arable acronyms analysed – a review of integrated arable farming systems research in western...

Ann. appl. Biol. (1994), 125, 399-438 Printed in Great Britain 399

Arable acronyms analysed - a review of integrated arable farming systems research in Western Europe

By J M HOLLAND’x2, G K FRAMPTON2, T CILG12 and S D WRATTEN3 ‘The Game Conservancy Trust, Fordingbridge, Hampshire SP6 1 EF, U K

2Department of Biology, University of Southampton, Bassett Crescent East, Southampton, SO16 7PX, U K

3Department of Entomology and Animal Ecology, Lincoln University, P 0 Box 84, Canterbury, New Zealand

(Accepted 22 June 1994)

Summary Arable production has come under increasing economic and environmental

pressures, especially in the last decade. These have derived from over-production, decreased farm incomes and a concern with the possible environmental effects of intensive pesticide use associated with such intensive cultivation. A number of long-term research programmes on integrated farming systems and their sustainability have recently been completed or are currently under way. In the UK, these include the ‘Boxworth’ project, ‘SCARAB’, ‘TALISMAN’, RISC, ‘LINK Integrated Farming Systems’, ‘LIFE’ and the demonstration-only programme ‘LEAF’. These projects are reviewed in terms of their objectives, designs and results to date, and are compared with some parallel programmes in Germany, the Netherlands, Switzerland and France.

Key words: Western Europe, integrated farming, farming systems, Boxworth, SCARAB, TALISMAN, RISC, LINK IFS, LIFE, LEAF, Lau- tenbach, INTEX, Nagele, Third Way

Introduction Since the late 1970s there has been considerable interest in reduced input or ‘integrated’

arable farming systems. This interest was generated because of changes in the economics of arable crop production, public pressure to reduce European Union (EU) food surpluses, environmental concerns and agronomic factors (Greig-Smith, 1992).

The price of cereals within the EU has declined in recent years although it is still maintained above the world market price through subsidies. Recently there has been pressure to reduce the level of Common Agricultural Policy (CAP) support and reduce the subsidies on international exports designated by the General Agreement on Tariffs and Trade (GATT) (Wall, 1992). The farmers’ production costs, however, have steadily increased whilst the ‘farm-gate’ price has decreased, with the result that farmers’ gross margins have declined (Murphy, 1990). The response of farmers to these changes has depended on (a) the relationship between fixed and variable costs, (b) the scope for diversification, (c) the potential yield capacity of farms and (d) farmers’ personal inclination (Prew, 1993). Some farmers have already adopted reduced-input systems in direct response 0 1994 Association of Applied Biologists

400 J M HOLLAND. G K FRAMPTON. T CILGI AND S D WRATTEN

to economic pressures. This is. however. without the fore-knowledge of the long-term ecological and agronomic implications of such changes.

The grain surpluses which have accumulated in the EU were a result of the EU policy for European self-sufficiency in food production. food security and maintenance of farm incomes (de Wit, Huisman & Rabbinge, 1987). In addition, the huge increases in yield per unit area, achieved by the development of high-yielding varieties, high inputs of fertilisers and agrochemicals combined with mechanisation have added to this surplus. In the last three decades. for example, the area of wheat in Europe has increased only by &9% whereas the total production has more than doubled; the world stockpiles during this period were on average 28% o f annual world consumption (Murphy, 1991). Consumption within the Organisation for Economic Co-operation and Development (OECD) countries is expected to remain static or to decrease. output is expected to rise further, but world trade is expected to increase only slightly. For example, world trade in wheat may increase by 20 million tonnes by the year 3000. but the EU alone will expect to produce 30 million tonnes in excess of consumption by the late 1990‘s (Murphy, 1991) without the implementation of set-aside whereby grain producers are subsidised for taking land out of production.

The environmental impact of farming which produces surpluses has also come under increased scrutiny in the last decade. As a result, in the UK new and comprehensive Codes of Good Agricultural Practice for the Protection of Water and Air have been published (Anon.. 1991). The Ministry of Agriculture. Fisheries and Food (MAFF) in the UK now provides free advice on conservation and pollution matters. Area-specific schemes have been implemented to manage nitrate leaching. These include ‘Nitrate Sensitive Areas’ where farmers are paid to manage their nitrogen inputs to prevent leaching into ground waters. ‘Nitrogen Advisory Areas‘ also exist for which detailed advice is given but no payments are made. Environmentally Sensitive Areas have also been established in which farmers are subsidised if they farm less intensively (Archer & Lord, 1993). On a wider scale the Countryside Commission administers the ‘Countryside Stewardship Scheme’ in which participants receive financial support and management advice for restoring or enhancing a range of traditional English landscapes and wildlife habitats (Anon., 1993). Pesticides have also received attention. Long-term monitoring studies by The Game Conservancy Trust have highlighted the inimical effects of modern farming methods on invertebrates and game birds such as the grey partridge (Perdisperdix L. ) (Potts, 1986). Furthermore, of 130 species nf birds which breed regularly on farmland in the UK. 16 have declined in population significantly. including nine of the 15 species which reside in the cereal ecosystem throughout the year (Marchant, Hudson. Carter & Whittington. 1990). The short- and long-term effects of pesticides on non-target invertebrates are discussed in detail in Burn (1989) and Jepson (1989). In the EU, the tolerance for pesticides in drinking water is 0.1 pg of any pesticide per litre of water (Roberts, 1989). These are the minimal detectable limits and do not reflect the toxicity of the pesticide; however. because of the high cost of purifying water the only economically viable method of achieving these limits is to change agricultural practices

‘There are also many agronomic factors to consider. The intensive use of pesticides has led to the development of resistance in insect pests, weeds and diseases, subsequently I-educing the efficacy and diversity of available crop protection products. Furthermore, products are constantly being withdrawn because of environmental and health considerations.. In combination with these factors and the decline in the production of new crop protection products owing to timescale and financial constraints (Finney, 1988), fewer new products are being developed to replace those lost. The basis of most arable farming systems is inputs of non-renewable resources such as fertilisers, fossil fuels and agrochemicals. Experience in the lJSA has already shown that some high-input farming systems can result

(Zadoks, 1992).

Review of integrated arable farming systems research in W Europe 40 1

in decreasing soil productivity, declining environmental quality, reduced profitability and threats to human and animal health (Reganold, Papendick & Parr, 1990).

The main response in the EU to excessive grain production has been a reduction in production-coupled subsidies, downward adjustment of price-support mechanisms and the implementation of set-aside (Potts, 1991). Set-aside schemes aim to take 15% of arable land out of food production on an annual rotational basis or 18% in a non-rotational scheme. This, however, may prove to be economically and socially unacceptable and may lead to intensification (higher inputs of agrochemicals giving greater production per unit area). The European Parliament has recognised this and foresees set-aside as a transition to ecologically sound farming, i.e. extensification, which involves lower inputs of agrochemicals and hence lower production (Potts, 1991). Extensification may be achieved through the development of integrated farming practices, but the majority of research on integrated production in the E U until the late 1970’s was conducted on a relatively small scale and was limited to specific aspects of the crop production system. Jordan (1989) listed many of the studies carried out in the UK on integrated production. A crop environment is, however, a highly interactive system and needs to be investigated as such. For example, nitrogen inputs affect crop physiology and subsequently susceptibility to pests and diseases (Vereijken, 1989~); ultimately the interaction of these determines yield and quality. There- fore any investigation of farming systems must assess all aspects of crop production if it is to be fully understood. These conclusions were drawn at an International Organisation of Biocontrol/West Palaearctic Region Sector (IOBC/WPRS) meeting in 1976 with the result that a study group was set up in 1981 to examine the possibility of developing research programmes on ‘management of experimental farming systems for integrated crop protection’. This group compiled guidelines for the design of experiments on integrated farming systems (Vereijken, 1986) and an IOBC/WPRS working group for integrated arable farming was initiated with researchers from nine countries (Vereijken, 1989~). This led to a number of projects on integrated farming being established. Descriptions of these projects are available in Vereijken & Royle (1989). In 1993 this group produced ‘Guidelines for Integrated Crop Production’ (El Titi, Boller & Gendrier, 1993). The opportunities that integrated farming may offer led to the establishment of long-term experiments at Nagele (the Netherlands) and Lautenbach (Germany) in the late 1970’s. In the UK the Boxworth project was the first large-scale, long-term experiment in which different farming systems were examined. Results from these experiments are now available and a number of other projects are under way. Some of these appear to have similar aims and designs. Each of the major projects in the UK and a range of projects in continental Europe are reviewed here in terms of their aims, methodology and main results. The key references on which these appraisals are based are given in Table 1. Experimental farms have also been established in Ireland (Johnstone), Italy (Florence) and Sweden (Logirden and Alnarp) while others are planned in each of the other EC-member countries.

The Boxworth Project (1981-1988)

Background The 1970’s was a decade of increasing pesticide (fungicides, herbicides, insecticides,

molluscicides and nematicides) use in UK cereals (Sly, 1981) and during this time an overall decline in cereal invertebrates was observed in southern England (Aebischer, 1991). There was no direct proof that these trends were connected but the possibility of a link, together with an increase in the frequency of applications per season and the increasing use of pesticide mixtures, raised concern that intensive pesticide use in cereals could be environmentally damaging.

402 J M HOLLAND, G K FRAMPTON, T CILGI AND S D WRATTEN

Table 1. Key references dewrrhing the urrrhle farming systems research projects reviewed Project Reterences

Bowor th

SCAR A B

TALISMAN

RISC

LIFE

LINK IFS

LEAF Lautenbach

INTEX

Net herlands

Third Way France

Greig-Smith (1991) Greig-Smith & Hard! (1992) Greig-Smith er a / . (1991)

SCARAB Annual reports (Unpuhl.) Cooper ( 1990) Frampton & Cilgi (1992. 1993. 1993)

‘TALISMAN Annual reports (Unpubl.) Cooper (1990) Jordan. Hutcheon k Perks (1990) Perks 8: Lane (1990) RISC Annual reports (Unpubl.) Easson (1993) Jordan (19x9. 1990) Jordan er d. (1990. 1993) Jordan & Hutcheon (199h, h ) LINK IFS Annual reports (Unpubl.) Holland (1994) Prew (1993) Wall (1992) Abel & Hill ( 1993) El Titi (1986. 1989. 1990. 1991) El Titi & Landes (1990) Gerowitt & Steinmann (1992) Przemeck k Lickfett (1991) Schmidt 8: Waldhardt (1992) Stippich & Bhchner (1992) Wildenhayn (1991) Vereijken (1985. 1YX9a. b. 1990) Wijnands & Vereijken (1986. 1992) M’ijnands (1990. 1992) Hiini (1989, 1990. 1993) Viaux. Roturier & Bouchet (1993)

The Boxworth project was set up to investigate the effects of this intensive use of pesticides in cereals on a range of wildlife including plants, birds, small mammals and arthropods. The first large-scale study in the UK to involve a long-term comparison of different farming systems. it was funded by the MAFF and undertaken in collaboration with a number of research organisations. The Boxworth project was concerned primarily with the environmental consequences of pesticide inputs; the study of other aspects of integrated farming systems. such as disease-resistant varieties and variety mixtures, occupied only a small part of the project (Greig-Smith, Frampton & Hardy, 1992). The main aim was to investigate the long-term impact of three pesticide regimes on farmland wildlife. There were also two subsidiary aims: to make economic comparisons between the three regimes and to assess their management (Greig-Smith & Griffin, 1992).

Design and methods The project was sited at Boxworth Research Centre, a MAFF cereal farm in eastern

England. The farm was divided into three units, each a block of contiguous fields (see Table 2 for summary of experimental design). This was to allow realistic monitoring of the effects

Tab

le 2

. Sum

mar

y of

exp

erim

enta

l des

igns

for

the

arab

le fa

rmin

g sy

stem

s re

sear

ch p

roje

cts

revi

ewed

(se

e te

xt fo

r de

tails

)

CR

= c

onve

ntio

nal

rota

tion

, IR

= in

tegr

ated

rot

atio

n. L

R =

less

roo

t cr

ops,

MR

= m

ore

root

cro

ps.

Proj

ect

Box

wor

th

SCA

RA

B

TA

LIS

MA

N

RIS

C

LIF

E

LIN

K IF

S

LE

AF

IL

aute

nhac

h

INT

EX

The

Net

herl

andr

I

Nag

ele

2 B

oree

5wol

d

3 V

rede

peel

Thi

rd W

av

Fran

ce

Num

ber

ot s

ites

1 3 3 2 1 6 6 1 3 1 I 1 3 4

Reg

ime\

Full

insu

ranc

e Su

perv

ired

In

tegr

ated

C

urre

nt f

arm

pra

ctic

e R

educ

ed i

nnut

anp

roac

h (S

ee T

able

6a)

(S

ee T

able

60)

C

urre

nt f

arm

pra

ctic

e1C

R

Inte

grat

ed f

arm

ing

ryst

cm/C

H

Cur

rent

far

m p

ract

lce/

IR

Inte

prat

rd f

arm

ing

Syst

emil

R

Con

vent

iona

l far

min

g pr

actic

e In

tegr

ated

far

min

g sy

stem

In

tegr

ated

far

min

g sy

stem

C

urre

nt f

arm

ing

syst

em

Inte

grat

ed f

arm

ing

ryst

em

1. C

onve

ntio

nal/

high

inp

ut

11. I

nteg

rate

d 11

1. R

educ

ed

IV

Extc

nsib

e V

Fa

llow

Con

vent

iona

l/ara

hle

Inte

grat

edJa

rabl

e O

rgan

ic/m

ixed

C

onve

ntio

nal

Con

vent

iona

l/L

R

Intc

grat

ed/L

R

Con

vent

iona

l In

tcgr

atcd

In

tegr

ated

/LR

~n

regr

ated

/tvf

R

Inte

grat

ed

Con

vent

iona

l

Con

vent

iona

l In

tegr

ated

Exp

erim

enta

l des

ign

Four

con

tiguo

us f

ield

s Fo

ur c

ontig

uous

fie

lds

Thr

ee c

ontig

uous

fie

lds

Scvc

n pa

ired

hal

f-he

ld5

Six-

I2

rand

omis

ed b

lock

s pe

r si

te

12 r

ando

mis

ed b

lock

s pe

r si

te

Fib

e p

aire

d bl

ock.

