Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Enhancing and Extending Enhancing and Extending Field Research Field Research
with Modeling of Agricultural Systemswith Modeling of Agricultural Systems----International CollaborationsInternational Collaborations
L. L. AhujaAhuja , L. Ma, S. , L. Ma, S. AnapalliAnapalli , and J. , and J. KoKo
USDAUSDA--ARS, Agricultural Systems Research ARS, Agricultural Systems Research
and Colorado State Univ., Fort Collins, COand Colorado State Univ., Fort Collins, CO
Need for Quantitative System Need for Quantitative System
Approaches in AgricultureApproaches in Agriculture
��Agriculture in the 21Agriculture in the 21stst century is much more complex century is much more complex due to: environmental concerns, limited water, climate due to: environmental concerns, limited water, climate change: droughts & uncertainty, global competition; change: droughts & uncertainty, global competition; biobio--energyenergy
�� Integrated and quantitative system approaches are Integrated and quantitative system approaches are needed as planning & decision tools for optimal needed as planning & decision tools for optimal management, & to help research develop them. System management, & to help research develop them. System models provide these approaches. models provide these approaches.
��Enhances understandingEnhances understandingof the experimental results & of the experimental results & complex interactions; cause & effect relationscomplex interactions; cause & effect relations
��Enables their synthesis, quantification & Enables their synthesis, quantification & extends extends resultsresultsto longer time periodsto longer time periods
��Helps Helps transfer resultstransfer results, their optimal application to other , their optimal application to other soils/ climates, and soils/ climates, and aids managementaids management
�� Identifies knowledge gapsIdentifies knowledge gapsto focus further research; to focus further research; reduces duplicationreduces duplication
��Good field data help Good field data help improve the modelsimprove the models
Integrating Field Research with System Models Helps Both
CUTTING EDGE CUTTING EDGE
FIELD FIELD
RESEARCHRESEARCH
SIMPLER DECISION SIMPLER DECISION
SUPPORT SYSTEMSSUPPORT SYSTEMS
for Farmers & for Farmers &
Ranchers, Ag Ranchers, Ag
Consultants, and Consultants, and
Action AgenciesAction AgenciesPROCESS MODELS PROCESS MODELS
OF AGRICULTURAL OF AGRICULTURAL
SYSTEMSSYSTEMS
Extend Research & Extend Research &
Applications Applications
THE GUIDING PRINCIPLE & VISIONTHE GUIDING PRINCIPLE & VISION
The CGIAR Science Council The CGIAR Science Council (2005) Research Priorities:(2005) Research Priorities:
•• ““ Modeling and the ability to combine data from Modeling and the ability to combine data from different sources, different sources, ……promises to revolutionize promises to revolutionize understanding of processes affecting management of understanding of processes affecting management of natural resources.natural resources.””
•• ““ Thanks to strategic accumulation of data, tools, and Thanks to strategic accumulation of data, tools, and modeling resources in the coming decade, one can modeling resources in the coming decade, one can expect the development of a more predictive approach expect the development of a more predictive approach to agricultureto agriculture”” . .
•• Policy to focus limited resources on a few Policy to focus limited resources on a few comprehensive field studies and then use models to comprehensive field studies and then use models to extend them to other locations and countries. extend them to other locations and countries.
Agricultural Agricultural SystemsSystemsResearchResearchUnitUnitFort Collins, ColoradoFort Collins, ColoradoU.S.A.U.S.A.
OUR UNITOUR UNIT ’’ S RESEARCH MISSIONS RESEARCH MISSION
Develop whole-system approaches to help optimize resource management & evaluate/ develop sustainable agricultural systems:
• Synthesis of disciplinary knowledge to the whole system level and collaborative research to fill knowledge gaps.
• Computer models of agriculture systems to help research, site-specific management, and create simpler decision aids.
• Decision support technology packages for farmers, ranchers, ag consultants and action agencies for planning and management.
