Trees in European crop fields:determining the trade-offs in

profitability and ecosystem regulation

Presentation made by Paul Burgess, Cranfield Universityat Farm Woodland Forum meeting,

Gregynog Conference Centre, 29 June 2005

Paul Burgess1, Anil Graves1, João Palma2,Felix Herzog2 , Yvonne Reisner2, Gerardo Moreno3,

Mercedes Bertomeu3, Christian Dupraz4, Fabien Liagre5,Herman van Keulen6 , L.D. Incoll7, David Pilbeam7,

and Terry Thomas8

1 Cranfield University, Bedfordshire, England; 2FAL, Switzerland; 3Extremadura University, Spain, 4INRA, Montpellier, France;

5APCA, Paris, France; 6Wageningen University, the Netherlands,7 University of Leeds, UK; 8BEAM (Wales) Ltd, Anglesey, Wales

Trees in European crop fields:determining the trade-offs in

profitability and ecosystem regulation

Acknowledgements

This research was funded through the European Union 5th Framework through the contract QLK5-

2001-00560 and the Swiss Federal Ministry of Science and Technology contract 00.0158.

Outline

1. Trees in fields2. Method3. Production4. Profitability5. Ecosystem regulation6. Conclusions

There has been a loss of wooded landscapes over much of Europe...

Tanner (1993)

...trees have become segregated from agriculture...

~60% decline of field trees in England 1980-1998

.. but some farmers have maintained agroforestry systems ...

.. and others have established trees in fields recently

2. Method

How to determine profitability and environmental effects?

•Difficulty in waiting 60 years

•Relative to arable cropping and forestry

Plot-SAFE

Yield-SAFE

Use of models to determine environmental and economic effects

Economic

Biophysical

Plot-scale Farm-scale

Farm-SAFE

Biophysical data

NShoot number

BcBiomassCrop

LcLeaf area

SThermal time

Water content

Leaf area

Biomass

θSoil

Lt

BtTree

Yield-SAFE: state variables

Bio-physical model (daily time step)

Weatherdata

Soil parameters

Crop parameters

Tree parameters

Tree yield

Crop-management(i.e crop rotation, date of tree planting)

Crop yieldArable system yields

Forest system yields

Silvoarable system yields

Yield-SAFE structure

Plot-SAFE: structure

Economic model at a plot scale(annual time step)

Bio-physical model(daily time step)

Weatherdata

Soils parameters

Crop parametersTree parameters

Tree yield

Tree value

Plot-management(i.e discount rate, labour cost)

Crop-management(i.e crop rotation, date of tree planting)

Crop yield

Crop finance

Modified management

Arable system yieldsForest system yields

Silvoarable system yields

Tree grantsTree costs

Summary of outputs

Farm-SAFE: structure

Forestry system

Arable system

Crop finance

Tree value

Tree grants

Tree costs

Plot-scale arable economics

Plot scale forestry economics

Plot scale agroforestry economics

Plots 1 - 4 Farm

Farm-scale forestry economics

Farm-scale agroforestry economics

Options and results

Farm-scale arable economics

Farm- scale results

Plot-scale results

Sensitivity analysis

PlantingSystem selection

Silvoarable system

Biophysical data

Financial data

Description and software

Calibrated output: forestry growth

Potential poplar growth in the Atlantic region simulated with Yield-SAFE compared to data from yield tables

0

200

400

600

0 2 4 6 8 10 12 14 16 18 20Time from tree planting (a)

Volu

me

(m3 h

a-1)

Calibrated output: wheat yield

Total wheat biomass predictions from Yield-SAFE calibrated to output from the comprehensive crop growth model STICS

0

10

20

30

0 100 200 300 400Time from planting (d)

Bio

mas

s (t

ha-1

) Yield-SAFE

STICS

Validation: UK Silvoarable Network

Site calibration for SilsoeThe model was calibrated for tree-only andcrop-only reference yields by modifying the transpiration coefficient and the harvest index

-2.4Timber yield (m3 tree-1) (30 years)8.2-Crop yield (t ha-1 a-1)

49

0.42

Poplar

51Harvest index (%)

0.32Transpiration coefficient (m3 kg-1)

Wheat

Validation: timber volume per poplar tree at Silsoe

0.0

1.0

2.0

0 10 20 30

Time from planting (a)

Tim

ber v

olum

e (m

3 tr

ee-1)

Yield-SAFE prediction Silsoe measurementsPoplar: yield class 13

Validation: relative crop yield with poplar agroforestry in the UK

0.0

0.5

1.0

0 10 20 30

Time from planting (a)

Rel

ativ

e cr

op y

ield

Relative arable yield Yield-SAFE predictionSilsoe measurements Leeds measurements

Conclusions: models

1. Validated daily-time-step biophysical model of the yields of arable, silvoarable and forestry systems over a tree rotation (up to 60 years)

2. Lack of data describing tree growth at low stand densities constrained model validation

3. Start simple

3. Production

Predicted effect of tree stand densityon timber production per tree

0.0

0.5

1.0

1.5

2.0

0 20 40 60

Time from tree planting (a)

Tim

ber v

olum

e (m

3 tr

ee-1

)