;

Seve

n-ni

ne

pair

ed b

lock

s pe

r si

te

Who

le t

arm

non

-exp

erim

enta

l Si

x pa

ired

blo

ch\

Thr

ee w

hole

far

m c

ompa

riso

ns

Thr

ee w

hole

far

m c

ompa

riso

ns

Four

who

le f

arm

com

pari

sons

Com

pari

son

of c

onve

ntim

al a

nd

no p

estic

ide

area

s w

ith r

emai

nder

of

fie

ld a

nd c

ompa

riso

n of

in

tegr

ated

and

con

vent

iimal

tar

mr

ot s

imila

r st

ruct

ure

Six-

I0 p

aire

d bl

ock,

per

sit

e w

ith

rota

tiom

in

deep

and

*ha

llow

so

h

No.

ot

rcpl

icat

es

per

trea

tmcn

t/ya

r

1 1 I 1 3-4 4 5

Siz

e of

eac

h A

ddit

iona

l sm

all s

uh-p

lot

ckpe

rim

enta

l un

it va

lidat

ion

tria

ls

12-1

3.6

ha

11 5

-17.

3ha

5 7-

10 7

ha

4-16

ha

Ye,

No

18 x

20m

(m

in.)

10

x 4

0 or

10 x

20

m

100

x 1X

m

Yes

Y

e5

Yes

1-3

>2.

5ha

(>72

m w

ide)

I 4

orX

ha

I 1

2-3.

8 ha

2 ha

Z ha

2 ha

16. 2

0 an

d 10

ha

12 rn w

ide

rtri

pr

12 x

30 m

bloc

k

1-5

ha

Yes

No

No

No

NO

5.

4

P s

tpeq

uain

el

JV3

1

INT

EX

I I1

11

1 IV

V

Nag

ele

CFP

IF

S 0

%

Bor

gesw

old

CFP

C

FP/L

R

IFS/

LR

Vre

depe

el

CFP

IF

S IF

S/L

R

IFS/

MR

Thi

rd W

ay

CFP

C

FP =

IFS

= W

Ps

IFS

WPS

Fran

ce

CFP

IF

S

OSR

/WW

/WB

3-c

ours

e O

SR/W

W/F

B/W

B

.(-co

urse

OSR

/WW

/FB

/WB

4-c

ours

e Fa

llow

P/?/

SBe/

WW

4-c

ours

e P/

?/SB

e/W

W 4

-cou

rse

~/W

W/C

arro

t/3 y

r pa

stur

e

PJSB

e/P/

WW

P/

SW/P

eas/

Gra

ss s

eed/

P/FB

/SB

e/W

W

P/SW

/Pea

s/G

rass

see

d/P/

FB/S

Be/

WW

P/SB

e/W

W/S

/P/S

Be/

M/P

ea

or B

Pl

SBe/

WW

/S/P

/SB

e/M

/Pea

or

B

P/SB

e/W

WJS

/P/M

JPea

JGra

ss s

eed

P/SB

e/W

W/S

/P/S

Be/

Car

ot/P

ea

or B

Var

ies

with

far

m h

ut s

ame

for

each

reg

ime

OSR

/WW

/WB

3-c

ourr

e

Cer

eals

+ B

C m

inim

um 3

-cou

rse

Cer

eals

+ BC

min

imum

3-c

ours

e

Plou

gh

Red

uced

tilla

ge

Plou

gh

Red

uced

tilla

ge

Plou

gh

Red

uced

tilla

ge

Min

imal

Plou

gh

Plou

gh

Red

uced

tilla

ge

Plou

gh

Red

uced

tilla

ge

Red

uced

tilla

ge

Red

uced

tilla

ge

Plou

gh

Red

uced

tilla

ge

Red

uced

tilla

ge

Plou

gh

Shal

low

Stan

dard

L

ate

Stan

dard

L

ate

Stan

dard

L

ate

Lat

e

Stan

dard

St

anda

rd

Lat

e

Stan

dard

L

ate

Lat

e L

ate

Stan

dard

L

ate

Lat

e

Stan

dard

L

ate

Hig

h yi

eldi

ng

Mix

ed/re

sist

ant

Hig

h yi

eldi

ng

Mix

ed/re

sist

ant

Hig

h yi

eldi

ng

Res

ista

nt

Res

ista

nt

Hig

h yi

eldi

ng

Hig

h yi

eldi

ng

Res

ista

nt

Hig

h yi

eldi

ng

Res

ista

nt

Res

ista

nt

Res

ista

nt

Hig

h yi

eldi

ng

Res

ista

nt/m

ixed

R

esis

tant

/mix

ed

Hig

h yi

eldi

ng

Res

ista

nt

Stan

dard

70

% r

educ

tion

50%

red

uctio

n N

one

Stan

dard

R

educ

ed +

orga

nic

Org

anic

Stan

dard

St

anda

rd

Red

uced

+ or

gani

c

Stan

dard

R

educ

ed +

orga

nic

Red

uced

+ or

gani

c R

educ

ed +

orga

nic

Opt

imum

R

educ

ed

Red

uced

0 p t

i m u m

R

educ

ed

Stan

dard

R

educ

ed

No

inse

ctic

ide

Non

e

Stan

dard

R

educ

ed

Org

anic

onl

y

Stan

dard

St

anda

rd

Red

uced

Stan

dard

R

educ

ed

Red

uced

R

educ

ed

Stan

dard

A

s la

st r

esor

t N

one

Stan

dard

R

educ

ed

No

No

No

No

Yes

Y

es

Yes

Yes

Y

es

Yes

No

Yes

Y

es

Yes

No

Yes

Y

es

No

No

No

Yes

N

o Y

es

No

No

No

No

No

No

No

No

No

No

No

Yes

Y

es

No

No

4 5.

@a Q

FI =

full

insu

ranc

e re

gim

e, S

UP

= su

perv

ised

reg

ime,

IN

T =

inte

grat

ed r

egim

e, C

FP =

con

vent

iona

l fa

rm p

ract

ice,

RIA

= re

duce

d in

put a

ppro

ach,

CC

P =

syst

em, O

rg =

org

anic

, WPs

= w

ithou

t pes

ticid

es, L

R =

less

root

cro

ps, M

R =

mor

e ro

ot c

rops

, WW

= w

inte

r w

heat

, BC

= b

reak

cro

p, O

SR =

oils

eed

rape

, SB

= sp

ring

barl

ey, W

B =

win

ter

barle

y, S

AS

= se

t-asi

de,

FB =

fiel

d be

ans,

M =

mai

ze,

P =

pot

atoe

s, S

Be

= s

ugar

bee

t, S

= s

corz

oner

a, B

= d

war

f be

ans.

curr

ent c

omm

erci

al p

ract

ice

(-s =

stan

dard

rota

tion,

-a

= a

ltern

ativ

e ro

tatio

n), L

IA =

low

inpu

t app

roac

h, M

IN =

Min

imum

Inpu

ts, I

FS =

inte

grat

ed fa

rmin

g

5 2 P

0

wl

406 J M HOLLAND, G K FRAMPTON. T CILGI AND S D WRATTEN

of pesticide use on mobile animals such as birds and small mammals. Each unit received one of three pesticide regimes during the treatment phase of the project (1983-1988). These were: (1) a ‘full insurance’ programme which involved high inputs and prophylactic treatments, imitating an intensive cereal production system of the late 1970’s; (2) a ‘supervised’ programme whereby pesticides were applied only if weeds, diseases and pests exceeded economic thresholds; (3) an ‘integrated’ regime using economic thresholds and husbandry practices which further reduced the need for pesticides. In practice there was little difference between the supervised and integrated regimes in terms of their pesticide inputs. Effects of fertiliser inputs were not investigated. At the start of the project (1981- 1983) all three experimental units received a ‘supervised’ pesticide regime to allow “baseline” monitoring of wildlife to be undertaken before the contrasting pesticide inputs were implemented.

It was decided at the outset to keep the pesticide regimes constant throughout the experimental period, even though current farming practices may have changed during the programme. Continuous winter wheat was grown throughout the experimental period with the exception of an oilseed-rape break crop every five years. Further details of the experimental design and research programme are shown in Tables 2 to 4 respectively.

Principal results The detailed results of the Boxworth project were reviewed in Greig-Smith et al. (1992)

and are summarised below and in Table 5 . Some beneficial epigeal arthropods virtually disappeared from full insurance fields for

the full five-year treatment phase of the project while others were less adversely affected (Vickerman, 1992). These inter-specific differences in responses were attributed to differences in species’ exposure to pesticides as a result of differences in their ecology, life- cycles and dispersal abilities (Burn, 1992; Vickerman, 1992). Effects of the full insurance regime on soil fauna also varied, with some species adversely affected and others apparently favoured (Frampton, Langton, Greig-Smith & Hardy, 1992). Overall, densities of herbivores and carnivores (predators and parasitoids) were approximately 50% lower in the full insurance than in the supervised and integrated regime fields, although detritivores did not show an overall adverse effect as a result of the full insurance regime (Vickerman, 1992). Some pest species were not always adversely affected; in some years populations of the grain aphid Sitobion avenue (F.) and rose-grain aphid Metopolophiurn dirhodum (Wlk.) (Homoptera) were highest in full insurance fields, perhaps reflecting lower numbers of predators; Burn (1992) showed that the full insurance regime reduced the predatory impact of the invertebrate fauna on artificial pest baits (Diptera pupae) and on aphid populations in some years.

Of the other wildlife monitored at Boxworth, lethal effects of pesticide use were detected only in wood mice (Apodernus sylvaticus L.), which were killed by broadcast molluscicide pellets (Johnson, Flowerdew & Hare, 1992). Some birds were exposed to organophosphorus insecticides (Hart et al., 1992; Thompson, Tarrant & Hart, 1992) but there was no conclusive evidence of an overall effect of the pesticide regimes on the breeding performance of the most common bird species (Fletcher, Jones, Greig-Smith & Hardy, 1992), nor on populations of rabbits (Tarrant & Thompson, 1992) or wild plants (Marshall, 1992).

Monitoring crop performance showed that the full insurance regime usually gave the highest yields but the supervised regime was equally, if not more profitable, despite lower yields. The integrated regime gave the lowest yields and lowest gross margins and attempts to reduce herbicide inputs below those of the supervised regime caused problems with grass weeds (Jarvis, 1992). However, as the integrated regime at Boxworth was a compromise

Review of integrated arable farming systems research in W Europe 407

* *

x x x * * *

*. * * *.

* * x x x *

* *. * * * *

* * x * x *.

* * * * * *

x *. * #

* * * *

* * *

* * *

x * * * x

* * * * *

x x * x *

x * * * x

c * * * * *


2 3 QJ

+

I1

I

+ +

/I c.

0

/ I

+ + + i


(Greig-Smith & Hardy, 1992), its relatively poor returns do not give a general indication of the potential economic benefit of integrated farming.

The Boxworth project highlighted some of the problems that can arise when managing reduced-input systems. These included: (1) the difficulty in keeping long-term pesticide regimes realistic under controlled experimental conditions (Greig-Smith & Griffin, 1992), (2) the difficultly of investigating simultaneously economic, agronomic and environmental factors because of compromises in scale us replication (Greig-Smith & Griffin, 1992) and (3) a failure to reduce the incidence of cereal diseases in an integrated regime by using cultivar mixtures (Yarham & Symonds, 1992).

Other information The environmental and economic studies within the Boxworth project were valuable

components but the restrictions imposed on economic studies by the demands of the environmental work suggested that subsequent studies would be better divided into those driven primarily by environmental concerns and those dictated by economic priorities. As a consequence, different experimental approaches are now being used in two follow-up studies to Boxworth, the environrhentally-driven SCARAB project and the economically- driven TALISMAN project (Cooper, 1990), which are described below.

The Boxworth project was not extended further in its original format on completion in 1988, mainly because the full insurance programme was unrepresentative of the then farming practice. Instead, it was continued in a limited form for a further three years to assess the potential for recovery of arthropod populations once the high-input full insurance regime was replaced by a lower-input supervised programme. This work showed that ground beetles, spiders and springtails recovered from the full insurance regime very slowly, indicating that effects of intensive pesticide use in cereals can last for years even after the intensity of pesticide use is reduced; recent European work has also shown persistent effects of pesticide use in other high-pesticide-input crops (e .g. hop gardens and vineyards; Filser, Fromm, Nagel & Winter, 1994).

SCARAB (Seeking Confirmation About Results At Boxworth) (1989-1996)

Background SCARAB was set up to investigate whether the results obtained from the Boxworth

project could occur more widely, at other locations, with different soil types, in a wider variety of arable crops and with typical 1990’s pesticide inputs. The project is one of two follow-up studies to the Boxworth project; SCARAB is primarily an environmental study with subsidiary economic and agronomic monitoring while TALISMAN (described below) is primarily concerned with economic and agronomic aspects of reducing pesticide inputs, with environmental monitoring having a subsidiary role. These two projects and RISC (also described below) are designed to complement one another. SCARAB is funded by MAFF and is being undertaken in collaboration with a number of research organisations. Its primary aim is to compare the long-term effects of two pesticide regimes; these are conventional and reduced pesticide input systems. SCARAB is concerned only with pesticide use; fertiliser inputs and other components of integrated farming systems are examined in some of the other studies described below.

Design and methods The two pesticide regimes are a ‘current farm practice’ (CFP) and ‘reduced input

approach’ (RIA). CFP is based on the current pesticide use for the location (based on


survey reports) as carried out by a technically competent, financially aware farmer and will accommodate any year-to-year changes in agricultural practice. RIA is based on the minimal use of fungicides and herbicides determined by crop monitoring and the use of thresholds. Insecticides are avoided completely unless major crop losses are anticipated. SCARAB has seven fields. ranging from 8 to 34 ha. located at three Agricultural Development and Advisory Service (ADAS) Research Centres in England: Drayton (Warwickshire), Gleadthorpe (Nottinghamshire) and High Mowthorpe (North Yorkshire) (Cooper, 1990). Cropping at each location follows a six- or seven-year rotation which is typical of the region. At Drayton Research Centre the rotation is two wheats and five years’ grass ley whilst at Gleadthorpe and High Mowthorpe Research Centres a six-course rotation is followed consisting of cereals and break crops (potatoes. spring beans and sugar beet (Gleadthorpe); spring beans and winter oilseed rape (High Mowthorpe)). In the first, “baseline”, year of the project all the fields received a CFP regime appropriate for the crop to allow invertebrates to be monitored in the absence of contrasting treatments. Since the autumn of 1990 one half of each field has received an appropriate RIA regime whilst the other half has continued to receive CFP pesticide inputs. Cropping. fertilisers and cultivations do not differ between the CFP and RIA halves. At each Research Centre the appropriate rotation is phased by starting it at a different crop in each field; this is to make information on all crops available sooiier than the full six years that would be needed if all fields entered the rotation at the same point. However, unlike the economically-orientated TALISMAN and RISC projects which also have phased rotations (see below). the SCARAB CFP and RIA regimes are not strictly replicated as the two or three split fields per site are at different rotational phases and subsequently each tield receives different pesticides during a growing season. Justification for this type of design is based on the premise that long-term monitoring should. as in the Boxworth project (Greig-Smith et 01.. 1992). detect effects of the regimes on invertebrates.

