Yield Gap Studiesthrough Comparative Performance

Analysis (CPA)

Presented are :

Current problems, mandate and methods.

Concepts to study sustainability aspects of

agricultural land use systems.

A case study to demonstrate CPA.

Dr. C.A.J.M. de BieITC, Enschede, The Netherlands

 Commission VII, Working Group VII/2.1 on Sustainable Agriculture

Pre-Symposium Tutorial

Title

Good yields

Nutrient removals

Soil degradation

Falling Yields

The Poverty Trap

High population

Shorter or no fallow periods

Low yields

Persistent soil degradation

Expansion onto unsuitable soils

The downward spiral to the poverty trap

Increasing population

Increasing population

Current problems, mandate and methods

1.Poverty Trap

Trends

Problems

Needs

Increase in cultivated area

Intensification (crops/ha; stocking density)

Heavier natural vegetation exploitation

Competition for same tract of land

Declining yields

Land and natural vegetation degradation

Assessment which land uses are relevant for which tracts of land and present needs

Assessment by land use type, which management minimizes environmental impacts while maximizing productivity

Land Use

Declining yields

Land and natural vegetation degradation

Assessment by land use type, which management minimizes environmental impacts while maximizing productivity

Sustainability

Study Goals

Trends

Fact: Detailed and reliable quantitative information on present landuse is scarce and often of low quality.

We need good land use data :

We need practical concepts and approaches :

to gather, manage, classify and map land use information.

to study various aspects of present day land use systems.

to address questions as put on record by the UNCED conference in Rio (1992; Agenda 21, Chapter 10), e.g.:

to identify options to solve future food requirements.

to understand and combat environmental degradation.

Our Mandate to study “Sustainable Agriculture”.

Mandate

Land Characteristics

Purpose(s)

Factor RatingsCropRequire-

ments s1

The Land UseRequirements

(LURs)

s2 s3 n

LQ’s

(de

ma

nd

sid

e)

diag

nost

icfa

ctor

s

one

tabl

e fo

rea

ch L

UT

Yield ranges

The LandQualities (LQs;

supply side)

Land

Uni

ts

Matching

Current “Land Evaluation” study method.

Based on generic crop-specific factor rating tables.

Typically ignores management effects.

Does not truly evaluate land use.

Evaluates “crops”.

Lack

of

met

hod ?

LE

Yield limiting / reducing factors not controlled at the researchstation

Yield Gap-0

Non-transferable technologyEnvironment and management constraints

Yield Gap-1

Market access Prices

Diminishing returnsGap-2A

Lack of inputsfarmers’ riskaversion stra-tegies

Gap-2

Gap-2B

Simulatedpotential‘researchstation’

yield

Experi-mental

maximum‘researchstation’

yield

Technical‘on-farm’

ceiling(= Potentialfarm yield)

Economic’on-farm’ceiling Actual

‘on-farm’yield

This level may differconsiderably from

plot to plot.

Current “Yield Gap” study entries.

Considers levels ‘fixed’; it omits variability between sites/areas.

Yield Gap

Current “Management-Yield” study approaches.

Based on technology transfer.

Focused on only few management aspects at a time.

Expensive.

Based on knowledge.

Management typically excluded.

Limited operational use.

Lack

of

met

hod ?

Conv.-Simul.

Bio-Physical Conditions

Socio-Economical Conditions

Land Use System

Other Land Use Systems

Livestock Systems

Context Goals

Inputs / Implements

Outputs /

Benefits

Soil / Terrain Climate / Weather Vegetation (Crops / Flora)

Wildlife (Fauna)

Infrastructure Operation Sequence

Land

Land Use Purpose(s)

Land Use

Land User(s)

Impact on land ( + or - )

Decision making / planningRequirements &

Suitability

Productivity

Impact on/from the environment

Interaction with secondary production

systems

The Concepts

The “Land Use System” (LUS) with ‘study entries’.

2.The Concepts

Failure to distinguish between the two has created much confusion.Land cover is a part of land, whereas land use is not.