• Techniques for more efficient development & maintenance of models & decision tools
More Recent Major Team ProductsMore Recent Major Team Products
•• The Root Zone Water Quality Model (RZWQM) to simulate The Root Zone Water Quality Model (RZWQM) to simulate management effects on water, water quality and crop management effects on water, water quality and crop production. production. Updated to RZWQM2Updated to RZWQM2--DSSAT & RZWQM2DSSAT & RZWQM2 --GIS GIS for precision spatial management & conservation effects assessment--CEAP .
•• GPFARM, a simpler whole farm/ranch decision support system GPFARM, a simpler whole farm/ranch decision support system for strategic planning. for strategic planning.
•• Object Modeling System: Create models from library of standObject Modeling System: Create models from library of stand--alone modulesalone modules
Root Zone Water Quality ModelRoot Zone Water Quality ModelModeling Management Effects on Water, Modeling Management Effects on Water,
Water Quality and Crop productionWater Quality and Crop production
Water Resources Publications, LLC
Distinguishing Features of Distinguishing Features of RZWQMRZWQM
• Agricultural management practices and their integrated effects on water, crop production, and environmental quality ( tillage, irrigation, fertilization, manure application, tile drainage, pesticide application, and crop rotation).
• Macropore/preferential flow.• Water table fluctuation and tile flow.• Chemical transport in runoff/percolation water.
Distinguishing Features Distinguishing Features of RZWQMof RZWQM
•• Detailed carbon/nitrogen dynamics with Detailed carbon/nitrogen dynamics with consideration of microbial populations.consideration of microbial populations.
•• Multiple year simulation for crop rotations with Multiple year simulation for crop rotations with capability of answering capability of answering ““whatwhat--ifif”” scenariosscenarios
•• Detailed cropDetailed crop--specific models from DSSAT specific models from DSSAT packagepackage
RZWQMRZWQM--RZWQM2 ApplicationsRZWQM2 Applications
�Extensively used in U.S & other countries to evaluate water quality/quantity impacts of ag management & develop sustainable systems.
�Adopted by EPA and used by pesticide
industry for pesticide registration
�Used by USGS for NAWQA program�China: water & N could be reduced by 50% w/o reducing corn yield
�Continues to play an important role in our new research projects.
A wholeA whole--farm decision farm decision support system for strategic support system for strategic planning: evaluation of planning: evaluation of alternate cropping system, alternate cropping system, rangerange--livestock systems, livestock systems, and integrated farming and integrated farming options for production, options for production,
GPFARM: A Farm Level DSSGPFARM: A Farm Level DSS
economics, and environmental impactseconomics, and environmental impacts
End UsersEnd Users: Farmers and Ranchers, Consultants, : Farmers and Ranchers, Consultants, Action Agencies, Extension, and ScientistsAction Agencies, Extension, and Scientists
GPFARM Applications
� Several invited presentations to Colorado Conservation Tillage Association & farmers
� MOU with CAWG: GPFARM distributed to 600 members; trained 150 members
� GPFARM-Range model has been extensively used for synthesizing research data from three range research stations in the Great Plains
The Object Modeling System The Object Modeling System (OMS) (OMS)
An Object Modeling System consists of a library of modules which facilitates the assembly of a modeling package, tailored to the problem, data constraints, and scale of application.
Collaborators: ARS, NRCS, USGS, Friedrich-Schiller University, Jena, GermanyHas been adopted by NRCS as a uniform system to deliver conservation technology and is being used to develop a new field to watershed scale model.
Agricultural Systems Research Unit
Model and Field Research a Precursor to Model and Field Research a Precursor to DcisionDcision Support ToolsSupport Tools
The Object Modeling System (OMS) facilitates the development of component-based models which benefits decision support tool development.
CEAP Watershed Compared to SWAT
• The semi-distributed SWAT concept considers distributed information within a sub-basin only statistically but not in terms of location.
• Important processes, e.g., lateral water /nutrient transport and specific management in some parts of a sub-basin cannot be simulated.
The fully distributed CEAP Watershed Modelconcept allows theconsideration of suchprocesses.