The emphasis of the research in the SCARAB project (Table 4) is on monitoring the effects of the pesticide regimes o n invertebrates and soil micro-biology. Populations of arthropods have been monitored since the start of the project by entomologists at Sou- tharnpton liniversity using pitfall traps and suction samples (Frampton & Cilgi. 1992, 1993, 1994) and MAFF researchers haare been monitoring earthworm populations. Soil microbiology and chemical composition are routinely analysed in selected fields by micro- biologists at University of Wales. Bangor (Jones. Jones & Johnson, 1993). The distribution and abundance of pests, weeds and diseases are examined by researchers at each of the Research Centres to monitor them in relation to control thresholds; crop yield and quality are recorded for the subsidiary economic appraisal.

Principal resiilts Preliminary results indicate that current (CFP) pesticide use in U K arable crops is

environmentally more damaging than using reduced inputs (RIA). Although the project was designed to study the overall long-term effects of the CFP and RIA regimes rather than those of specific pesticide applications. some individual broad-spectrum insecticides have caused obvious reductions in catches of some Carabidae (ground beetles), Staphylinidae (rove beetles). Linyphiidae (money spiders) and Collembola (springtails). Temporary elimination of several species of Carabidae and Collembola occurred, particularly after applications of chlorpyrifos and dimethoate (Cilgi, Wratten. Frampton & Holland, 1993; Frampton &L Cilgi, 1994): populations of some species of Collembola did not recover within 18 months o f a pesticide application and these results show that they could be valuable indicators of non-target pesticide effects (Frampton, 1994). The severity of pesticide effects has varied between years. species. crops and locations and it would be premature to draw


any firm conclusions until the full range of crop-site-species-pesticide combinations has occurred. However, already it is clear that persistent adverse effects of pesticides, like those seen at Boxworth, can occur with conventional 1990’s pesticide inputs and that adverse effects of conventional pesticide use may occur in other arable crops besides cereals.

Overall, soil biomass levels were similar under the two regimes, but some transient effects of soil type and short-term pesticide use were found. At High Mowthorpe Research Centre, applications of herbicides (bentazone and glyphosate) and fungicides (metalaxyl and chlorothalonil) produced a temporary stimulation of soil biomass, whereas the insecticide pirimicarb significantly reduced total and fungal biomass. At Gleadthorpe Research Centre the sandy loam soil produced more variable results and individual pesticide effects could not be ascertained (Jones et al., 1993).

Weed monitoring up to 1992 indicated that, compared with the CFP herbicide use, the efficiency of the half-rate RIA herbicide treatments varied from very effective to inadequate. Weed control tended to be poorer in broad-leaved crops (oilseed rape and beans) than in cereals (Ogilvy, Green, Groves & Jones, 1993). Although SCARAB is not concerned primarily with low inputs of herbicides, monitoring weeds could be a valuable aid to interpretation of the results of invertebrate monitoring as the presence of weeds may affect the activity and distribution of non-target arthropods (Speight & Lawton, 1976).

In the first two years after implementing CFP, yields exceeded those of RIA for the majority of crops although differences were sometimes small. Where variable inputs were lower for the RIA compared to the CFP regime, higher gross margins were achieved.

To draw firm conclusions from these results would be premature as they represent only two years’ monitoring. However, it is clear that whilst CFP regimes often gave the best financial returns, the RIA treatment can economically out-perform CFP under certain cropping situations.

Other information

and Frampton & Cilgi (1993). The designs of the Boxworth and SCARAB projects are compared in Cilgi et al. (1993)

T A L I S M A N (1 989-1 996) (Towards A L o w Input System Minimising Agrochemicals and Nitrogen)

Background TALISMAN is the second long-term MAFF study that was set up to extend the infor-

mation obtained from the Boxworth project to a wider range of crops, soils, locations and to typical current agrochemical inputs. Its main objective is to measure the economic and agronomic effects of implementing cropping systems with lower agrochemical and nitrogen inputs than conventional farming systems. It also aims to provide information on the scale of compensation necessary to attract farmers to adopt such lower-input systems. Environmental monitoring forms a subsidiary part of the project. In these respects TAL- ISMAN is very similar to, and complements, the RISC project which is sited in Northern Ireland and is described in detail below.

Design and methods The experiment was initially located at four ADAS Research Centres in England.

Three of these are shared with the SCARAB project and the fourth is at Boxworth in Cambridgeshire. After harvest 1991, the Gleadthorpe TALISMAN site was discontinued

412 J M HOLLAND. G K FRAMPTON, T CILGI AND S D WRATTEN

due to unpredictable variation in soil but preliminary results from that site have been published (e.g. Clarke er al.. 1993).

At all TALISMAN sites there are two farming systems: ‘Current Commercial Practice” (CCP) and “Low Input Approach“ (LIA); and two six-course rotations: “standard” and “alternative” (Table 6a). The standard rotation consists of winter sown cereals and break crops whilst the alternative rotation comprises mainly spring-sown cereals and break crops which have a lower requirement for nitrogen and pesticides (Clarke et al . , 1993). The CCP system represents typical current nitrogen and pesticide inputs and uses the ADAS “Fertiplan” fertiliser planning service (Goodlass, 1991) to determine nitrogen use. The LIA is a low nitrogen and pesticide input system in which nitrogen is applied at 50% of the CCP rate. CCP pesticide use is based on current ADAS thresholds in conjunction with field monitoring, with widely-used products applied at full recommended rates. The priority in the LIA is to not apply pesticides unless a yield loss of more than 10% is expected, in which case up to 50% of the recommended rate may be applied. An additional farming system, “Integrated Low Input Approach” (ILIA). in which cultural measures such as different cultivations. sowing dates and cover crops are used to reduce the impact of lower nitrogen and pesticide use. is also employed at one of the TALISMAN sites (Table 6a) .

At each site the rotations were phased by starting them respectively in the first (phase 1) and fourth (phase 2) crops of the sequence to take some account of the effects of seasonality and provide information on all crops after three rather than six years. Altogether there are 10 treatments comprising different combinations of the farming systems, the rotations and their phases (Table 6a). Six of the treatments are common to all four TALISMAN sites and form the main experimental comparisons; the additional four treatments are limited to one or two of the TALISMAN sites but are complemented by very similar treatments at two sites in Northern Ireland as a component of the RISC project (Table 6a) .

The economic and. secondarily. environmental effects of the TALISMAN treatments are being compared using replicated plots of a minimum size of 18 m x 20 m, with rotation and farming system as the main treatments. Interactions between the CCP and the LIA rates of herbicides, fungicides and insecticides are examined in addition to the main treatments by using sub-treatments (Table 6 b ) by dividing each main plot into five equal sub-plots. At each site the plots are arranged within a field in a randomised block design with 12 m-wide mown grass strips separating the plots; there are three replicates of each treatment at three of the sites and four at Boxworth. This allows a rigorous statistical appraisal of the economic consequences of reducing pesticide and nitrogen inputs and has the advantage of examining these regimes in homogenous blocks of land duplicated in localities with different soil types.

Agronomic monitoring (Table 4) in the sub-plots includes assessments of the incidence of weeds, pests and diseases in conjunction with crop morphology, yield, quality and economic gross margins. Plant seedbanks and soil nematode populations have also been monitored in selected plots since the spring of 1991 by researchers at the Scottish Crop Research Institute. An additional block of plots adjacent to a field boundary is used at each site solely for assessing effects of the main treatments on arthropods. These plots, which are not sub-divided but are otherwise identical to others at the site, are sampled routinely using pitfall traps and suction samplers.

Principal results In 25 comparisons between the CCP and the LIA made up to 1992, the LIA gave

measurable yield losses on 12 occasions but only four of these were greater than 5%. Gross margins were lower in the LIA than the CCP on seven occasions, but differences were mostly very small (Clarke et al.. 1993).

High weed infestations in some LIA crops illustrated the potential longer-term problems

? 2 %

S'

ip" 2

I t

r

\ fi

Tre

atm

ent

Rot

atio

n Ph

ase

Syst

em/R

egim

e B

oxw

orth

G

lead

thor

pe

Dra

yton

H

ighM

owth

orpe

H

illsb

orou

gh

Gre

enm

ount

L

1 St

anda

rd

1 C

FP/C

CP

Yes

Y

es

Yes

Y

es

Yes

Y

es

2 2

Stan

dard

1

LIA

Y

es

Yes

Y

es

Yes

Y

es

Yes

s 3

3 St

anda

rd

2 C

FP/C

CP

Ye5

Y

es

Yes

Y

es

Yes

Y

es

3 4

Stan

dard

2

LIA

Y

es

Yes

Y

es

Yes

Y

es

Yes

S'

5 A

ltern

ativ

e 1

CFP

/CC

P Y

es

Yes

Y

es

Yes

Y

es

Yes

6

Alte

rnat

ive

1 L

IA

Yes

Y

es

Yes

Y

es

Yes

Y

es

00 2

7 A

ltern

ativ

e 2

CFP

/CC

P Y

es

Yes

Y

es

Yes

ip"

8 A

ltern

ativ

e 2

LIA

Y

es

Yes

Y

es

Yes

2 9

Stan

dard

2

INT

IIL

IA

Yes

Y

es

10

Alte

rnat

ive

2 IN

TII

LIA

Y

es

Yes

11

St

anda

rd

2 M

IN

Yes

Y

es

i! 12

A

ltern

ativ

e 2

MIN

Y

es

Yes

n 3

CFP

= c

onve

ntio

nal f

arm

pra

ctic

e, C

CP

= c

urre

nt c

omm

erci

al p

ract

ice,

LIA

= lo

w in

put a

ppro

ach,

IL

IA &

IN

T =

Inte

grat

ed L

ow I

nput

App

roac

h. M

IN =

M

inim

um I

nput

s.

S. 3 '", s.

Tab

le 6

(a).

Sum

mar

y of

tr

eatm

ents

in T

AL

ISM

AN

and

RIS

C p

roje

cts

00

A

A

RIS

C s

ites

TA

LIS

MA

N s

ites

n co s

h 8 P

w

CL

414 J M HOLLAND. G K FRAMPTON. T CILGl AND S D WRATTEN

Table 6( b ) Sub-plot treatnients u e d in TALISMAN aiid RISC \ l lh plot Fungicide t-ierhiiide I nwct icide

I CFP C C P CFP CCP ( F P CC‘P - I 1A LIA 1 IA 3 1 I A CFP C C P (FP’CCP 4 CFP ( C P LIA CFP’CCP 5 CFP C C P CFP CCP LIA

7

that can arise if herbicide inputs are reduced. Monitoring weed seeds revealed that plots at Boxworth that had previously been sown with oilseed rape had higher densities of Brassica seeds in the soil. whilst plots previously under linseed had higher densities of Poa seeds. The long-term implications of these results are not known but they demonstrate that an accumulation of weed populations in the seed bank may be difficult to avoid unless the correct combination of cropping and herbicides is achieved. So far, no other effects of the treatments on weed seeds have been detected.

Preliminary results for the range of pesticides tested in TALISMAN suggest that reducing the rates of many pesticides can give significant savings (Bowerman, 1993), although this is likely to be achieved less easily in broad-leaved break crops than in cereals.

Preliminary results of arthropod monitoring have shown effects of individual pesticide applications on some Carabidae. Arachnida and Collembola but it is difficult to judge the importance of these because catches have also shown considerable variation in the absence of pesticide applications (Hancock et a / . . 1994).

Densities of nematodes at two of the sites were greater where oilseed rape had been sown than where linseed or spring beans were grown. perhaps reflecting an effect of continuous over-winter vegetation cover. Some apparently detrimental effects of full-rate applications of nitrogen and herbicides on predatory nematodes were observed but require verification through continued monitoring.

Other information As with all small-plot studies TALISMAN has a number of potential limitations; the

presence of many crop varieties within a small area could influence the development of disease and pest infestations and could limit the predictive powers of the results, particularly with regard to insect pests and diseases. Some of these potential limitations may be overcome by the “phasing” of the rotations (Cooper, 1990), which gives some temporal as well as spatial replication.

The environmental monitoring (Table 4) is limited to an assessment of invertebrate populations by pitfall sampling and suction sampling in the block of plots adjacent to the field boundary at each site. This should provide an indication of the short-term effects on the more vulnerable species, and provide feedback on those pesticides which might warrant more study in the detailed invertebrate monitoring of the SCARAB project. TALISMAN aims specifically to examine the effects of reducing pesticide and nitrogen inputs and includes the impact of cultural measures only at High Mowthorpe. Other components of an integrated farming systems (IFS) approach are not included, such as the careful management of field margins. which can enhance natural control (El Titi. 1991; Sotherton. 1991).

RISC (Reduced Input Systems of Cropping) (1991-2000)

Buckground The RISC project was established and is funded by the Department of Agriculture for


Northern Ireland (DANI) to provide information for Government policy divisions on the scale of compensation necessary to attract farmers to change from current farming systems to those with reduced pesticide and nitrogen use, which are thought to be environmentally more acceptable. RISC will also assess the impact of such changes in husbandry practices on crop yields, farm management and environmental indicators. The aim of the project, like that of TALISMAN (see above), is to measure the economic and agronomic consequences of adopting cropping systems which use lower levels of agrochemicals and nitrogen than conventional cropping systems. RISC is sited in Northern Ireland and was designed to be complementary to TALISMAN in England; the designs of the two projects are similar, with closely matching protocols, but they take account of the different cropping practices and pesticide use in Northern Ireland. RISC, like TALISMAN, concentrates on the economic effects of low-input, environmentally benign arable systems, with environmental monitoring of invertebrates of secondary importance.

Design and methods RISC is sited at two farms and started in 1991 at Hillsborough, County Down and in 1992

at Greenmount, County Antrim. The project is planned to continue until 2000. Two six- course rotations are used: a standard rotation (potatoes, winter wheat, winter barley, oilseed rape, winter wheat, spring barley) and an alternative rotation comprising crops with inherently lower demands for pesticides and nitrogen (two-year grass ley, potatoes, winter wheat, spring barley, winter barley). The cow slurry produced as a by-product of the grass ley enterprise will be used pre-sowing to fertilise the other crops in the alternative rotation. As in TALISMAN, the rotations are phased, by starting them in both the first and fourth crops of the sequence, so that information is available for all crops after three rather than six years.