Some examples: The cover "forest" is identified by its physical components such as

vegetation structure, height, density, and extent. The use of “forest” is dictated by its purpose(s) like: rubber production, conservation, recreation, timber production, or shifting cultivation.

The land cover “grassland”, distinguished by the presence or dominance of grass (herbaceous vegetation) may be used for hay production, grazing, recreation, etc.

Land cover is defined as: "The vegetation (natural or planted) or man made constructions (buildings, etc.), which occur on the earth surface … …”.

Land use is defined as: “A series of operations on land, carried out by humans, with the intention to obtain products and/or benefits through using land resources”.

Land cover can be determined by direct observation.Obtaining land use information requires communication with the land user.

What is Land Cover and what is Land Use ?

Cover/Use

- climate- weather

- landform- terrain- soil

- flora (incl.crops)

- fauna

- results of past land use (incl. infrastructure)

Implementsused

InputsOutputs

Land Use System

Land Land Use

Purpose(s)

Aimed at[Species/Services

-Products/Benefits]

combinations.

Operation

Sequence

Details onOperations

Impact on land( + or - )

Repetition …

Land problems: limits growth reduces yields

Land use aims to modify land to reduce land problems

Repetition

Operation Sequences

Grazing Fallowing

1989198819751969 1979

Rainfed Cropping

J F M A M J J A S O N D

1988 1989

Observations

Operations

… many aim to control growth limiting, and yield reducing land aspects.

… many relate to growth limiting, and yield reducing land aspects.

Ploughing Harvesting Fallow

Pest AttackGermination

Trampling Hail Storm

Rill Erosion

WeedingSeeding

NPK Applic.Illustrating land use

operations

and land obser-vations

The “Operation Sequence” impacts on ‘sustainability’ aspects.

Land Use

Land

Land Use System

Oper.Seq.

Yie

ld

they address: growth limiting yield reducing land modifying aspects of LUSs.

Feasible

Problems

Management

Plot-to-plot variability

ProblemsProblemsProblemsProblemsProblems

What do sustainability studies do ?

CPA relates differences in land and management aspects to differences in system performances.

CPA uses survey data from many plots.

CPA studies this gap.

Sust.Studies

Livestock ProductionSystem(s)

Agricultural Holding

(in Administrative Unit-A)

Plot Plot

PlotPlotPlot

Land

use-1

Parcel-1 Parcel-2 Parcel-3

Soil/Terrain Unit-1 Soil/Terrain Unit-2

Agro-Ecological Zone

HouseholdSystem

Other ActivitiesIncl. off-farm

Land

use-2

Land

use-4

Land

use-3

Land

use-5

A Plot defines spatially a unique Land Use Systems.

Holding

Flow of Information Flow of Materials

tenancy arrange- ments

biophysical circumstances

inputs and

outputs

goals, labour / capital / inputs and implements availability, knowledge, flexibility, awareness, social acceptance, risk perception, etc.

Farm System

socioeconomic circumstances

Decisions

by Holder

Household System

Other Activities

Land Use System - 1

Land Use System - 2

Livestock Production

Systems

Components

A Farm is the unit that controls several Land Use Systems.

F.System

Date of image

Land use is mapped using crop calendar data and RS-images.

Crop Calendar

Interview data

Observation by surveyor

Photo / Image characteristics 1D-features (tone, color), as related to:

crop calendars, cropping patterns and other land use operations

Infrastructure

2D-features, such as:

field sizes, shapes and patterns

internal patterns (textures, grids, mottles)

line features

3D-features (on APs):

vertical structure

no. of layers

holding/holder information (profile)

site aspects (tenancy arrangement, cadastral no., distance to holding, infrastructures present)

land use system (plot) aspects for the period considered:

a-priori land use class

crops grown / services provided (% of area, numbers, etc.)

land use purposes

operations (crop calendar / cropping pattern):

operation name; species involved; % of plot involved; period / periodicity / duration and task times; main power source

labor and material inputs and implements used

products / benefits obtained

observations by land user:

soil related (workability, infiltration rate, fertility status, etc.)

weather related (hail storm, dry period, etc.)

crop related (pests, diseases, lodging, wilting, etc.)

plot size, coordinates, slope, position, etc.

crops (residues) and infrastructure present in / around the plot

land cover data (crop condition, growing stage, weed incidence, biomass, height)

ground cover status (bare soil, mulch, crop residues)

specific observations (soil characteristics, tillage condition, erosion status, hydrological aspects, pests / diseases incidence, evidence of grazing)

Relevant land use data to survey.

