The interaction of cultivations with time of drilling and seed rates on black-grass populations and control.

Following the MAFF review in August 2000 the emphasis within this project was changed and in the final year two field sites were set up at Boxworth (ADAS) and Woburn (IACR-Rothamsted).

Both experiments were adversely affected by the extremely wet weather during autumn 2000/spring 2001 when rainfall was the highest on record. While the early drilled wheat was sown as planned in mid – late September 2000 at both sites, the later sown winter wheat was not drilled until November at Boxworth and January 2001 at Woburn and spring wheat until May 2001. However, despite these limitations both experiments achieved their key objectives.

The key conclusions were;

  • There was no clear evidence at either site that shallow stubble cultivations prior to the main cultivation or immediately after harvest encouraged black-grass to germinate.
  • Early drilling resulted in higher black-grass populations than later drilling at both sites (approx. 2x greater).
  • Ploughing was done 11 and 28 days pre-early drilling at Boxworth and Woburn respectively. This resulted in lower black-grass populations in the crop at Woburn (only had early drill date) but only at the ‘normal’ drill date at Boxworth.
  • Reduced crop seed rate was associated with higher black-grass infestations at the early sowing date at both Boxworth and Woburn (only one date) but was not significant at the ‘normal’ sowing at Boxworth.
  • There was no clear evidence that the use of glyphosate pre-cultivation/sowing reduced the black-grass infestations in the crop at Woburn (only ‘normal’ date). At Boxworth where there were both sowing dates there were significantly less panicles in the ‘normal’ sown crop where glyphosate had been used but not from the early sowing.
  • Spring sown wheat resulted in substantially lower black-grass infestations (only 18% or 36% of those in September or January sown wheat respectively).
  • There was no evidence that autumn cultivations influenced the subsequent black-grass infestation in spring-sown wheat. The effect of cultivation as compared with leaving stubbles uncultivated during the winter was neutral in relation to black-grass.
  • The decline of black-grass seeds in the seedbank at both sites was consistent with previous studies (70-80% annual decline rate).

The various aspects within this project were all aimed in different ways at addressing the MAFF policy of reducing herbicide inputs. This can be done by understanding the key conditions when herbicides work best and hence when it is appropriate to reduce herbicide doses but also when they will work less well and an alternative strategy would be more appropriate. The field experiments in the final year demonstrated the importance of understanding the influence of cultivations, stubble management and drilling date in contributing to grass weed control in winter cereals as part of an integrated strategy.

The cultivation work is being continued under LINK funding (SAP 127).

CSG 15 (1/00)1

Project
title / IMPROVED MANAGEMENT OF GRASS WEEDS IN CEREALS.
????? / MAFF
project code / CE0612
Scientific report (maximum 20 sides A4)
To tab in this section press the tab key and the Control key together
Press the DOWN arrow once to move to the next question.

CSG 15 (1/00)1

Project
title / IMPROVED MANAGEMENT OF GRASS WEEDS IN CEREALS.
????? / MAFF
project code / CE0612

Grass weeds, notably black-grass (Alopecurus myosuroides), bromes (Bromus spp.) and wild-oats (Avena spp.) are threatening the sustainability of intensive cereal production on heavy land by, for example, black-grass and wild-oats developing resistance to herbicides and in surface water contamination from the use of residual herbicides to control these weeds.

DEFRA policy is to develop strategies for both reducing pesticide inputs and alleviating resistance. Other work is looking at resistance, this project concentrates more on ways of minimising herbicide use.

This report will cover four different areas of work relating to the potential to adjust herbicide dose. These were; a) the interaction of weather factors with the activity on black-grass of the herbicides isoproturon and clodinafop-propargyl b) the use of fluorescence to identify the timescale of herbicide entry into plants and hence whether there was a need or not to re-spray c) the use of leaf growth rate measurements to indicate whether plants are growing actively or under stress allowing herbicide dose to be adjusted and d) the interaction of cultivations with time of drilling and seed rates on black-grass populations and the levels of control.

a) The interaction of weather factors with the activity of herbicides isoproturon and clodinafop-propargyl for the control of black-grass.

METHODS

Each season field trials were undertaken on a heavy clay soil of Hanslope series, in a different field at ADAS Boxworth, known to be relatively free of natural populations of grass-weeds. Plots measured 6 x 3 m. In 1997/1998 there were three blocks in a fully randomised design. In 1998/99 and 1999/2000 there were four blocks in a randomised block split plot, with herbicide timings as the main plots.

Black-grass seed (of known non-resistant source) was broadcast by hand at a rate of 400 seeds m-2 prior to drilling winter wheat (cv. Equinox) on the same day, on 2 October 1997, 19 October 1998 and 12 October 1999 respectively.

The herbicide clodinafop-propargyl (CP) was applied at a rate of 15, 30, 60 (full rate) g a.i. ha-1, and isoproturon (IPU) at 625, 1250, 2500 (full rate) g a.i. ha-1, with an untreated control for each date. Sprays were applied on eight occasions between 29 October 1997 to 25 February 1998, on ten occasions between 23 November to 14 April 1999 and on 15 occasions between 12 November 1999 and 6 April 2000 respectively (Figure 1).