There are four regimes (Table 6a) of pesticide and nitrogen use in the RISC project: (1) Current Farming Practice (CFP) which aims to optimise inputs and maximise productivity using recommended top-yielding varieties. Pesticide use is based on ADAS and DANI surveys and fertiliser use is determined by DANI recommendations, with slurry applied to the alternative rotation; (2) Low Input Approach (LIA) is based on restricted nitrogen and pesticide use and aims to use a maximum of 50% of the overall CFP inputs applied each season (reductions of 50% are not likely to be achieved for every chemical). Pesticides will be omitted unless a yield reduction of more than 10% is anticipated; (3) Integrated Low Input Approach (ILIA) has pesticide and nitrogen inputs no greater than those of LIA but will incorporate other husbandry practices (such as drilling dates, choice of disease-resistant varieties and use of cover crops) in an attempt to reduce the economic and environmental impact of the LIA regime; (4) Minimum Inputs (MIN) is an integrated regime based on the use of slurry or farmyard manure to fertilise crops in the alternative rotation but no fertilisers will be used in the standard rotation except to prevent crop failure. Insecticides will be avoided, and fungicides and herbicides used only as a last resort to secure a viable crop. The policy for the MIN regime may require revision as the project develops. Combinations of the two rotations and four input regimes in the RISC project give a total of 12 treatments, some of which are shared with TALISMAN (Table 6a).

The RISC treatments at each site are arranged in a randomised block design with four blocks of plots within a field to give four replicates of each treatment. Plots in one of the blocks near a mature field boundary are large (10m X 40 m) to allow for invertebrate monitoring using pitfall traps; the remaining three blocks contain plots measuring 10 m X 20 m. In addition to the main treatments, the plots in the latter three blocks are each divided into five sub-plots (lOm x 4 m) to examine interactions between the CFP and the LIA fungicide, herbicide and insecticide inputs (Table 6b).


Priticipal results As only two years of the project have elapsed it is too early to put a strong emphasis on

the results of RISC. especially as the study has been planned to continue for a further eight years. Some time will be required to establish the full significance of the preliminary results summarised here.

The CFP regime gave the lowest gross margins in spring barley, winter barley and one of two oilseed rape crops. but provided both the highest yield and gross margin in potatoes, perhaps reflecting the response of a nitrogen-sensitive crop to the CFP and the LIA differences in fertiliser inputs. In the oilseed rape crop at Hillsborough the low variable costs of MIN helped this treatment to produce the highest gross margin, despite the yield being the lowest of all treatments. Oilseed rape at the Greenmount site exemplified one of the management problems that may occur in a low-input rotation; a serious wild-oat problem led to the decision to forage-harvest the crop to prevent oats seeding and affecting the following crop. Overall, the initial results suggest that a conventional (CFP) farming approach may not always be necessary to achieve the best financial returns, although too few crops have yet been studied in RISC to allow firm conclusions to be drawn.

Some initial differences between treatments in catches of Carabidae have emerged. For example, in spring barley plots Bembidion tetracolum say showed consistently higher LIA than CFP catches whereas the reverse pattern occurred for Nebria breuicollis (F.), with higher CFP catches. In oilseed rape Pterostichus melanarius (Ill.) was considerably more numerous in traps in the MIN than in the INT treatment. However, as in TALISMAN, the RISC environmental monitoring plots are not replicated so further monitoring will be needed to establish whether these differences in catches represent true effects of the different treatment regimes, natural population fluctuations, or variations in activity.

Other information RISC is similar in its aim and design to TALISMAN and these projects will complement

each other in their economic and environmental monitoring programmes (see Table 4). There are close links between researchers in the economically orientated RISC, TALISMAN and LINK IFS (see below) projects and the environmentally orientated SCARAB project. Together. these projects should provide a considerable amount of information on the economic and environmental consequences of adopting reduced-input farming systems.

Further information on RISC is given in Easson (1993) and in the RISC Report for 1991 and 1992 Seasons (Agricultural Research Institute of Northern Ireland, Hillsborough, Co. Down; unpublished).

There are many other farming practices besides pesticide and nitrogen use which may have a profound ecological impact and these can be manipulated to produce a more intricate farming system. The ecological and economic effects of these are being explored in the LIFE and LINK Integrated Farming Systems Projects which are described below.

LIFE (Less-Intensive Farming atid Environmental Research) (1989-1 999)

Background This project, funded by MAFF, was the first to develop and investigate a fully integrated

farming system in the UK. The objective is to provide fundamental information on the effects, interactions and ecological implications of an integrated farming systems approach for cereal growing in short rotations, and to develop and evaluate systems of less-intensive production which are both economically and ecologically sound and sustainable in the long-


term. The aims are specifically to: 1. Decrease the cost and improve the environmental safety of arable farming in the UK. 2. To develop integrated control strategies which are compatible with environmental protection (Jordan, 1990).

Design and methods The experiment was begun in 1989 at Long Ashton near Bristol, South-west England and

occupies five large fields totalling 23 ha. Four farming approaches are compared within each of the fields, so the comparisons are replicated five times. The approaches are based on the combinations of two rotations and two farming systems: a conventional and an integrated rotation, each with a standard farm practice system and a less-intensive system which is managed within the Guidelines for Integrated Crop Production (El Titi et al., 1993). Smaller experimental areas within the main treatments are used to examine the yield response to selected components of the system to identify the most influential components. Five-course fully phased rotations of cereals and break crops have been devised for the conventional and integrated rotations.

A range of crop husbandry practices and strategies (Table 3) has been employed to reduce pesticide inputs in the less-intensive systems and these will be flexible over the experimental period. For weed control, herbicides are chosen according to the weed spectrum but must have low environmental hazard, low total active ingredient and low cost. Mechanical weeding combined with reduced doses is used where possible. Disease control is based upon the combination of appropriate rotation, sowing date, resistant cultivars and lower applied nitrogen which reduces the disease incidence and severity facilitating the exploitation of a single-fungicide spray strategy. The pest control strategy relies upon either predictive models (Kendall, Brain & Chinn, 1992) or on site-specific damage thresholds (Glen et al., 1993). All the fields are surrounded by hedgerows with herbaceous vegetation and more recently these have been enhanced by the addition of a 2 m-wide grass and wild flower mixture strip between the crop and field boundary (Greaves & Marshall, 1987).

In each field unit, the incidence of pests, diseases and weeds is monitored, and soil physical, chemical and biological parameters are measured. Collaborative studies are in progress examining soil organic matter content, soil hydrology, run-off, erosion and nitrate leaching. Studies on fertiliser and pesticide movement in the soil are planned. Crop yields, quality parameters and husbandry records provide a detailed economic analysis of each system. Selected species of Carabidae, Staphylinidae and Arachnida are monitored with pitfall traps, and Lumbricidae (earthworm) biomass is also measured.

Principal results The first three years of this study revealed that the less-intensive system using the

conventional rotation was cost effective, providing a 16% higher average profit compared with the standard farm practice system using the conventional rotation even though lower yields were produced (Jordan, Hutcheon & Glen, 1993). This increased profit was achieved through reductions in applied nitrogen (29-31%), herbicides (15-19%), fungicides (>80%) and plant growth regulators (100%). Overall pesticide use was reduced in the integrated system through fewer sprays rather than reduced dose rates. The less-intensive system using the integrated rotation increased profits by only 3% compared with the conventional system. This was achieved through reductions similar to those described above with the addition of a 90% reduction in insecticides.

A marked improvement in catches of some beneficial invertebrates was achieved in the lower input areas. Greater numbers of Carabidae, Staphylinidae, Linyphiidae and Diplo-

41 8 J M HOLLAND. G K FRAMPTON. T C‘ILGI AND S D WRAITEN

poda (millipedes) were caught. Nehria hrevicollis (Carabidae) was highly significantly more abundant in the lower input areas in contrast to RISC where it was favoured in the CFP areas. There were no significant effects of rotation. chemical inputs, fields or crops on overall invertebrate numbers. Lumbricidae biomass was 16-20% greater in the lower input areas of both rotations. No significant effects on microflora populations were detected, however. although populations varied between fields; associations between crops and systems in cellulose degradation rates were apparent. Improvements in soil structure, organic matter content. residual soil nitrogen and nitrate leaching have been found.

Other information Results from the first rotational cycle will be available at the end of 1994, when a complete

analysis of the systems will be made. The LIFE project has now received further funding from the Commission of the European

Communities (CEC) and in conjunction with collaborators in France and Germany will put the research studies into practice on a wider scale. In the autumn of 1992 two pilot farm schemes were initiated. one at Trerulefoot in Cornwall, the other at Cirencester in Gloucester5hire (both in south-west England), in which two commercial farms were converted to the less-intensive system devised in LIFE. A six-course rotation of cereals, break crops and set-aside is planned in conjunction with the use of disease resistant varieties, minimal tillage, later sowing dates and reduced nitrogen fertilisers. Closer monitoring of weeds. pests and diseases combined with manipulation of timing and application rates will aim to reduce agrochemical inputs (Jordan et a l . , 1993).

LINK Integrated Farming Systems ( IFS) (1992-1 997)

Background This project is part of a larger programme of research on ‘Technologies for Sustainable

Farming Systems’ aiming to investigate new techniques and farm management practices which will ultimately contribute to new production systems (Parker, 1992). The LINK title arose because these projects aim to encourage collaborative research between industry and the research base (academia. research institutes and organisations) (Wall, 1992) with funding jointly by the UK Government and agrochemical industry. These production systems must be profitable, meet environmental requirements. ensure greater conservation of resources and meet consumer requirements (Parker, 1992). The LINK IFS project is the largest within this programme of research and was set up to compare on a farm scale conventional and integrated farming systems in terms of economics. and agronomic and environmental impact. The project is funded by MAFF, Scottish Office Agriculture and Fisheries Devel- opment, Home Grown Cereals Authority. Zeneca Agrochemicals and the British Agro- chemical Association. The organisations responsible for the six sites are: ADAS Research Centres (Boxworth, High Mowthorpe and Rosemaund); Institute of Arable Crops Research (IACR) Rothamsted and William Scott Abbott Trust; The Game Conservancy Trust and Southampton University: and the Scottish Agricultural College.

Design and methods Two farming systems are compared: a conventional farm practice (CFP) which represents

the agricultural practice for that area. as carried out by a technically aware, cost-conscious, risk-averse farmer. The second. an integrated farming system (IFS) which is based on a cost conscious. environmentally concerned farmer who seeks to use non-agrochemical methods of management as his starting point with agrochemicals used as needed to achieve


a profitable commercial crop (Prew, 1993). Six sites were chosen to represent a range of climatic zones, soil types and agronomic practices in the main arable production areas of the UK. The project will run for five years through a five-course rotation of cereals, break crops and set-aside. There are a minimum of seven pairs of plots per site ensuring that each course of the rotation is present every year and at least two courses of the rotation are replicated each year. Each plot has a minimum size of 2.5 ha and width of 70 m although this is usually exceeded. On completion of the project there will be a comprehensive statistical evaluation using data within and between all sites for the whole experimental period. The experimental design follows closely that of LIFE but extends the number of sites and plot size.

The baseline data were collected in 1992 before the split-field treatments were established in the autumn of that year. An extensive range of crop husbandry practices will be implemented in the IFS area to reduce the need for fertilisers and agrochemicals. These include later sowing dates, disease resistant varieties, reduced tillage, cover crops, mechanical weeding and lower inputs of fertilisers and pesticides (Table 3). In addition, the sites will use one or more of the following: Conservation Headlands (Sotherton, Boatman & Rands, 1989), ‘beetle banks’ (Thomas, Wratten & Sotherton, 1991) and Phacelia tanacetifolia Benth. flower borders (Hickman & Wratten, 1994) or wildflower borders to enhance beneficial insect populations. The farming practice will remain flexible for both systems over the duration of the experiment and will incorporate any new developments as they arise. Difficulties are foreseen in building a distinction between the systems because economics are currently driving farming into a reduced-input system. A management group will ensure that a distinction between the systems remains and that there is consistency between sites.

At present the emphasis of the research is on agronomic and economic factors (Table 4). This involves the detailed monitoring of weeds, the seed bank, diseases, pests, crop morphology and soil minerals. The impact of specific crop protection practices is evaluated at four sites using a series of small plot experiments in which only one variable is altered. Yield and quality, combined with measurements of all chemical inputs and cultivation costs, will be used to compare the economics of each system. The environmental monitoring concentrates only on beneficial invertebrates and earthworm populations. If, however, new resources become available these studies may be extended (Holland, 1994). No results have been published to date but Annual Reports for the baseline year and first treatment year are available.

LEAF (Linking Environment And Farming) This non-experimental project was launched in 1991 to promote integrated crop man-

agement (ICM) which combines environmentally-friendly farming whilst maintaining an economically viable production system. This is to be achieved by careful selection of rotations, cultivations, varieties, fertilisers and agrochemical inputs. Field margins and non- crop habitats will also be managed to enhance populations of beneficial insects and wildlife. The project has been funded by the European crop protection industry for three years. By January 1993, 14 LEAF demonstration farms had been established throughout the UK, with the prospect of more farms being established at a later date. These farms will be used to demonstrate to visiting groups the techniques and economics of ICM. Each farm was selected according to strict guidelines by an advisory board comprising specialists from agricultural production and research, the food industry, conservation and consumer groups. An executive committee with extensive knowledge of agricultural production provides a series of crop guidelines and principles for each farm to follow.

420 J M HOLLAND. G K FRAMPTON, T CILGl AND S D WRATTEN

N o evaluative research is planned and it remains up to each farmer to evaluate the economics of the ICM system in terms of the farm's profitability, in comparison with that of other farmers and their previous results.

Luicterzhach, German.v (1978-1989)

Background German agriculture in the 1970's was undergoing changes similar to those in the UK with

more intensive use of fertilisers and pesticides and large-scale monoculture (El Titi, 1989, 1991). Concern over the impact on wildlife of such farming practices, declining farm incomes and the limited success in introducing integrated pest management in European agriculture (El Titi, 1990; Vereijken. 1986) led to the initiation of the Lautenbach project. The specific objectives of the study were to examine whether: a. An integrated farming system could maintain pest populations below economically

b. Use of off-farm inputs (pesticides, fertilisers and fuels) could be reduced and optimised. c. Integrated farming is sustainable. d. Integrated farming would enhance the environment by decreasing pollution and increas-

e. The effects of landscape components on crop productivity can be evaluated in the long

damaging levels.

ing the diversity of beneficial species.

term (El Titi. 1989, 1990. 1991).

Design and methods Two farming systems were compared at Lautenbach; the Current Farming System (CFS)

us Integrated Farming System (IFS). The conventional system was typical of the practice for the surrounding region. In the integrated system inputs were reduced through changes in husbandry practices whilst aiming to maintain profits. Six pairs of plots (each 4 or 8 ha) in six fields, totalling 36 ha. were maintained in the two systems; one plot of each pair was in a different system and a sub-plot of one ha split between the two systems was used as a monitoring area. Starting in 1978. each system followed a six-course rotation of 60% cereals, 25% sugar beet and 15% legumes. The choice of variety and rotation was limited because the farm was producing commercially viable seed crops.