Survey Data

Surveydata

Interviewdata

Performance = f ( land, land use )

Function to analyze the surveyed Land Use data.

Data on the operation sequence

Observations on land (crop) by the surveyor / land user

Yield

Impact on lande.g. erosion, bulk density, or salinity data

Math Function

A CPA example.

The impact of land use and land aspects on yields of sticky rice in Phrao, Thailand.

Yield reported by farmers (21x) were vali-dated through crop-cuttings.

0 1000 2000 3000 4000 5000 6000kg / ha

0

5

10

15

20

Co

un

t

Yields of 63 paddies variedfrom almost ‘0’ to almost 5500 kg/ha.

The CPA study did explain 83% of the shown yield variability.

Reported Yld. = 0.915 x Crop cutting Yld. (R-Sq=73%)

Correlation Coeff. = 85%

0

1000

2000

3000

4000

5000

6000

0 1000 2000 3000 4000 5000 6000 Crop Cutting Yield (kg/ha)

Rep

ort

ed Y

ield

(kg

/ha)

1:1 line

regression line

The yields

3.CPA Example

0

50

100

150

200

250

300

Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr

Sunshine duration in Chiang Mai (h/month; 1954-80)

Precipitation in Phrao (mm/month; 1952-85)

ET-pan in Chiang Mai (mm/month; 1965-80)

P

ET-pan

Humid period

Sunshine duration

Multi-Temporal TM image analysis…………Climate represented in a P-ETo diagram.

The multi-temporal NDVI image is used to map different crop calendars (next slide…).

Multi-Temp.

5000 7500 520000 2500 5000

2500

5000

7500

2140000

2500

5000

7500

2150000

Paddies with low cropping intensity (one crop per annum)

Paddies with high cropping intensity(two or three crops per annum)

Villages

In the Phrao floodplain water flows freely across bunded paddies.

Areas with a high cropping intensity could be mapped by using the 3 TM-images.

Distance from weirs strongly affects the cropping intensity, drought risk, and yields.

If water is available, a second or third crop can be grown after rice.

The fa

cts

Phrao Map

160

Ploughing

Puddling

Levelling (harrowing)

Trans-planting

Chemical weeding (39x)

(Day no.) 230 300 370

NPK-Application(42x)

Harvesting

(9 Jun) (18 Aug) (27 Oct) (5 Jan)

Avg.date

9 July

13 July

16 July

8 Aug.

13 Aug.

5 Sep.

7 Dec.

By site, data on the operation sequence were recorded.

Shown are the starting dates by operation.

Oper.Seq.

Y N N N 9 2893

Y N Y N 2 2109

Y Y N N 3 2069

N Y N N 12 2605

N Y Y N 5 2019

N N Y N 13 2101

Period of Water ShortageCount

Average yields(kg/ha)

Establish-ment

Vegeta-tive

Heading + Flowering

Yield formation

N N N N 19 3932

Impact of Water Shortage on Yield.

NN NN NN NN 1919 39323932

H2O-Shortage

Impact of Selected Pests and Diseases on Yield.

Rice Blast 63 -64 None Yes

Leaf blight 10 -34 37% with RB Exclude

Brown Leaf Spot 18 -17 None Yes

False smut 41 -12 20% with RB Exclude

Stem Borers 17 -12 None Try

Black bugs 7 -8 None Try

Sheath rot 12 0 - No effect

Foot rot 2 0 - No effect

Infection rates (%)

Pests / Disease Freq.Correlationwith yields

Inter-CorrelationsUse in

M. Regr. ?

Pest/Diseases

Period of lodging Freq.Average riceyields (kg/ha)

Average lodging-%

During Heading/Flowering 15 2300 34 (7.5-60.0)

During Yield Formation 9 1988 21 (12.5-27.5)

During Ripening 8 3585 20 (12.5-27.5)

None 31 3187 0

Impact of Lodging on Yield (severity and impact).