Black-grass growth stage at treatment ranged between 1-2 leaves at the first timing up to 2-3 tillers at the last timing in each season.

Herbicides were applied using a knapsack sprayer and 3m boom, operating at a pressure of 2.0 bars delivering 225 litres ha-1 through 02 F 110 nozzles set at a height of 35cm above the target leaf. No other grass-weed herbicides were applied to the trials, but normal farm practice occurred for all other inputs, in all years.

Effectiveness of the herbicides was assessed by counting panicles of black-grass on 19 June 1998 and 22 June 1999 and 27 June 2000 respectively, using 5 x 0.1m-2 quadrats per plot.

The mean percentage reduction in black-grass panicle numbers from the untreated control plots were calculated for each herbicide at all doses and timings. Log fractions of label dose to response curves were fitted to these data using combined controls and the Whadley’s problem variant of probit analysis (Ross, 1987). From the fitted curves the log dose and its 95% confidence limits required to kill half the black-grass plants (ED50) was calculated and the values back transformed to fractional dose. In practice much higher levels of kill are required but these higher levels are near the asymptotes, and thus can only be very poorly estimated and are unsuitable for further analysis. Meteorological data collected were daily totals of rainfall (R, mm), solar radiation (I, MJ m-2) windrun (W, km) and daily maximum (TX, C) and minimum (TN, C) temperatures and 10 (TU, C) and 20cm (TL, C) deep average daily soil temperatures. Dry Bulb (TD, C) and wet bulb (Tw,, C) temperatures were recorded at 0900 daily and vapour pressure (V, mbar) and relative humidity (H, %) calculated. Potential evaporation (P, mm d1), was also calculated using standard methods (Thompson et al.,1981). The data were collected from a meteorological station within 1 km except TD and TW , which were collected at University of Cambridge Botanic Gardens 10 km from the site. All the above factors, summed into periods of up to 14 days before and after spraying were then included in a stepwise multiple linear regression analysis. The factors explaining most of the variation in ED50 for each of the two herbicides were identified.

RESULTS

The mean number of black-grass panicles in the untreated control plots for the three seasons was 244, 260 and 285 plants m-2 respectively.

The percentage reduction in black-grass panicles, compared to the untreated controls by the full dose of IPU and CP (Figure 1), illustrated a wide variation in the level of control with both herbicides in all seasons.

In 1997/98 the efficacy of the two herbicides varied considerably between the eight spray occasions with ED50 varying between 0.42 and 0.94 for IPU and 0.22 and 0.67 for CP (Figure 2). There was no simple relation of ED50 with time for either herbicide, although there was the suggestion of a mid season increase in effectiveness in CP. For IPU activity was greater on 29 October 1997, 6 November 1997, 1 December 1997 , and 25 February 1998 when the ED50 values were below 0.52. For CP activity was greater between 25 November 1997 and 15 December 1997, when the ED50 values were all below 0.41. There were considerable differences in the relative performance of IPU and CP between dates. IPU was less active than CP on 25 November 1997, 15 December 1997 and 22 January 1998 contrasting with the greater activity of IPU than CP on 6 November 1997. On the other dates ED50 was very similar for the two herbicides. The overall balance of effectiveness throughout the season was in favour of CP.

Figure 1. The % reduction from the untreated controls of black-grass, sprayed with isoproturon (IPU – open columns) and clodinafop- propargyl (CP – hatched columns), at range of spray timings in 1997/98 and 1998/99. Standard errors are shown as a vertical line where greater than the thickness of the edge of the histogram bar.

Figure 2. The ED50 values for clodinafop-propargyl (CP – hatched columns) and isoproturon (IPU – open columns) applied to black-grass, at a range of spray timings in 1997/98 and 1998/99.

In 1998/99 as in the previous season there was variation in the efficacy of the herbicides between application dates with ED50 varying between 0.28 and 1.17 for IPU and 0.32 and 1.51 for CP. These values extended the range in ED50 encountered in the previous year, except for the most effective application for CP. The general pattern was of lower activity for both herbicides early in the season compared with later in the season. For IPU activity was greatest on 11 January 1999 (ED50 0.28) and relatively high between 21 January 1999 and 14 April 1999 with ED50 below 0.52, the discriminating value used in the previous season. For CP there were only two dates (21 January 1999 and 14 April 1999) when there was greater effectiveness than the value (0.41) previous discriminating value. Again, there were considerable differences between the activities of IPU and CP on the different dates. Of the nine comparisons IPU was more active on five occasions with larger differences earlier in the season On two occasions CP was more active than IPU and on the remaining dates (23 November 1998 and 12 February 1999) the activities of the two herbicides were similar. The overall balance of effectiveness in this season was in favour of IPU, contrasting with the previous season.