Recent Examples of Model Recent Examples of Model Application to Enhance and Application to Enhance and
Extend field ResearchExtend field Research
1. Water Quality Studies1. Water Quality Studiesin Tilein Tile--Drained Cropping Systems Drained Cropping Systems
(Nashua, Iowa) (Nashua, Iowa)
•• 36 136 1--acre plots in Nashua, IA under tile drainage (at 120 cm acre plots in Nashua, IA under tile drainage (at 120 cm depth); Variety of data over time from 1978 to 2003depth); Variety of data over time from 1978 to 2003
•• Two crop rotations: continuous corn (CC) and cornTwo crop rotations: continuous corn (CC) and corn--soybean (CS) soybean (CS) rotations rotations
•• Four tillage systems: moldboard plow (MP), ridge till (RT), chiFour tillage systems: moldboard plow (MP), ridge till (RT), chi sel sel plow (CP), and noplow (CP), and no--till (NT)till (NT)
•• FertlizersFertlizers: anhydrous ammonium (AA) from 1977 to 1993; UAN : anhydrous ammonium (AA) from 1977 to 1993; UAN from 1993 to 1998from 1993 to 1998
•• Swine manure from 1998 to 2003: fall & spring applications Swine manure from 1998 to 2003: fall & spring applications
Experimental DesignExperimental Design
1975 1980 1985 1990 1995 2000 2005
Ave
rage
Yie
ld (Kg
ha-1)
0
2000
4000
6000
8000
10000
12000
14000 Measured corn yieldSimulated corn yieldMeasured soybean yieldSimulated soybean yield
1988 1990 1992 1994 1996 1998 2000 2002 2004
Ave
rage
Yea
rly T
ile F
low (cm
)
0
10
20
30
40
50Measured yearly tile flowSimulated yearly tile flow
1988 1990 1992 1994 1996 1998 2000 2002 2004Ave
rage
Yea
rly N
load
(kg
N h
a-1)
0
20
40
60
80
100
Measured yearly N loadSimulated yearly N load
Corn: RMSE=1223, r2=0.52 Soybean: RMSE=626, r2=0.37
RMSE=4.9, r2=0.74
Yearly N load: RMSE=12.6, r2=0.71
Year
1988 1990 1992 1994 1996 1998 2000 2002 2004
Flo
w
Wt C
onc.
(m
g N L
-1)
0
20
40
60
80
100
Measured flow weighted N Conc.Simulated flow weighted N Conc.
Flow weighted N Conc.: RMSE=5.4, r2=0.69
Synthesis of Synthesis of Data:Data:
Average values Average values across across treatmentstreatments
Using Validated Model to Extend Results to Using Validated Model to Extend Results to multiple years and Create multiple years and Create
Simpler Decision ToolsSimpler Decision Tools
Simulated yearly water balance and crop production averaged overSimulated yearly water balance and crop production averaged over 2424--yr for different crop yr for different crop rotation, tillage, and drainage scenarios. Results for cornrotation, tillage, and drainage scenarios. Results for corn --soybean rotation were taken as soybean rotation were taken as
averages from CS and SC phases of the rotation (Ma et al., 2007baverages from CS and SC phases of the rotation (Ma et al., 2007b ).).Scenarios Drain
flow(cm)
Lateral** flow(cm)
Runoff(cm)
ET*** (corn, cm)
Corn yield(kg/ha)
Corn biomass(kg/ha)
ET (soybean, cm)
Soybean yield (kg/ha)
Soybean biomass (kg/ha)