Lodging

180 190 200 210 220 230 240 250Planting date (day of the year)

90

100

110

120

130

140

150

160

Leng

th c

rop

gro

win

g p

erio

d (d

ays)

90 100 110 120 130 140 150 160Length crop growing period (days)

0

1000

2000

3000

4000

5000

6000

Yie

ld (

kg/h

a)

Delayed planting of sticky rice is associated with shorter crop-growing periods that cause in turn lower yields.

The crop is photo-sensitive. Other factors cause the ‘noise’.

Planting Date

Linear Multiple Regression 8 steps model

Dependent Variable = Rice Yield (kg/ha) S.E. = 482 N = 63 Adj.R2 = 84% Stepwise forward solution Coeff. Prob.

Constant: 2283 R2 when Independents entered

Incidence of Rice Blast (%) 41.3 -43.22 0.0% If water shortage during Heading/Flowering 61.1 -607.66 0.0% Lodging at Yield Formation stage (%) 69.1 -52.93 0.0% If 3 sequential crops grown (H2O-avail.) 74.7 937.76 0.0% Lodging at Heading/Flowering stage (%) 79.1 -13.58 0.1% If the farmer considers his soil "good" 81.7 386.75 0.4% Incidence of Brown Leaf Spots (%) 84.1 -32.65 0.5% Length of the crop growing period (days) 85.7 13.01 1.7%

Model estimation through multiple regression.

Model Estim.

measured values (v)

c x v

Independents coeffi-cient

(c)

av

g.

be

st

av

g.

be

st

Yie

ld g

ap

constant 2283 1 1 2283 2283

If 3 sequential crops grown (H2O avail.) 937 0.079 1 74 938 864

Length of the crop growing period (days) 13.01 120 157 1572 2043 471

Incidence of Rice Blast (%) - 43.22 15.8 5 - 681 - 216 465

If the farmer considers his soil "good" 386 0.44 1 172 387 215

If H2O shortage during Heading/Flowering - 607.66 0.32 0 - 193 0 193

Lodging at Yield Formation stage (%) - 52.93 3.0 0 - 158 0 158

Lodging at Heading/Flowering stage (%) - 13.58 8.1 0 - 110 0 110

Incidence of Brown Leaf Spots (%) - 32.65 3.2 0 - 104 0 104

Estimated yields (kg/ha): 2856 5434

Actual yields (kg/ha): 2855 5437

Estimated yield gap (kg/ha): 2578

Expected yield at Sampatong Rice Research Station : 4378

Potential yield at Sampatong Rice Research Station : 6253

Quantified break-down of the yield gap by constraint.

Yield Gap

0

1000

2000

3000

4000

5000

6000

7000Y

ield

(k

g/h

a)

farmers fields research station

weedingsoil typevarietylodgingwater stressdiseases

timely plantingcropping system

Potential Yield

Yie

ld G

ap

Actual Yield

Average Yield

Through regression analysis:• the yield constraints were identified,• and impacts by constraint quantified …

The re

sults

Bar Diagram

Relevant land characteristics:

The yield constraints and their relative importance:

Advise to local organizations:

Final statements on the yield constraints.

• water shortage (41%)

• diseases incidence (22%; rice blast / brown leaf spot)

• late planting (18%)

• lodging (10%)

• poor soil condition (8%)

• Water-loss from paddies, defined by the soil infiltration rate (= site specific)

• Water shortage, defined by the distance from weirs (= map unit specific)

• Plant breeders must concentrate on resistance to drought, diseases, and lodging.

• Extension services are best concerned with water management, timely planting, and control of diseases.

The re

sults

Conclusion