A range of crop husbandry practices including non-inversion tillage, reduced nitrogen, mechanical weeding, different row spacings and managed field margins were used to reduce inputs in the IFS system (Table 3). The IFS crop husbandry practices continued to evolve over the 12-year experimental period as a result of input from farmers and researchers.

A comprehensive range of detailed economic, agronomic and environmental investigations was carried out (Table 4). In addition, the effect of planting hedgerows on crop pests, beneficial insects and crop yields was assessed.

Principal results The Lautenbach project showed that IFS were economically viable in comparison with

the conventional practice; the gross margin for wheat crops and all crops considered together were consistently higher for the IFS despite an average yield reduction of 7% over the 12 years. This was achieved through overall reductions in nitrogen fertilisers (18%) and pesticides (37%) for all crops. Incidence of stem-base diseases and powdery mildew (Ery.szphe gruminis DC.) were lower in the IFS plots, although leaf blotch incidence (Seproria nodoriim Berk.) was 5-12% higher (El Titi, 1991). Populations of annual and perennial weed seeds increased in the IFS plots: however, these increases were not reflected in the


cost of weed control over the 12 years (El Titi, 1989). Many of the environmental indicators showed increases over the duration of the experiment. For example, there was a six-fold increase in populations of Lumbricidae, and more Collembola, gamasid mites (Gasmasina: Mesostigmata), Carabidae and Staphylinidae were recorded in the IFS than in the CFS plots (El Titi & Ipach, 1989). Soil humus content and soil workability were also improved in these areas (El Titi, 1990).

So far, the Lautenbach project has identified the combination of minimal tillage and incorporation of crop residues as an important component of any IFS system.

Other information One of the aims of the Lautenbach project was to demonstrate integrated farming and it

has achieved much success in disseminating this information to the farming community. Some of this success can be attributed to the fact that Lautenbach was run as a commercial farm. Although this limited the range of husbandry practices which could be implemented, it ensured that the practices remained practical and economical. Much of the valuable information from Lautenbach was derived from the fact that this was a comprehensive, interdisciplinary study permitting investigations of many interacting factors.

INTEX, Germany (1989-1997)

Background In 1989 a new programme of research, “INTEX”, was initiated involving scientists from

the University of Gottingen, coordinated by the Research Centre for Agriculture and the Environment. The aims are to develop further an integrated farming system and to examine the economic and ecological effects of extensification in arable production (Wildenhayn, 1992).

Design and methods This project covers a total area of 94 ha and is located at three sites in Lower Saxony,

Northern Germany. Four farming systems (conventional, integrated, reduced input and extensive) and five years’ fallow (Table 2) are compared at each site. At two sites (Reinshof and Marienstein) each course of a three or four-course rotation is present each year. At the third site (Eickhorf) only one crop per year is produced for systems I11 and IV. The crop husbandry practices are shown in Table 3. These include reduced inputs of mineral fertilisers and pesticides, in addition to more positive measures such as the use of disease resistant and mixed cultivars, reduced tillage systems, mechanical weed control and wider field margins to encourage beneficial invertebrates.

A comprehensive range of ecological parameters is measured by specialist departments at the University of Gottingen. Soil studies include investigations of soil structure, water balance, microbial activity, nutrient balances and Lumbricidae populations. The effect of each farming system on plant pathogens is monitored in conjunction with small plot trials to confirm results from varying fungicide strategies. The species composition and abundance of crop pests and their natural enemies is being investigated by entomologists. Further studies on invertebrates are under way with the emphasis on the abundance and diversity of epigeal and edaphic arthropods, especially beneficial species. The Geobotanical group is investigating the secondary succession of fallow lands by assessing plant community structure, soil seed-bank and nitrogen levels in soils and plants. The Weed Research group is concentrating on the effects of mechanical weed control. Studies on herbicide degradation and leaching and their subsequent effects on the soil microflora are also under way. The economic aspects of each farming system are being analysed and ultimately the goal is to

422 J M HOLLAND. G K FRAMPTON. T CILGI AND S D WRATI‘EN

model the experimental results so that the impact of extensification at the whole farm level can be assessed.

Princilial results Results from the first three years were difficult to interpret because there are no within-

year crop replications. and differences were found in soil type within and between the systems and between the experimental sites. In addition. yields were influenced by variable weather conditions, in particular by unusually dry periods. This shows the benefit of long temporal scales which ensure that the farming system is tested under a more extensive range of climatic conditions.

Preliminary results show that integrated farming of winter oilseed rape (25% less nitrogen and 70% fewer pesticides than the high input system) seems to be possible without any reduction in yield. In contrast, oilseed rape yield losses in the reduced-input system (50% less nitrogen. no insecticides) were 33% compared with 71% in the extensive system (no nitrogen, no pesticides). The corresponding results for winter wheat were reductions of 22% in the integrated system. 23% in the reduced-input system, and 54% in the extensive system. I n winter barley the reductions were 9% (integrated), 13% (reduced) and 499k (extensive) (M Wildenhayn. unpublished data).

Investigations of the soil nutrient budget (Przemeck & Lickfett, 1992) revealed a clear influence of soil type and location on residual nitrate content, which can be more important than the effects of different rates of nitrogen fertilisation. So for best possible use of nutrient supply. crop choice and crop rotation seem to be more significant than reduction of fertilisation.

Zoological investigations (Stippich & Buchner. 1992) show that responses of many animal taxa were similar within cultivation systems (e.g. ploughing or reduced tillage) but differed between systems. indicating that soil cultivation measures may be important. The most diverse fauna was present in the integrated system. Compared with cultivated fields. carabid species diversity and the abundance of some other beneficial arthropods was considerably higher on fallow lands.

Where reduced tillage and mechanical weed control were both used in a farming system, higher numbers of weed species and greater weed cover was recorded (Schmidt & Waldhardt. 1992). The most agronomically important increases were by species such as Galium aparine L. at Reinshof, Alopecurus myosuroides Huds. at Marienstein and Apera spica-venti (L.) Beauv. at Eickhorst. These reached sufficient densities to undermine the sustainability of farming systems utilising the above methods (Przemeck & Lickfett, 1992).

Nagele. the Netherlands (1979- )

Background This project, on the development of farming systems (DFS), was started following

opposition in the 1970’s to intensive agricultural production and increasing interest in alternative agricultural methods. Three organic farms at Nagele in the North-east polder were chosen for this long-term research programme. Three farming systems were compared on a farm scale: a conventional arable farm, an integrated arable farm and a biodynamic/ organic mixed farm. The aim of the conventional system was to maximise financial returns, whilst in the integrated system the aim was to achieve similar financial returns but with the minimum level of fertiliser, pesticide and machinery inputs. In the organic system the aim was to be self-supporting in fertiliser and fodder. The primary objectives were to: a. Develop the integrated and organic farming systems in theory and test them in practice. h. Evaluate the results of the systems, based on their specific aims.


c. Compare the organic and integrated systems with the conventional system (Vereijken, 1990).

Design and methods The organic farming system was started in 1977 while the other two systems were started

in 1979. The husbandry practices for each system are outlined in Table 3. A four-year rotation was chosen in the conventional and integrated systems because it maximised profits. A longer rotation was used in the organic system because greater control of soil-borne pests and diseases and increased soil fertility may be achieved. Pesticide use in the integrated system was reduced by a combination of later sowing dates, reduced nitrogen inputs, disease-resistant varieties and mechanical weed control. Selective chemicals and lower rates of pesticides were applied when excessive economic loss was expected. As a result of these practices some yield loss was accepted in the integrated and organic systems. Successful husbandry practices developed in the organic system were sometimes incorporated into the integrated system, an option not available in the other research projects.

Detailed assessments were made on the economic and agronomic impact of each farming system, but only limited environmental monitoring was carried out.

Principal results The economics of each farming system received particular attention at Nagele. For the

period 1985-88, the total returns were 6% lower for the integrated compared with the conventional system but the operational costs were similar. However, because of lower variable costs, the net surplus in the integrated system was higher despite a reduction in yield (Wijnands, 1990). The organic system produced the lowest net surplus. This was because of high production costs which were not offset by the higher premium obtained for the organic crops and because the labour productivity (net surplus divided by labour cost) was half that of the other two systems (Vereijken, 1990). The amount of pesticide applied in the period 1985-87 was reduced in wheat (60%), potatoes (96%), sugar beet (63%). peas (12%), onions (57%) and winter carrots (66%) in the integrated compared with the conventional system (Vereijken, 19893). The average number of pesticide treatments per season (1985-88) in the conventional system was 9.2 per field compared with 4.3 in the integrated system (Wijnands, 1990). Average inputs (1985-88) of mineral potassium, phosphorus and nitrogen were reduced by 66%, 100% and 60% respectively in the integrated system because of the change to organic fertilisers. Average (1985-88) nitrate-nitrogen levels in drain water were 10,s and 3 mg litre-' for the conventional, integrated and organic systems (Wijnands, 1990). In the last two systems, nitrogen mineralisation after harvest was successfully recovered by post-harvest cover crops.

Only limited results have been widely published on the agronomic and environmental assessments. Predator species diversity, abundance and guild composition were examined in a range of crops. The main factor affecting predator abundance was crop type. Crops which established extensive ground cover early in the season favoured higher populations of carabids, staphylinids and arachnids (Booij & Noorlander, 1992). The farming system had some influence on species composition and abundance with certain species being confined to particular crops and farming systems.

Nagele, Vredepeel and Borgerswold (1986 )

Background To extend the research to the major soil types and crop types for arable farming in the

Netherlands, two further experimental farms located in the north-east (Borgerswold) and


south-east (Vredepeel) were included in 1986 and 1989 (Wijnands & Vereijken, 1992). The integrated strategy was developed according to the soil type and crop types typically grown but also aimed to reduce disease and soil fertility problems which had developed as a result of the previous husbandry practices.

Design and methods Detailed descriptions of the integrated systems are available in Wijnands & Vereijken

(1992) and are summarised in Table 2. The conventional and integrated reference systems are compared at Nagele. whilst at Borgeswold an additional conventional system includes fewer root crops. as did the integrated system. At Vredepeel the conventional reference system is compared with three integrated systems (reference, more and less root crops). The economic feasibility of each system is determined from gross margins, net farm profits and labour returns. Environmental feasibility is determined from input levels of fertilisers and pesticides.

Principal results The gross margins for each system were mainly dependent on the chosen rotation. At

Nagele gross margins were similar for both systems. whilst at Borgerswold the integrated low root crop system performed better than the equivalent conventional system, but was poorer than the conventional reference system. Gross margins at Vredepeel were dominated by the crop rotation and differences between the management of the systems had not had sufficient time to exert their effect on agronomic factors.

For the period 1986-90, pesticide use for the three sites has been reduced by S0-6S%, excluding nematicides, and by 85-9S% including nematicides, in the integrated systems (Wijnands & Vereijken, 1992). Reductions in herbicide inputs of 60-75% were achieved through mechanical control and band spraying techniques. Fungicides were reduced by SO- 65% , depending on the location. by using resistant cultivars, appropriate rotations, moderate nitrogen fertilisation and close monitoring of infection periods. Insecticide use was minimised by use of control thresholds. band-spraying and reduced doses. Resistant varieties have removed the need for nematicides in the integrated systems. As in the original Nagele experiment, the incorporation of organic manures reduced the need for inorganic fertilisers by supplying 100% of the required phosphorus, 60-100% of the potassium and 60-70% of the nitrogen. Overall, the total nutrient inputs were lower in the integrated systems and this, combined with the use of cheaper organic manures. has lowered fertiliser costs.

Other information To extend the development of integrated farming to an even greater diversity of soil types

and farm and management conditions, five pilot groups each of eight farms are converting to integrated farming during 1990-94 (Wijnands, 1992). using the protocols developed at the experimental farms. Detailed baseline assessments were carried out on soil fertility and soil-borne pests and diseases for three years prior to the start of the project. Following this period the performance of each farm has been determined from assessments of the economics (financial returns and profits, direct and fixed costs. labour inputs), agronomic aspecis (cropping techniques; levels of soilborne pests, diseases and weeds; yield quality and quantity: yield stability) and environmental compatibility (pesticide and nutrient inputs, soil fertility, nitrate loss). Economic and environmental data were compared either with that from the baseline years for each farm or with a reference base such as that from the national economic survey of arable farms. Results from the first year showed reductions in total use of pesticides in the integrated systems (kg ha-' active ingredient) of 36% compared


with 1987-89. The use of organic manure and an integrated nutrient strategy reduced phosphorus inputs by 1&30 kg ha-’ and total nitrogen inputs by an average of 45 kg ha-’. Detailed economic analyses are not yet available. The response by farmers to the new system was enthusiastic particularly because managing such a system improved their awareness of crop production processes (Wijnands, 1992).

The concept of, and training in, IFS is being extended to the farming community (Wijnands, 1992). The concept and experimental development of integrated farming has also been extended to other production systems in the Netherlands including flower bulb production (Raven & Stokkers, 1992), fruit production (Schenk & Wertheim, 1992), glasshouse horticulture (Wells, 1992), nursery stock (Dolmans, 1992), dairy farming (Aarts, Biewinga & van Keulen, 1992), grassland farms (Hermans & Vereijken, 1992) and pig production (Den Hartog, 1992).

Third Way project, Switzerland (1981- )

Background Swiss agriculture cannot compete on the EC open market owing to high production costs

so produce is therefore only for the national market, through which the country is 60% self-sufficient. Swiss consumers may be prepared to pay more for home-produced food if it is of high quality and ‘ecologically’ produced (Hani, 1989). Therefore, the ‘Third Way’ project (the ‘first-way’ being high input, the ‘second-way’ a reductionist approach) was set up to develop a long-term, sustainable agricultural system whilst safeguarding the environment. Three commercial farms have been converted to integrated farming with the following aims: a. To maintain and promote natural pest, weed and disease regulation factors. b. To reduce external inputs (fertilisers, pesticides, energy). c. To maintain income, employment and social structure (Hani, 1989).

by the Swiss Technical College of Agriculture. The project was originally operated on a voluntary basis until 1988 when it was adopted

Design and methods Two farms are located at Ipsach; one is a mixed farm of 16 ha, the second an arable farm

of 20 ha with fattening pigs. The third farm is at Schlosswil and is also a mixed farm (10 ha). The whole of each farm is managed using an integrated system. Within each experimental field a strip 12 m wide running the length of the field is farmed using a conventional system and an area, 30 m x 12 m, is managed using no pesticides. To allow for factorial analysis and to extend the ecological experiments other conventionally farmed sites are also used.