CC-NT-FD* 12.2 13.2 6.8 57.0 7878.3 18426.0 --- --- ---
CC-MP-FD 10.6 11.4 6.8 60.4 7862.2 18384.4 --- --- ---
CC-CP-FD 11.7 12.7 6.8 58.0 7886.9 18445.1 --- --- ---
CC-NT-CD 8.7 15.2 7.4 57.8 7920.3 18433.5 --- --- ---
CC-MP-CD 7.2 13.4 7.3 61.3 7908.0 18382.9 --- --- ---
CC-CP-CD 8.3 14.7 7.4 58.8 7921.7 18433.2 --- --- ---
CS-NT-FD 13.8 15.5 6.9 57.6 7879.7 18410.2 47.9 2971.5 8190.5
CS-MP-FD 12.8 14.4 6.9 58.9 7915.0 18443.1 50.9 3052.2 8544.2
CS-CP-FD 13.5 15.2 6.9 57.9 7915.7 18460.0 48.7 3001.3 8270.7
CS-NT-CD 9.8 18.2 7.6 58.4 7927.3 18428.6 48.4 3024.4 8281.8
CS-MP-CD 8.7 17.1 7.5 59.7 7959.0 18472.6 51.6 3098.3 8638.8
CS-CP-CD 9.5 17.9 7.6 58.8 7958.3 18487.6 49.3 3052.2 8356.0
* CC: continuous corn; CS: corn-soybean rotation; NT: no-till; MP: moldboard plow; CP: chisel plow; FD: free drainage; CD: controlled drainage.** Lateral groundwater flow below the tiles***Evapo-transpiration
Scenarios Flow-weighted N concentration (mg/L)
N loss in Tile
N loss to lateral flow
Net Mineraliza-tion
Denitrification
N uptake in corn biomass
N uptake in soybean biomass
N fixation
∆ inorganic N (1979-2002)
∆ organic N (1979-2002)
CC-NT-FD* 15.0 18.4 19.1 73.3 10.4 224.1 --- --- 85.7 1642.6
CC-MP-FD 17.4 18.4 17.5 70.3 8.6 223.7 --- --- 81.2 1637.6
CC-CP-FD 16.1 18.9 19.2 73.3 9.7 224.3 --- --- 83.2 1639.6
CC-NT-CD 15.3 13.4 21.9 72.2 11.3 224.4 --- --- 86.0 1647.6
CC-MP-CD 18.3 13.1 20.6 69.2 9.4 224.3 --- --- 81.2 1641.6
CC-CP-CD 16.6 13.7 22.0 72.1 10.5 224.7 --- --- 83.2 1644.6
CS-NT-FD 11.8 16.3 18.1 116.5 12.5 225.0 318.8 242.8 49.8 1672.7
CS-MP-FD 13.5 17.3 18.6 118.0 9.5 225.8 333.4 252.0 45.4 1793.6
CS-CP-FD 12.7 17.2 19.0 117.0 10.8 226.0 322.1 246.5 52.3 1679.9
CS-NT-CD 11.7 11.4 21.2 115.8 13.3 225.6 322.4 246.3 49.5 1727.6
CS-MP-CD 13.4 11.7 22.1 117.4 10.1 226.5 336.9 254.9 46.3 1798.6
CS-CP-CD 12.6 11.9 22.2 116.2 11.5 226.8 325.4 249.7 52.0 1684.7
* CC: continuous corn; CS: corn-soybean rotation; NT: no-till; MP: moldboard plow; CP: chisel plow; FD: free drainage; CD: controlled drainage.∆ = change from 1979 to 2002.
Simulated yearly nitrogen balance under different c rop rotation,Simulated yearly nitrogen balance under different c rop rotation, tillage, and drainage scenarios. tillage, and drainage scenarios. Units are in Kg N/ha unless stated otherwise. N app lication rateUnits are in Kg N/ha unless stated otherwise. N app lication rate was 202 kg N/ha for CC and 168 kg was 202 kg N/ha for CC and 168 kg N/ha on corn for CS and SC. Results for cornN/ha on corn for CS and SC. Results for corn --soybean rotation were taken as averages from CS and soybean rotation were taken as averages from CS and SC phases of the rotation (Ma et al., 2007b).SC phases of the rotation (Ma et al., 2007b).
(Akron, CO)
2. Dryland Copping Systems Studies in the Semi-Arid
Central Great Plains(Akron, CO)
Alternative Crop Rotation Exp.Alternative Crop Rotation Exp.