50 trees/ha 113 trees/ha Forestry

Champlitte

Predicted effect of tree stand density on timber production per hectare

0

50

100

0 20 40 60

Time from tree planting (a)

Pred

icte

d st

and

volu

me

(m3

ha-1

)

Forestry 50 trees/ha 113 trees/ha

Champlitte

Predicted effect of tree stand density on relative yields of a wheat/wheat/oilseed

rotation

0.0

0.2

0.4

0.6

0.8

1.0

0 10 20 30 40 50 60

Time from tree planting (a)

Rel

ativ

e cr

op y

ield

Arable 50 trees/ha 113 trees/ha

Champlitte

113 trees/ha

50 trees/ha

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Relative crop yield over tree rotation

Rel

ativ

e tr

ee y

ield

ove

r tre

e ro

tatio

n

Effect of tree density on relative productivity

Spain France

Champlitte

Production determined for selected locations, tree and crop species in Europe

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Relative crop yield over tree rotation

Rel

ativ

e tr

ee y

ield

ove

r tre

e ro

tatio

n

Poplar Cherry Oak Pine Walnut

Integrating trees and crops can result in improved site productivity

Spain France

4. Profitability

Effect of system on profitability

Champlitte

-200

0

200

400

600

800

No grants With grants

Equi

vale

nt a

nnua

l val

ue(4

% d

isco

unt r

ate)

(€ h

a-1 a-1

)

Arable Silvoarable Forestry

Effect of system on profitability

Champlitte

-200

0

200

400

600

800

No grants With grants

Equi

vale

nt a

nnua

l val

ue(4

% d

isco

unt r

ate)

(€ h

a-1 a-1

)

Arable Silvoarable Forestry

Profitability, with grants, of arable and an agroforestry system (113 trees/ha) at Champlitte

Agroforestry€434 ha-1 a-1 Arable

€473 ha-1 a-1

Champlitte

Farm-scale analysis (2005 grants; 4% discount rate)

0

20

4060

80

100O

ak

Pine

Che

rry

Wal

nut

Popl

ar

Tree species

Prop

ortio

n of

cas

es(%

)Agroforestry more

profitable than forestryAgroforestry more

profitable than agriculture

0

20

40

60

80

100

Oak

Pine

Che

rry

Wal

nut

Popl

ar

Tree species

Conclusions: production and economics

Agroforestry results in greater site productivity than growing trees and crops separately

In France, agroforestry• is often the most profitable way of

establishing walnut, cherry and poplar in arable areas

• with walnut and poplar is competitive with arable monocultures

• with current grants, cherry is less profitable than arable systems

5. Environmental regulation

• Soil erosion • Groundwater recharge• Nitrogen leaching• Carbon sequestration

Predicted effect of agroforestry and contours on soil erosion at Champlitte

0.0

0.2

0.4

0.6

0.8

Non-contour Contour

Pred

icte

d ac

tual

soi

l los

s (t

ha-1

a-1

)

Arable 50 trees/ha 113 trees/ha

Reduced soil erosion when

combined with contours

Champlitte

Predicted effect of agroforestry on groundwater recharge at Champlitte

0

200

400

600

0 10 20 30 40 50

Gro

undw

ater

flow

(m

m a

-1)

0

600

1200

Prec

ipita

tion

(mm

a-1

)

Arable 113 trees/ha Precipitation

Champlitte

Predicted effect of agroforestry on groundwater recharge at Champlitte

0

50

100

150

Year 1-20 Year 21-40 Year 41-60Mea

n an

nual

gro

undw

ater

rech

arge

(mm

)

Arable 50 trees/ha 113 trees/ha

Champlitte

Predicted effect of agroforestry on nitrogen leaching at Champlitte

0

50

100

0 10 20 30 40 50Time from tree planting (a)

Nitr

ogen

leac

hing

(kg

ha-1

a-1

)

Arable 113 trees/ha

Champlitte

Predicted carbon sequestration over 60 years using wild cherry at Champlitte

0

50

100

Arable 50trees/ha

113trees/ha

ForestPred

icte

d ca

rbon

seq

uest

ratio

n (t

ha-1

)

Champlitte

Conclusions: EnvironmentA biophysical model was used to predictkey environmental effects of an arable monoculture and agroforestryat different tree densities,however validation is required

As tree density increased, predicted:–soil erosion decreased–groundwater recharge decreased–nitrogen leaching decreased–carbon sequestration increased

Trade-offs in profitability and environmental regulations

Agroforestry€434 ha-1 a-1 Arable

€473 ha-1 a-1

Reduced soil loss Groundwater recharge

Reduced flooding

Reduce nitrates

Agroforestry€434 ha-1 a-1

Arable€473 ha-1 a-1

Reduced soil loss

Reduce nitratesC sequestration

Groundwater recharge

Reduced flooding

Trade-offs in profitability and environmental regulations

Agroforestry€434 ha-1 a-1

Arable€473 ha-1 a-1

Reduced soil loss

Reduce nitratesC sequestration

Groundwater recharge

Reduced flooding

Trade-offs in profitability and environmental regulations

Conclusions: determining trade-offsAlthough agroforestry may not be the most profitable option to an individual farmer, it may still be the best option for society.

A bio-economic model, linked to some environmental models, can be used to compare key economic and environmental effects of arable, forestry and agroforestry systems