The extensive range of husbandry practices used in the integrated system includes lower rates of nitrogen, organic manures, cover crops, undersowing, resistant and/or mixed varieties, selective pesticide use (based on toxicity to man and non-target species, persist- ence, mobility, resistance risk, reduced application rates and spot treatments), later sowing dates, mechanical weeding and reduced soil tillage. In addition 5-10% of the land (field margins, slopes and unmanured rough meadows) is uncultivated to enhance beneficial insect and wildflower populations. Each year the husbandry practices are evaluated and improved using previous years’ experience and outside knowledge.

The economics and physical yields of each system are evaluated for the within-field comparisons and compared with conventional farms of similar structure. Crop diseases, pests and weeds are assessed within each system. Several bioindicators are used to determine

420 J M HOLLAKD. G K FRAMPTON. 7' CILGI AND S D WRATTEN

the environmental impact of each farming system. This includes monitoring populations of Lumbricidae. Nematoda. Araneae. Carabidae. Entomophthoraceae, Staphylinidae, Syr- phidae and parasitic Hymenoptera. Cellulose degradation rates in the soil are also recorded.

I'rincipl rrsri/t.s When the gloss margins f o r the integrated systems (Ipsach farms) were compared with

those for conventionally managed farms. it higher gross margin was obtained for winter wheat (5-857). sugar beet (7-15%) and potatoes (17%), but a lower one for maize (4%). For the within-field comparisons ( 1986-89). the gross margins were similar for the integrated and conventional systems. but the no-pesticide regime produced a 13% lower margin. The integrated y'stem involved higher labour costs which are not incorporated in the gross mayins . The mean number of pesticide treatments per field per year was reduced to 1.2 in the integrated system compared with 4.4 in the conventional system. No insecticides, neniaticides or plant growth reguiators were applied in the integrated system.

The ~Cologiciil investigations re\.ealed that Lumbricidae populations had been increased $11 thc integrated system and higher rates of cellulose degradation occurred. Weeds within the field5 and herbaceous borders along hedges enhanced the populations of beneficial ~.~rganisins such as Syrphidae. Staphylinidae. some Carabidae species, parasitic Hymenop- tera. Araneae and Entomophthoraceae. Aphids (Homoptera) appeared earlier in the integrated ssstem but failed to reach threshold levels.

O t k r iiiformariori The "Third Way' project has so far shown that integrated farming is economically feasible

and substantial reductions in inorpnic inputs are possible. The yield and financial results m a ) underestimate the impact of the integrated system because the adjacent conventional 5 y t e m may be benefiting from some of the integrated practices such as the crop rotation and use o f resistant culti\,ars. Furthermore. because of the experimental design, beneficial insect5 could have moved between the systems and subsequently the level of biological control i n the con1,entional system may have been higher than expected. Thus, the scale of such jtudizs i s an important factor which can influence the interpretation of results.

essments in neighbouring conventionally managed farms would help to evaluate the responses detected here.

Integrated farming in France is seen as an effective instrument for extensification and an altern;rti\e n a y to reduce overproduction and declining farm incomes than a switch to a world-market orientated s p t e m (Viaux. Roturier & Bouchet, 1989). Simulation studies which evaluated the economic viability of extensification by comparing a conventional system with tw:o integrated systems provided promising results. The gross margin was only 7"; lower for the integrated systems and some crops responded better to lower inputs. To cvtend the k n o n ledge and to investigate the reliability of integrated farming, experimental trial5 Lvcre set up to compile and compare economic and agronomic data on conventional and integrated systems. These were established by Institute Technique des Cereales et des Fourrases and the Association de Coordination Technique Agricole.

DP \ ig i1 (ii 1 ( i 171 rJth od I

(. o ~ i ~ cntiond m d integrated farming are being compared in trials established in four


contrasting regions. Plots of 1-5 ha are used. The aim in the integrated system is to reduce the need for agrochemicals by exploiting natural control and genetic resources, combined with lower target yields. The range of husbandry practices employed is shown in Table 2 and includes the use of at least a three-course rotation, minimal tillage, later sowing dates, lower sowing densities, disease-resistant varieties, lower nitrogen fertilisation and use of thresholds and models to determine pesticide application rates and timings.

Economic assessments involve recording all fixed and variable costs, labour inputs and outputs, so allowing economic indicators such as gross margins, direct margins and net margins to be calculated for each crop and rotation. Overall results from each site have also been calculated. Agronomic data on crop establishment, crop growth rates, pest, disease and weed levels, yields and crop quality are also recorded.

Principal results Results from the first one or two years indicated that the integrated system achieved

reductions in input costs of approximately 34% for all four sites compared with the conventional system. This was attributed to reductions in fungicides by 57-75%, insecticides by 40-73%, fertilisers by 2845% and seed rates at some sites. Herbicide savings were difficult to achieve primarily because of the non-inversion tillage practice and high densities of competitive weeds at some sites. There was considerable variation between crops and sites in the reductions in herbicide achieved, with 42-75% for wheat, 26% and 59% for peas, 43% for sunflower, 39% for barley and 17% for oilseed rape. Yields in the integrated system were either equivalent to or lower than the conventional system, depending to some extent on incidental weed and disease infestations. Overall gross margins varied between sites. At Courseulles a better gross margin was obtained in the integrated system and was primarily attributed to wheat and sugar beet crops, whilst at Boigneville the gross margin was 30% lower in the integrated system because of poor crop establishment and relatively high input costs. At Saint-Hilaire, overail gross margins were similar for the two systems but there was considerable variation between crops. In the integrated system gross margins for barley were 1000 FF ha-' higher but 29% lower for oilseed rape. At Montgaillard, overall gross margins were similar although differences were found between crops. The gross margin for sunflowers was favoured by the integrated system, but this was not the case with peas.

Other information This project has shown that considerable variation can be expected in the performance

of crops at different sites, emphasising the importance of multi-site investigations. These differences may be attributed to differences in soil type, climate and sporadic pest infestations, the effects of the latter being detected only by long term investigations. Furthermore, in the initial years of a study, an integrated system may gain from the low incidence of weeds and the presence of sufficient P and K soil reserves and only in the long term will such a system's success or failure be revealed. Difficulties in weed control using non- inversion tillage have already been experienced and will have to be overcome if sustainability is to be maintained.

Discussion

Experimental design The projects outlined in this paper have had to confront a number of problems associated

with designing and implementing large-scale field projects. Some of these have been


overcome in more recent projects, although a number still remain. largely because of financial and logistic constraints. The first problem is the question of scale. Ideally the scale of each project should be tailored to meet the objectives. Where, however, a large range of investigations is carried out there may be conflict. For example. environmental monitoring may require the largest attainable spatial scale to assess highly dispersive species whereas a smaller scale may be required for agronomic investigations in which inherent variability must be minimised. A small spatial scale is particularly important if the environmental or agronomic factors are influenced by soil type, which may exhibit greater variability over increasing distance. Soil type may influence soil nutrient and water-holding capacity, soil organic matter. soil fauna and the seed bank. Subsequently these may indirectly influence crop growth. yields and the economics of the production system. Smaller reference areas within each plot with comparable environmental conditions may be used for some assessments to reduce such variation.

For studies of small mammals, birds and some invertebrate species the spatial scale must be sufficient to minimise the impact of movement between different farming systems. If movement does occur this may confound treatment affects directly by increasing or decreasing the actual numbers recorded in each system. In addition, changes in species abundance or diversity may influence other factors which are monitored; for example, the distribution of predator species may affect pest abundance. Even the 53 ha full insurance block at Boxworth may have been too small for the study of some small mammals and birds. In the SCARAB project the emphasis is on invertebrate studies but whether the size of the fields will he sufficient for such a study remains to be seen. Some Carabidae have been recorded moving up to 17 m h-' (Wallin & Ekbom. 1988) and therefore within-year and between- year redistribution of certain species is likely in all the studies reviewed here. This may be greater where plot size is small (such as in TALISMAN and RISC), or where pitfall traps in neighbouring plots are nearby (e.g. in LIFE and LINK IFS). The transects of pitfall traps arranged either side of the cropping system boundary in the SCARAB study may help quantify movement between plots by invertebrate species. Artificial barriers can be used to prevent the movement between adjacent plots of some ground-dwelling invertebrate species; however. this may confound interpretation of results because the extent to which particular species are restricted cannot always be identified. More appropriately treatment effects may be more readily identified by selecting indicator species with limited powers of dispersal. To assess the more mobile species, demonstration farms (e.g. LEAF farms) could provide the opportunity for even larger scale, environmental monitoring studies, particularly if adjacent farms with similar field dimensions, soil types, crops and non-crop habitats could be found.

The choice of experimental design is also complicated in large scale studies, where there are several objectives to attain, because of financial or logistical constraints. A single site with many replicates may allow more extensive orthodox statistical analysis, more intensive testing of the experimental systems and more intensive monitoring. Ultimately, however, the techniques must be extended to further sites if nationwide applicability is to be attained. In contrast. multi-site studies may incorporate a greater range of environmental variables and husbandry practices, but may forfeit replication and detailed monitoring. To a certain extent the lack of orthodox replication in some of these studies may be compensated for by their ability to include a wider range of variables into multivariate analytical techniques than a stringent spatial design would permit. In projects like SCARAB, TALISMAN, LINK IFS and INTEX which have several more years to run, all the possible crop-species- pesticide-soil type combinations have not yet occurred: for these and similar projects multivariate analyses may be more appropriate after several years of representative data have been accumulated. The option of using multivariate analytical methods in these


projects derives from the flexibility of the analytical techniques rather than the layout of the projects.

Finally, the flexibility of a regime will have an important bearing on its realism and has to be decided at the outset to ensure consistency over the experimental period. A problem which arose at Boxworth was that over the duration of the experiment there was a trend in the UK away from high inputs of agrochemicals and virtually continuous cereal crops to lower inputs and more break crops. This outdated the inflexible full insurance regime so that it no longer typified current farming practice. This inflexible approach was necessary to allow sufficient time for environmental differences to develop and be detected. The rapid changes in commercial arable production, which occur in response to reductions in price support and market prices, continue to present experimental difficulties because farmers are cutting inputs further, narrowing the gap between conventional and integrated farming systems. This may be overcome in long-term studies only by adopting a flexible approach in both systems and by incorporating more positive measures such as habitat manipulation to enhance predator numbers (Thomas et al., 1991) in the integrated systems, as opposed to the reduced-input “reductionist” approach. Changes in husbandry practices, however, during the experimental period may limit repetition of individual practices available for statistical analysis.

Future husbandry practices The range of husbandry practices which are currently used in the integrated farming

systems described in these experiments could be extended further to include, for example: use of the stale seedbed technique (whereby land is cultivated to encourage weed ger- mination prior to mechanical destruction during drilling), undersowing with legumes, higher seed rates (Millington, Stopes & Woodward, 1990) and the use of mixed varieties (Wolfe, 1990). The last method would be easiest to implement if crops are designated for the animal feed market, but would require the cooperation of the relevant industries if for milling or malting. Natural control of weeds, pests and diseases can also be enhanced by increasing crop diversity within the field by multiple cropping, whereby two crops are grown simultaneously in the same field, as is practised throughout the tropics (Edwards & Stinner, 1990). This can be achieved by growing alternate rows of two or more crops (row intercropping), growing alternate drill widths of two crops (strip intercropping) (Edwards, Brust, Stinner & McCartney, 1992) or growing strips wide enough to allow independent husbandry practices (strip cropping) (Francis, 1986). Undersowing is another form of intercropping. Whether intercropping improves pest control was examined in a survey comparing intercropping with monoculture using annual crops (Risch, Andow & Altieri, 1983). For populations of monophagous insect species, 58% decreased, 3% increased, 23% showed a varied response and 15% no change in the intercropped system compared with monoculture. Row and strip intercropping may, however, present difficulties in the choice of crop protection chemicals and harvesting, whereas strip cropping may be a more practical alternative for highly mechanised crop production systems.

Some of these practices are already incorporated in organic production systems. It is unlikely that integrated farming will develop into a fully organic approach because, at present, organic arable production provides a lower income (Woodward & Lampkin, 1990). In addition, organic production is most suitable for mixed farms where reliable supplies of animal manures are available. Integrated farming can, however, also benefit to the same extent from the use of animal manures where available, but the costs of switching to mixed farming would be prohibitively expensive for the majority of arable-only farms. Careful calculations would be required to balance organic manure requirement with livestock production in order to prevent a surplus of livestock products. The change to organic


production would be costly, estimated at an annual cost of 300-500 ha-’ (Murphy, 1991) and would require an extensive training programme. Finally, the relatively slow growth of the organic produce market may indicate that either consumers are not prepared to pay the extra premium for such produce. that its quality is deemed inferior or that supply and quality are inadequate for large retail outlets. Nevertheless. the assumed environmental benefits of organic farming have stimulated interest among consumers that was not foreseen a decade ago. Even if organic farming does prove to be economically difficult, as currently seems likely, it is not known at present just how far the integrated approach can be taken in the direction of organic farming before economic viability will be lost.

One of the most important components of any IFS system is the crop rotation (El Titi et al . , 1993). This should be chosen to achieve the maximum benefits of natural pest, weed and disease regulation whilst maintaining or enhancing soil structure and fertility (Jordan, 1992). All of the projects reviewed here, with the exception of the Boxworth project, meet the minimum requirement of four crops as designated in the guidelines for Integrated Production (El Titi et a l . , 1993). However. with rare exceptions, because of experimental constraints only one type of rotation is tested for each farming system at each site. This represents a fraction of the rotations possible with the current range of commercially grown crops. The number of rotations can be extended through the development of pilot farms, but where this has been done. for example in the Netherlands, it has been without any accompanying detailed monitoring programme. Agronomic, economic and limited environmental assessments for a range of rotations should be carried out on a smaller spatial scale in conjunction with the main programme, if financial resources are available, to validate findings from the experimental projects.

Ultimately, if an integrated system is to be developed for farmers, their cooperation and input should be sought throughout the study to ensure that the system is operationally feasible in terms of such factors as labour management, machinery availability and crop- walking requirements. I n return for farmers’ inputs, researchers can determine what information is required in order for farmers to switch to an integrated system (El Titi, 1989). The risk potential of such a system should also be determined. Later sowing, for example of winter cereals, may reduce the need for autumn herbicides and insecticides but operationally may not be feasible across the whole farm and because of the approaching winter. favourable drilling conditions may occur infrequently.

Eualuatir1e research The research component for the projects reviewed here has many common objectives.