-23 dryland rotations (beginning 1991)
-Wheat, corn, millets, sunflower. peas
Millet Yield
Measured Grain Yield (kg ha -1)
0 1000 2000 3000 4000 5000
Sim
ulat
ed G
rain
Yie
ld (
kg h
a-1)
0
1000
2000
3000
4000
5000WCM
Corn Yield
Sim
ulat
ed G
rain
Yie
ld (
kg h
a-1)
1000
2000
3000
4000
5000WCMWCF
Wheat Yield
Sim
ulat
ed G
rain
Yie
ld (
kg h
a-1)
1000
2000
3000
4000
5000WCMWF (CT)WF (NT)WCF
Millet Biomass
Measured Biomass (kg ha -1)
0 3000 6000 9000 12000 15000
Sim
ulat
ed B
iom
ass
(kg
ha-1
)
0
3000
6000
9000
12000
15000 WCM
1 to 1 line
average 1 standard deviation
1 to 1 line
average 1 standard deviation
1 to 1 line
circled points are from calibration data set
average 1 standard deviation
1 to 1 line
circled points are from calibration data set
average 1 standard deviation
Wheat Biomass
Sim
ulat
ed B
iom
ass
(kg
ha-1)
0
3000
6000
9000
12000
15000 WCMWF (CT)WF (NT)WCF
average 1 standard deviation
1 to 1 line
Corn Biomass
Sim
ulat
ed B
iom
ass
(kg
ha-1)
0
3000
6000
9000
12000
15000 WCMWCF
average 1 standard deviation
1 to 1 line
SYNTHESIS – Biomass and Yield in 4 Rotations
W F(C T)-F
Soi
l wat
er (
mm
)
100
200
300
400
500 PredictedMeasured
W F(N T)-F
Soi
l wat
er (
mm
)
100
200
300
400
500PredictedMeasured
C T vs. N T - pred icted C TN T
C T vs. N T - m easured CTNT
(a) (c)
(b ) (d)
wheat
1992 1994 1996 1998 2000 2002
wheat wheat wheat wheat
wheat wheat wheat wheat wheat
wheat wheat wheat wheat wheat
1992 1994 1996 1998 2000 2002
wheat wheat wheat wheat wheat
Measured and predicted (RZWQM) total profile (180 cm) soil waterMeasured and predicted (RZWQM) total profile (180 cm) soil water under under wheat fallow (WF) for the beginning fallow data set. (a) convenwheat fallow (WF) for the beginning fallow data set. (a) conventional tillage tional tillage (CT), (b) no(CT), (b) no--till (NT), (till (NT), ( c)comparisonc)comparisonof predicted soil water under WF(CT) of predicted soil water under WF(CT) and WF(NT), (d) comparison of measured soil water under WF(CT) aand WF(NT), (d) comparison of measured soil water under WF(CT) and nd WF(NT) (WF(NT) (SaseendranSaseendranet al., 2005).et al., 2005).
ET - composite
Estimated ET, cm(other losses assumed zero)
0 20 40 60
Sim
ulat
ed E
T, c
m
0
20
40
60
1:1 lineWFWCMWCF
RMSE = 5.35 cm
R2 = 0.80E = 0.92
Wheat-Corn-Fallow (No-Till)F
ract
ion
orga
nic
mat
ter
in 5
cm s
oil l
ayer
0.014
0.016
0.018
0.020
0.022
0.024
0.026
0.028
WCF-WWCF-CWCF-F
WF(CT)-WWF(NT)-W
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Measured (1995)
FOM in the first 5 cm soil layer under W, C, and F phases FOM in the first 5 cm soil layer under W, C, and F phases of WCF, and W phase of WF(CT) and WF(NT) during of WCF, and W phase of WF(CT) and WF(NT) during 1992 to 2002. 1992 to 2002.
Using Validated Model to Explore Using Validated Model to Explore Drought Management StrategiesDrought Management Strategies
Summer Crop Selection in Summer Crop Selection in Rotation with WheatRotation with Wheat
Based on Initial Water Based on Initial Water
Cumulative probabilities Cumulative probabilities of Grain yields of of Grain yields of Corn, Corn, ProsoProsomillet and millet and Canola, and Biomass Canola, and Biomass of Foxtail millet and of Foxtail millet and Triticale planted with Triticale planted with Plant Plant AvailalbleAvailalbleWater at planting Water at planting from 0 to 100%from 0 to 100%
Note:Note: PAW in the soil PAW in the soil profile below 45 cm profile below 45 cm was constrained not to was constrained not to exceed 50% of PAW exceed 50% of PAW at field capacity.at field capacity.
0.0
0.2
0.4
0.6
0.8
1.0
0 1000 2000 3000 4000 5000 6000 7000Yield, kg/ha
cum
ulat
ive
prob
abili
ty
10% PAW
25% PAW
50% PAW
75% PAW
100% PAW
a) Corn
0.0
0.2
0.4
0.6
0.8
1.0
0 1000 2000 3000 4000 5000 6000 7000Yield, kg/ha
cum
ulat
ive
prob
abili
ty
b) Proso millet
0.0
0.2
0.4
0.6
0.8
1.0
0 3000 6000 9000 12000 15000Biomass, kg/ha
cum
ulat
ive
prob
abili
ty
d) Foxtail millet
0.0
0.2
0.4
0.6
0.8
1.0
0 3000 6000 9000 12000 15000Biomass, kg/ha
cum
ulat
ive
prob
abili
ty
e) Triticale
0.0
0.2
0.4
0.6
0.8
1.0
0 1000 2000 3000 4000 5000 6000 7000Yield, kg/ha
cum
ulat
ive
prob
abili
ty
c) Canola
Predicting Peak Standing Crop (rangeland biomass) a t different Predicting Peak Standing Crop (rangeland biomass) a t different levels of initial soil water Contents (SWC) and levels of initial soil water Contents (SWC) and
AprilApril --May precipitationMay precipitation
500
750
1000
1250
1500
1750
2000
2250
130 140 150 160 170 180 190
Apr.-May precipitation, mm
Pea
k st
andi
ng c
rop,
kg
ha-1
SWCi = 0.2SWCi = 0.3SWCi = 0.4
Driest
Normal
Wettest
Limited Irrigation Studies on Corn Limited Irrigation Studies on Corn in the Central Great Plainsin the Central Great Plains
Synthesis of Data: Grain Yield of Corn Synthesis of Data: Grain Yield of Corn
Measured yield, kg ha-1
0 2000 4000 6000 8000 10000 12000 14000
Sim
ulat
ed y
ield
, kg
ha-1
0
2000
4000
6000
8000
10000
12000
14000
1:1 lineIrrigatedRainfed
RMSE(rainfed) = 464 kg ha-1
RMSE(irrigated) = 982 kg ha-1
Using Validated Models to Explore Using Validated Models to Explore
alternative Scenarios for Managing Limited alternative Scenarios for Managing Limited Irrigation WaterIrrigation Water
Rain out experiments with different levels of irrigation Rain out experiments with different levels of irrigation in Cornin Corn
(Seasonal PET=90 cm, irrigated weekly)(Seasonal PET=90 cm, irrigated weekly)
1000
3000
5000
7000
9000
11000
10 20 30 40 50 60 70 80 90 100
Irrigation, cm
Yie
ld, K
g/ha
Uniform 20%V, 80%R 40%v, 60%r
(N not limiting )
1000
3000
5000
7000
9000
11000
10 20 30 40 50 60 70 80 90 100
Irrigation, cm
Yie
ld, K
g/ha
Uniform 20%V, 80%R 40%v, 60%r
(b) Silt Loam