The economics of each farming system have been assessed in all projects as have the impact of the different regimes on weeds, pests and disease levels (Table 4). Soil mineral levels were evaluated in the majority of projects to evaluate nutrient cycling and to aid the determination of fertiliser requirements. Non-target invertebrates were the most widely used indicators of environmental fitness. Soil fertility was frequently determined from estimates of Lumbricidae populations and cellulose degradation rates. Because of the similarity in the research objectives of these projects, standard assessment procedures could be developed and used in future studies, facilitating direct comparisons between sites for verification of results. Progress towards this has already been made with the formation of a working group on Integrated and Ecological Arable Farming Systems (IAFS and EAFS) which involves representatives from each of the EC-member-countries and five countries from outside the EC (Finland, Norway, Poland, Sweden and Switzerland). The group’s objectives are to coordinate their research methodology and extension services, exchange results and provide information on IAFS and EAFS to scientists and farmers within their country .


Many of the projects described here have provided excellent opportunities for agronomic and environmental monitoring. This could in some cases be expanded to provide greater understanding of the interactive components of the crop production system and ultimately lead to better management. Commonly, financial constraints limit investigative studies with the result that further experimentation may be required. Furthermore, the projects under way at present will run through only one cycle of the rotation. Whether this is sufficient for the assessment of agronomic and environmental factors is debatable. We have regarded the studies reviewed here as long-term mainly to distinguish them from the many single-season agricultural experiments that are undertaken in the UK and continental Europe each year. Quite how long a study needs to run will depend on the spatial scale of the experiment and the organisms being studied and it may be difficult to decide whether a study would reveal further effects if extended in time. Studies of farmland flora, for example, could require many years because the long-term viability of the soil seed bank may obscure effects of farming practices in the short term. Studies of long duration have the advantage not only of covering a more representative range of temporally variable biotic and abiotic conditions. but also they have more chance of including atypical conditions which could be just as important. At Boxworth, for example, weather conditions in 1986 led to an abnormally late winter insecticide application to full insurance fields and exceptionally poor crop cover, and these factors may have been responsible for triggering a long-term decline in the population of the carabid Agonum dorsale (Pont.) (Vickerman, 1992) and for adverse changes in several groups of Collembola (Frampton et al., 1992).

Overall results Those studies which have specifically aimed to develop an integrated farming system

(LIFE, Lautenbach, Nagele, Third Way) show that for the overall rotation, gross margins can be maintained or increased relative to conventional systems (Table 5). Some variation between the profitability of different crops was found. Yields were usually lower in the integrated systems but this was compensated for by lower inputs and therefore lower costs of fertilisers and agrochemicals combined with improvements in nutrient retention. In the majority of studies, weed populations were increased in the integrated systems as a result of minimal tillage and lower herbicide inputs. Whether these were sufficient subsequently to reduce yield, inhibit harvesting procedures or increase seed cleaning costs has not been reported. Problems of weed control were greatest at Lautenbach, where seed crops in which weeds and crop volunteers could not be tolerated were produced. The success of this project despite such restrictions indicates that crops produced using an integrated system can reach the highest quality. The effect of integrated farming on pests and diseases was less well defined and considerable differences were reported for different crops and locations. This may be a result of the sporadic nature of pest and disease incidence. Only small-plot validation studies can help unravel the complex of influencing factors which govern pest and disease levels. One of these main factors is the rate of nitrogen inputs (Vereijken, 1989a) and this needs further investigation.

The main environmental indicators assessed, notably non-target arthropods and Lum- bricidae, showed higher populations in the lower-input or integrated areas in every study where they were assessed (Table 5). Increases in invertebrate populations could be attributed to lower pesticide inputs and/or enhancement of their habitats, for example through greater weed species diversity and abundance, improved field margin management and change to non-inversion tillage practices. Trials in Austria showed up to a 30-fold increase in Lum- bricidae populations following a non-inversion tillage and sowing system (‘Horsch’) compared with ploughing (Berger, 1990).

Although some of the studies considered above have broadly similar objectives, this does

432 J M HOLLAND. G K FRAMPTON, T CILGI AND S D WRATTEN

not necessarily mean that their results are comparable. Boxworth and SCARAB were set up specifically to investigate only the pesticide component of IFS and are better equipped to assess environmental effects than are economically-driven studies such as TALISMAN and LIFE. Conversely. it would be unwise. for example, to make comparisons between the limited integrated regime of Boxworth and the detailed integrated component of the LIFE, LINK IFS, Lautenbach and Nagele projects. However, by comparing these large-scale, long-term studies it has been possible here to give an overview of the European effort currently engaged in research leading towards environmentally benign, economically acceptable. low-input arable farming.

Complete results from many of the above projects will not be available until the mid-late 1990's. by which stage. with the predicted decline in the rate of subsidies, economics will have driven farm inputs to their economic minimum. The adoption of an integrated system appears to maintain profits and reduce inputs whilst achieving environmental benefits, as demonstrated by some of these projects; the merits of a true organic approach are at present less clear. although trends in the USA and New Zealand are more positive (Wratten, 1993). Further work is. however. required to develop more novel approaches to arable crop production. utilising the experience of organic growers and widening the methods of arable crop production.

Glossary of Acronyms and Abbreviations ADAS

CAP - Common Agricultural Policy CCP - Current Commercial Practice (In TALISMAN project) CEC - Commission of the European Communities CFP - Current Farm Practice (in RISC and SCARAB projects)

- Conventional Farm Practice (In LINK project) CFS - Current Farm System (in Lautenbach project) D A M - Department of Agriculture. Northern Ireland DFS - Development of Farming Systems (project name in the Netherlands) EAFS - Ecological Arable Farming Systems ELI - European Union (previously European Community) GATT - General Agreement on Tariffs and Trade IACK - Institute of Arable Crops Research, UK IAFS - Integrated Arable Farming System ICM - Integrated Crop Management (same as IFS) IFS - Integrated Farming System (same as ICM) ILIA - Integrated Low Input Approach (in TALISMAN project; same as INT) INT - Integrated Low Input Approach (in RISC project; same as ILIA) INTEX - Project name in Germany IOBC ' WPRS - International Organisation for Biological Control/West Palaearctic

LEAF - Linking Environment And Farming (project name in UK) LIA - Low Input Approach (in RISC and TALISMAN projects) L I F t - Less-Intensive Farming and Environmental Research (project name in

L I N K - LINK Integrated Farming Systems (project name in UK) MAFF

- Agricultural Development Advisory Service (an executive agency of MAFF, UK)

Region Sector

UK)

- Ministry of Agriculture, Fisheries and Food (UK)


MIN OECD RISC SCARAB

TALISMAN

THIRD W A Y - Project name in Switzerland

- Minimum Input farming system (in RISC project) - Organisation for Economic Co-operation and Development - Reduced Input Systems of Cropping (project name in UK) - Seeking Confirmation About Results A t Boxworth (project name in

- Towards A Low Input System Minimising Agrochemicals and Nitrogen UK)

(project name in UK)

Acknowledgements J M Holland is funded by The Game Conservancy Trust on the LINK IFS project. G K

Frampton and T Cilgi are currently funded by MAFF to study invertebrate aspects of the SCARAB project. These employers are gratefully acknowledged.

W e also thank L Easson, A El Titi, F Hani, V Jordan, S E Ogilvy, N W Sotherton, P Vereijken, P Viaux, M Wildenhayn and J Young for their comments on this manuscript.

References Aarts H F M, Biewinga E E, van Keulen H. 1992. Dairy farming systems based on efficient nutrient

Abel C, Hill P. 1993. Focus on LEAF. Farmers Weekly, pp. 3 H 1 , 22 January 1993. Aebischer N J. 1991. Twenty years of monitoring invertebrates and weeds in cereal fields in Sussex.

In The Ecology of Temperate Cereal Fields, pp. 305-332. Eds L G Firbank, N Carter, J F Darbyshire and G R Potts. Oxford: Blackwell Scientific Publications.

Anon. 1991. Code of good agriculturalpractice for the protection of water. London: MAFF Publications. Anon. 1993. Handbook for Countryside Stewardship. Cheltenham, Gloucestershire: Countryside

Commission. Archer J, Lord E. 1993. Results from the pilot nitrate scheme. In Solving the nitrate problem, pp. 29-

32. London: MAFF Publications. Berger H K. 1990. Effects of alternative soil cultivation and sowing techniques on epigeic fauna. Swiss

Journal of Agricultural Research 29:385-386. Booij C J H, Noorlander J. 1992. Farming systems and insect predators. Agriculture, Ecosystems and

Environment 40: 125-135. Bowerman P. 1993. Sequels to the Boxworth Project - Studies of environmental, agronomic and

economic effects of reduced crop inputs. Journal of the Royal Agricultural Society of England

Burn A J. 1989. Long-term effects of pesticides on natural enemies of cereal crop pests. In Pesticides and non-target invertebrates, pp. 177-194. Ed. P C Jepson. Wimborne, Dorset: Intercept

Burn A J. 1992. Interactions between cereal pests and their predators and parasites. In Pesticides, Cereal Farming and the Environment, pp. 110-131. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Cilgi T, Wratten S D, Frampton G K, Holland J M. 1993. The long-term effects of pesticides on beneficial invertebrates - lessons from the Boxworth project. Pesticide Outlook 4(4):30-35.

Clarke J H, Bowerman P, Young J E B, Cook S K, Jones A E, Groves S J, Green M. 1993. Effect of recommended and reduced rate herbicides on weed number, yield and gross margin in TALISMAN: report on the first two years. Proceedings of the Brighton Crop Protection Conference - Weeds, pp.

Cooper D A. 1990. Development of an experimental programme to pursue the results of the Boxworth project. Proceedings of the Brighton Crop Protection Conference - Pests and Diseases, pp. 153-162.

management. Netherlands Journal of Agricultural Science 40:285-300.

154: 45-60.

1009-1014.


Den Hartog L A. 1992. Towards integrated pig production in the Netherlands. Netherlands Jolirnal of Agriculturul Science 40:3 15-326.

de Wit C T, Huisman H, Rabbinge R. 1987. Agriculture and its environment: are there other ways? Agricultural Systems 23:211-236.

Dolmans N G M. 1992. Integrated nursery stock production. Netherlands Journal of Agricultural Science 40269-276.

Easson L. 1993. Reducing the input of agrochemicals in cropping systems. Agriculture in Northern Ireland 7:20-21,

Edwards C A, Stinner B R. 1990. The use of innovative agricultural practices in a farm systems context for pest control in the 1990s. Proceedings of the Brighton Crop Protection Coriference - Pests and Diseases. pp. 679-684.

Edwards C A, Brust G E, Stinner B R, McCartney D A. 1992. Work in the United States on the use of cropping patterns to promote natural enemies of pests. Aspects of Applied Biology 31, Inter- pretrrtion of Pesticide Effects o n Beneficial Arthropods, pp. 139-148.

El Titi A. 1986. Management of cereal pests and diseases in integrated farming systems. Proceedings of' f h e Brigitton Crop Protection Conference - Pests and Diseases, pp. 147-156.

El Titi A. 1989. Research on integrated farming systems at Lautenbach. IOBCIWPRS Bulletin 19891 XIZ/5. pp. 21-36. Wageningen, The Netherlands.

El Titi A. 1990. Farming systems research at Lautenbach. Germany. Swiss Journal of Agricultural Research 29:237-237.

El Titi A. 1991. The Lautenbach project 1978-89: integrated wheat production on a commercial arable farm. south-west Germany. In The Ecology of Temperate Cereal Fields, pp. 399411. Eds L G Firbank. N Carter, J F Darbyshire and G R Potts. Oxford: Blackwell Scientific Publications.

El Titi A, Ipach W. 1989. Soil fauna in sustainable agriculture: results of an integrated farming system at Lautenbach. FRG. Agriculture, Ecosystems and Environment 27:561-572.

El Titi A, Landes H. 1990. Integrated farming systems of Lautenbach: a practical contribution toward sustainable agriculture in Europe. In Swtainable Agricultural Systems, pp. 265-286. Eds C A Edwards. R Lal. P Madden. R Miller and G House. Ankeny, Iowa: Soil and Water Conservation Society.

El Titi A, BolIer E F, Gendrier J P. 1993. Integrated Production - Principles and technical guidelines. IOBC,'WPRS Bulletin 16( 1 ) . 96 pp.

Filser J . Fromm H, Nagel R F, Winter K. 1994. Long-lasting negative effects of intensive agricultural management on the biodiversity of the soil fauna. Plant and Soil (In press).

Finney J . 1988. World crop protection prospects: demisting the crystal ball. Proceedings of the Brighton Crop Protection Conference - Pests and Diseases. pp. S l 4 .

Fletcher M R, Jones S A, Greig-Smith P W, Hardy A R. 1992. Population density and breeding success of birds. In Pesticides, Cereal Furrning and the Environment. pp. 16@174. Eds P W Greig-Smith. G K Frampton and A R Hardy. London: HMSO.

Frampton G K. 1994. Sampling to detect effects of pesticides on epigeal Collembola (springtails). Aspects of Applied Biology 37. Saniplitrg to Make Decisions. pp. 121-130.

Frampton G K, Cilgi T. 1992. Long-term effects of pesticides on arthropods in UK arable crops: preliminary results from the "SCARAB" project. Aspects ofApplied Biology 31, Interpretation of Pesti- cide Efyects on Beneficial Arthropods, pp. 69-76.

FramptonG K, Cilgi T. 1993. Boxworth and SCARAB: approaches to the long-term study of side effects of pesticide use on invertebrates. Proceedings A NPP-BCPC Second International Symposium on Pesticide Application Techniques. Strasboitrg, 22-24 September 1993, pp. 399406.

Frampton G K. Cilgi T. 1994. Long-term effects of pesticides on Carabidae in UK farmland: some initial results from the "SCARAB" project. In Carabid beetles - Ecologv and Eriolution, pp, 433438. Eds K Desender et ul. Dordrecht: Kluwer Academic Publishers.

Frampton G K, Langton S D, Greig-Smith P W, Hardy A R. 1992. Changes in the soil fauna at Bnxworth. In Pesticide-y, Cereal Farming and tke Eniiironment. pp. 132-143. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Francis C A. 1986. Multiple Cropping Systems. New York: MacMillan Publishing Co. Gerowitt B, Steinmann H -H. 1992. Weed research. In I N T E X : Developing an integrated farming


system and investigating ecological effects of an extensification in arable production - Field guide 1992, pp. 24-25. University of Gottingen, Germany: Research centre for Agriculture and the Environment.

Glen D M, Spaull A M, Mowat D J, Green D B, Jackson A W. 1993. Crop monitoring to assess the risk of slug damage to winter wheat in the United Kingdom. Annals ofApplied Biology 122:161- 172.

Goodlass G. 1991. Fertilizer planning for crop rotation. Proceedings of the Symposium by Novia Scotia Crop Development Institute, Crop Rotation and Resource Management, Nouia Scotia, February 25-

Greaves M P, Marshall E J P. 1987. Field margins: definitions and statistics. In Field Margins, pp. 3- 10. Eds M J Way and P W Greig-Smith. BCPC Monograph 35.