Irrigation, mm 0 100 200 300 400 500 600 700 800
Gra
in yield, k
g ha
-1
2000
4000
6000
8000
10000
12000
20% V, 80% R40% V, 60% R50%V, 50% R
(c) Sandy Loam
Irrigation, mm 0 100 200 300 400 500 600 700 800
Gra
in yield, k
g ha
-1
2000
4000
6000
8000
10000
12000
20% V, 80% R40% V, 60% R50%V, 50% R
(a) Clay Loam
0 100 200 300 400 500 600 700 800
Gra
in yield, k
g ha
-1
2000
4000
6000
8000
10000
12000
20% V, 80% R40% V, 60% R50%V, 50% R
Grain Yield (20:80, 40:60 & 50:50 split Grain Yield (20:80, 40:60 & 50:50 split -- 94 year average)94 year average)
N balance N balance
(20:80 split (20:80 split -- 94 year average)94 year average)
N k
g ha
-1
0
20
40
60
80
100
Clay loamSilt loamSandy loam
(a) Leached N
N k
g ha
-1
100
150
200
250
300
Clay loamSilt loamSandy loam
(b) Uptake N
Irrigation, mm
100 200 300 400 500 600 700
N k
g ha
-1
0
20
40
60
80
100
Clay loamSilt loamSandy loam
(c ) Mineralization(c ) Mineralization
Limited Water Optimization Tool InterfaceLimited Water Optimization Tool Interface
Screen 1 – lowest level of economic input
Screen 3 – greatest economic input detail
Screen 2 – greater economic inputdetail
Production FunctionProduction Functionfor Water Optimizerfor Water Optimizer
Silt loam soil
ET/Irrigation, in0 5 10 15 20 25
Yld
, Bus
hels
0
20
40
60
80
100
120
140
160
Irr vs. Yld (simulated average for 97 yrs)
Yld= 25.4+10.4*ETa-dY=yd+(ym-yd)*(1-(1-dirr/dm)^(1/wue))
Akron
Sandy loam soil
ET/Irrigation, in0 5 10 15 20 25
Yld
, Bus
hels
0
20
40
60
80
100
120
140
160
Irr vs. Yld (simulated average for 97 yrs,)
Yld= 33.5+10.3*ETa-dY=yd+(ym-yd)*(1-(1-dirr/dm)^(1/wue))
Clay loam soil
ET/Irrigation, in
0 5 10 15 20 25
Yld
, Bus
hels
0
20
40
60
80
100
120
140
160
Irr vs. Yld (simulated average for 97 yrs)
Yld= 22.8+9.9*ETa-dY=yd+(ym-yd)*(1-(1-dirr/dm)^(1/wue))
Simulate yield response to irrigation and ETcrop
Simulated corn grain yield Simulated corn grain yield response to initiation of response to initiation of irrigation at different soil irrigation at different soil water depletion levels at water depletion levels at Akron, ColoradoAkron, Colorado
(a) number of irrigations and (a) number of irrigations and irrigation amounts simulated irrigation amounts simulated with the 94with the 94--yr weather record yr weather record in response to initiation of in response to initiation of irrigations at soil water irrigations at soil water depletions until 10 to 90% depletions until 10 to 90% plant available water (PAW), plant available water (PAW), andand
(b) cumulative probabilities (b) cumulative probabilities for corn grain yields simulated for corn grain yields simulated with the 94with the 94––yr weather record yr weather record in response to initiation of in response to initiation of irrigations at soil water irrigations at soil water depletions until 10 to 40% depletions until 10 to 40% PAW. Bars represent one PAW. Bars represent one standard deviation above the standard deviation above the mean.mean.