Greig-Smith P W. 1991. The Boxworth experience: effects of pesticides on the fauna and flora of cereal fields. In The Ecology of Temperate Cereal Fields, pp. 333-372. Eds L G Firbank, N Carter. J F Darbyshire and G R Potts. Oxford: Blackwell Scientific Publications.

Greig-Smith P W. 1992. Origin and aims of the Boxworth project. In Pesticides, Cereal Farming and the Environment, pp. 1-5. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Greig-Smith P W, Griffin M J. 1992. Summary and recommendations. In Pesticides, Cereal Farming and the Environment, pp. 200-215. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Greig-Smith P W, Hardy A R. 1992. Design and management of the project. In Pesticides, Cereal Farming and the Environment, pp. 6 1 8 . Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Greig-Smith P W, Frampton G K, Hardy A R. Eds. 1992. Pesticides, cereal farming and the enuironment. London: HMSO. 281 pp.

Hancock M, Frampton G K, Cilgi T, Jones S E, Johnson D B. 1994. Ecological aspects of SCARAB and TALISMAN studies. Proceedings of the 13th Long Ashton International Symposium: Arable Ecosystems for the 21st Century (In press).

Hani F. 1989. The Third Way, a research project in ecologically orientated farming systems in Switzerland, pp. 51-66. IOBC/ WPRS Bulletin 1989/X11/5. Wageningen, The Netherlands.

Hani F. 1990. Farming systems research at Ipsach, Switzerland - The “Third Way” project. Swiss Journal of Agricultural Science 29:257-271.

Hani F. 1993. Weiterentwicklung umweltschonender Bewirtschaftungssysteme - projekt ‘Dritter Weg’. Swiss Journal of Agricultural Science 32:341-364.

Hart A D M, Thompson H M, Fletcher M R, Greig-Smith P W, Hardy A R, Langton S D. 1992. Effects of summer aphicides on tree sparrows. In Pesticides, Cereal Farming and the Environment, pp. 175-193. Eds P W Greig-Smith, G K Frampton and A R Hardy. London:HMSO.

Hermans C, Vereijken P. 1992. Integration of animal husbandry and nature conservation on grassland farms. Netherlands Journal of Agricultural Science 40:301-314.

Hickman J, Wratten S D. 1994. Use of Phacelia tanacetifolia (Hydrophyllaceae) as a pollen resource to enhance hoverfly populations (Diptera:Syrphidae) in sweetcorn fields. IOBC/ WPRS Working Group ‘integrated control in cereal crops.’ Bulletin 1993 17(4): 156-167.

Holland J M. 1994. A MAFF/LINK project - integrated farming systems. IOBC/WPRS Working Group ‘Integrated control in cereal crops. ’ Bulletin I993 17(4):237-244.

Jarvis R H. 1992. Performance of crops and economics of production. In Pesticides, Cereal Farming and the Environment, pp. 50-59. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Jepson P C. 1989. The temporal and spatial dynamics of pesticide side-effects on non-target invertebrates. In Pesticides and non-target invertebrates, pp. 95-128. Ed. P C Jepson. Wimborne, Dorset: Intercept.

Johnson I P, Flowerdew J R, Hare R. 1992. Populations and diet of small rodents and shrews in relation to pesticide usage. In Pesticides, Cereal Farming and the Environment, pp. 144-156. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Jones S E, Jones A L, Johnson D B. 1993. Effects of differential pesticide inputs on the size and composition of soil microbial biomass: results from the Boxworth and SCARAB projects.

27th, 1991, pp. 69-71.


Proceedings of the lrzternational Symposium on Environmental Aspects of Pesticide Microbiology, Sigtuna, Sweden. pp. 3&36.

Jordan V W L. 1989. Farming systems research on arable crops in the UK. In Current status of integrated arable farming systems research in Western Europe, pp. 67-76. IOBC/WPRS Bulletin 1989!XII/S. Wageningen. The Netherlands.

Jordan V W L. 1990. Long Ashton low input farming and environment (LA LIFE). Swiss Journal of Agricultural Research 29:389-390.

Jordan V W L. 1992. Opportunities and constraints for integrated farming systems. In Proceedings of the 2nd E S A Congress, Warwick Universi t~. pp. 318-393.

Jordan V W L, Hutcheon J A. 1994~ . Less intensive integrated farming systems for arable crop production and environmental protection. Proceedings No. 346, The Fertilizer Society. 32 pp.

Jordan V W L, Hutcheon J A. 19943. Less-Intensive Farming and the Environment (LIFE): an integrated farming systems approach to arable production. Proceedings of the 13th Long Ashton lriternutional Symposium: Arable Ecosystems f o r the 21st Century (In press).

Jordan V W L, Hutcheon J A, Glen D M. 1993. Studies in technology transfer of integrated farming systems considerations and principles for development. Booklet. AFRC Institute of Arable Crops Research. Long Ashton. Bristol.

Jordan V W L, Hutcheon J A , Perks D A. 1990. Approaches to the development of low input farming systems. Crop Protection in Organic (itid Low Input Agriculture. BCPC Monograph 45, pp. 9-18. Suffolk: Lavenham Press.

Kendall D A, Brain P, Chinn N E. 1992. A simulation model of the epidemiology of barley yellow dwarf virus in winter sown cereals and its application to forecasting. Annals of Applied Ecology 29:414-426.

Marchant J H, Hudson R J , Carter S P, Whittington P A. 1990. In Population Trends in British Breeding Birds. Tring: British Trust for Ornithology.

Marshall E J P. 1992. Patterns of distribution of plant species in the fields and their margins. In Pesticides. Cereal Farming and the Enuironment. pp. 68-81. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Millington S, Stopes C, Woodward L. 1990. Rotational design and the limits of organic systems-the stockless organic farm. Crop Protection i n Organic and Low Input Agriculture. BCPC Monograph 45. pp. 163-173. Suffolk: Lavenham Press

Murphy M. 1990. Food consumption. free trade and demand for agrochemicals. Proceedings of the Brighton Crop Protection Conference - Pests and Diseases. pp. 457466.

Murphy M. 1991. An economic perspective on cereal farming, food consumption and the environment. In The Ecology of Temperate Cereal Fields. pp. 69-94. Eds L G Firbank. N Carter, J F Darbyshire and G R Potts. Oxford: Blackwell Scientific Publications.

Ogilvg S E, Green M R, Groves S J, Jones A E. 1993. SCARAB - the impact of less intensive herbicide use on the diversity and distribution of weed species in three arable rotations. Proceedings of the Brighton Crop Protection Conference - Weeds. pp. 1241-1246.

Parker M M. 1992. Sustainable agriculture and the market place. Food Matters Worldwide 17: 10. Perks D A, Lane P. 1990. Towards a low input system minimising agrochemicals and nitrogen

Potts G R. 1986. The Partridge: Pesricides, Predation and Conseruation. London: Collins. Potts G R. 1991. The environmental and ecological importance of cereal fields. In The Ecology of

Ternperate Cereal Fields. pp. 3-21. Eds L G Firbank. N Carter, J F Darbyshire and G R Potts. Oxford: Blackwell Scientific Publications.

Prew R. 1993, Development of integrated arable farming systems for the UK. Proceedings of the HGCA Cereafs R & D Conference. pp. 242-254.

Przemeck E, Lickfett T. 1992. Investigation of the soil nutrient budget. In I N T E X : Developing an integruted farrningsystem and investigating ecological effects of an extensification in arableproduction - Field guide 1992, pp. 14-15. University of Gottingen, Germany: Research centre for Agriculture and the Environment.

Raven P W J, Stokkers R . 1992. Development of integrated flower bulb production. Netherlands Joiirriul of Agricultural Science 40:25 1-36.

(TALISMAN). Swiss Journal of Agricultural Science 29:395-396.


Reganold J P, Papendick R I, Parr J F. 1990. Sustainable agriculture. Scientijic American, June 1990,

Risch S J, Andow D, Altieri M A. 1983. Agroecosystem diversity and pest control: Data, tentative

Roberts T. 1989. Pesticides in drinking water. GIFAP Bulletin 15(3):3.1. Schenk A M E, Wertheim S J. 1992. Components and systems research for integrated fruit production.

Netherlands Journal of Agricultural Science 40:257-268. Schmidt W, Waldhardt R. 1992. Geobotanical investigations. In INTEX: Developing an integrated

farming system and investigating ecological effects of an extensijication in arable production - Field guide 1992, pp. 22-23. University of Gottingen, Germany: Research centre for Agriculture and the Environment.

Sly J M A. 1981. Review of usage of pesticides in agriculture, horticulture and forestry in England and Wales 1975-1979. Pesticide Usage Survey Report 23. London: MAFF. 94 pp.

Sotherton N W. 1991. Conservation Headlands: a practical combination of intensive cereal farming and conservation. In The Ecology of Temperate Cereal Fields, pp, 373-398. Eds L G Firbank, N Carter, J F Darbyshire and G R Potts. Oxford: Blackwell Scientific Publications.

Sotherton N W, Boatman N D, Rands M R W. 1989. The ‘Conservation Headland’ experiment in cereal ecosystems. The Entomologist 108: 135-143.

Speight M R, Lawton J H. 1976. The influence of weed cover on the mortality imposed on artificial prey by predatory ground beetles in cereal fields. Oecologia 23:211-223.

Stippich G, Biichner S. 1992. Zoological investigations. In INTEX: Developing an integrated farming system and investigating ecological effects of an extensijication in arable production - Field guide 1992. pp. 20-21. University of Gottingen, Germany: Research centre for Agriculture and the Environment.

Tarrant K A, Thompson H M. 1992. Exposure of rabbits to aphicides. In Pesticides, Cereal Farming and the Environment, pp. 157-159. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Thomas M B, Wratten S D, Sotherton N W. 1991. The creation of ‘island’ habitats in farmland to manipulate populations of beneficial arthropods: predator densities and emigration. Journal of Applied Ecology 28:906917.

Thompson H M, Tarrant K A, Hart A D M. 1992. Exposure of starlings, house sparrows and skylarks to pesticides. In Pesticides, Cereal Farming and the Environment, pp. 194-199. Eds P W Greig- Smith, G K Frampton and A R Hardy. London: HMSO.

Vereijken P. 1985. Alternative farming systems at Nagele: preliminary results and prospects. In Sustainable agriculture and integrated farming systems 1984 conference proceedings. Eds C Edens, C Fridgen, S L Battenfield. East Lansing, USA: Michigan State University Press.

Vereijken P. 1986. From conventional to integrated agriculture. Netherlands Journal of Agricultural Science 34:387-393.

Vereijken P. 1989~. Experimental systems of integrated and organic wheat production. Agricultural Systems 30:187-197.

Vereijken P. 1989b. From integrated control to integrated farming, an experimental approach. Agriculture, Ecosystems and Environment 26:37-63.

Vereijken P. 1990. Research on integrated arable farming and organic mixed farming in the Nether- lands. Swiss Journal of Agricultural Research 29:249-256.

Vereijken P, Royle D J. Eds. 1989. Current status of integrated arable farming systems research in Western Europe. IOBC/WPRS Bulletin 1989/XII/5, Wageningen, The Netherlands. 76 pp.

Viaux P, Roturier C, Bouchet C. 1993. Integrated arable farming systems in France. Proceedings of Conference on Integrated Arable Farming Systems, Nitra, 18-20 August, 1993, pp. 24-33.

Viaux P, Fourgeroux A, Hilaire A, Cavelier A, Lescar L. 1989. Planning of integrated farming systems experiments in France, pp. 1620. IOBC/WPRS Bulletin 1989/XII/5. Wageningen, The Netherlands.

Vickerman G P. 1992. The effects of different pesticide regimes on the invertebrate fauna of winter wheat. Pesticides, Cereal Farming and the Environment, pp. 82-108. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Wall C. 1992. A LINK collaborative research programme on technologies for sustainable farming

pp. 72-78.

conclusions and new research directions. Environmental Entomology 12:625-629.

43 8 J M HOLLAND. G K FRAMPTON, T CILGl AND S D WRATTEN

systems. Proceedings of the Brighton Crop Protection Conference - Pests and Diseases, pp. 1107- 1114.

Wallin H, Ekbom B S. 1988. Movements of carabid beetles (Coleoptera: Carabidae) inhabiting cereal fields: a tracing study. Oecologiu 77:3943.

Wells G W H. 1992. Integrated production for glasshouse horticulture. Netherlands Journal of Agricultural Science 40:277-284.

Wijnands F G . 1990. Farming systems research in the Netherlands: perspectives for integrated arable and organic mixed systems. Crop Protection in Organic and Low Input Agriculture. BCPC Monograph 45, pp. 139-146. Suffolk: Lavenham Press.

Wijnands F G. 1992. Evaluation and introduction of integrated arable farming in practice. Netherlands Journal of Agricultural Science 40:239-250.

Wijnands F G, Vereijken P. 1986. Integrated cropping of winter wheat as an example of integrated farming. Proceedings of the Brighton Crop Protection Conference - Pests and Diseases, pp. 1235- 124 I .

Wijnands F G , Vereijken P. 1992. Region-wise development of prototypes of integrated arable farming and outdoor horticulture. Netherlands Journal of Agricultural Science 40:225-238.

Wildenhayn M. 1992. INTEX - Ein interdisziplinares Forschungsvorhaben zur Entwicklung integrierter Anbausysterne und zur Erfassung okologischer Wirkungen von ExtensivierungsmaBnahmen. Mit- reilltrigen der Gesellschaft ,fur Pffanzenbauwissetzschnfren 437-40.

Wolfe M S. 1990. Intra-crop diversification: disease, yield and quality. Crop Protection in Organic and LOM' l tpi i t Agriculture. BCPC Monograph 45, pp. 105-114. Suffolk: Lavenham Press.

Woodward L, Lampkin N. 1990. Organic agriculture in the United Kingdom. Crop Protection in Orgunic and Low Input Agriculture. BCPC Monograph 45. pp. 19-30. Suffolk: Lavenham Press.

Wratten S D. 1993. Compatible use of insecticides and beneficial insects: prospects for sustainable pest control. Proceedings ANPP-BCPC Second lnrernntional Symposium on Pesticide Application Techniques, Strasbourg, 22-24 September 1993. pp. 393-398.

Yarham D J, Symonds B V. 1992. Diseases. In Pesticides, Cereal Farming and the Environment, pp. 3749. Eds P W Greig-Smith, G K Frampton and A R Hardy. London: HMSO.

Zadoks J C. 1992. The costs of change in plant protection. Journal of Plant Protection in the Tropics,

(Received 18 July 1993) 9:151-159.

arable acronyms analysed – a review of integrated arable farming systems research in western...

Documents