SOIL WATER DEPLETION LEVEL FOR IRRIGATIONSOIL WATER DEPLETION LEVEL FOR IRRIGATION
Yield, kg ha -10 2000 4000 6000 8000 10000 12000 14000
Cum
. pro
babi
lity
0.0
0.2
0.4
0.6
0.8
1.0
10% PAW20% PAW30% PAW40% PAW50% PAW60% PAW
(b) Cum. probability vs. grain yield
Soil water depletion levels, % plant available wate r (PAW)10 20 30 40 50 60 70 80 90
No.
of i
rrig
atio
ns
0
10
20
30
40
50
60
Irrig
atio
n, m
m
0
100
200
300
400
500
600
No. of irrigationsIrrigation, mm
(a) No. of irrigations and irrigation amount
Late Initiation of irrigation Late Initiation of irrigation -- Comparison between Comparison between soilssoils
Y ie ld , k g h a-10 2 0 0 0 4 0 0 0 6 0 0 0 8 0 0 0 1 0 0 0 0 1 2 0 0 0 1 4 0 0 0
Cum
. pro
babi
lity
0 .0
0 .2
0 .4
0 .6
0 .8
1 .0(b ) S i l t lo a m
Y ie ld , k g h a-10 2 0 0 0 4 0 0 0 6 0 0 0 8 0 0 0 1 0 0 0 0 1 2 0 0 0 1 4 0 0 0
Cum
. pro
babi
lity
0 .0
0 .2
0 .4
0 .6
0 .8
1 .0
1 0 % P A W2 0 % P A W3 0 % P A W4 0 % P A W5 0 % P A W6 0 % P A W
(a ) C la y lo a m
Y ie ld , k g h a-10 2 0 0 0 4 0 0 0 6 0 0 0 8 0 0 0 1 0 0 0 0 1 2 0 0 0 1 4 0 0 0
Cum
. pro
babi
lity
0 .0
0 .2
0 .4
0 .6
0 .8
1 .0
(c ) S a n d y lo a m
3. Water and N Management for 3. Water and N Management for WheatWheat--Maize DoubleMaize Double--Cropping Cropping
System in ChinaSystem in China
Climate Change Effects on Climate Change Effects on Cropping Systems Cropping Systems
in the Central Great Plains in the Central Great Plains
Wheat Yield in WCF for Projected ScenariosOver the Baseline Experiment Years, 1992 to 2007
Yield (kg/ha)
0 2000 4000 6000
0.0
0.2
0.4
0.6
0.8
1.0
BL2025205020752100
Yield (kg/ha)
0 2000 4000 6000
Pro
babi
lity
0.0
0.2
0.4
0.6
0.8
1.0
BL415550693 836
Yield (kg/ha)
0 2000 4000 6000
0.0
0.2
0.4
0.6
0.8
1.0
BLS1.9W0.8S2.7W1.6S3.5W3.2S4.3W3.2
Yield (kg/ha)
0 2000 4000 6000
Pro
babi
lity
0.0
0.2
0.4
0.6
0.8
1.0
BLS90S80S70S60
CO2 (ppm) Temperature (oC)
Precipitation (%) Projected years
Maize Yield in WCF for Projected ScenariosOver the Baseline Experiment Years, 1992 to 2007
Yield (kg/ha)
1000 2000 3000 4000 5000
0.0
0.2
0.4
0.6
0.8
1.0
BL2025205020752100
Yield (kg/ha)
1000 2000 3000 4000 5000
Pro
babi
lity
0.0
0.2
0.4
0.6
0.8
1.0
BL415550693 836
Yield (kg/ha)
1000 2000 3000 4000 5000
0.0
0.2
0.4
0.6
0.8
1.0
BLS1.9W0.8S2.7W1.6S3.5W3.2S4.3W3.2
Yield (kg/ha)
1000 2000 3000 4000 5000
Pro
babi
lity
0.0
0.2
0.4
0.6
0.8
1.0
BLS90S80S70S60
CO2 (ppm) Temperature (oC)
Precipitation (%) Projected years
Thank you very much for your Thank you very much for your attention attention
International CollaborationsInternational Collaborations
1.1. Enhancing applications of existing models in Enhancing applications of existing models in field researchfield research
2.2. Easier parameterization of models for field Easier parameterization of models for field scientistsscientists——effective properties of soils/cropseffective properties of soils/crops
3.3. Improving model components, e.g., water Improving model components, e.g., water stress response, N uptake, management stress response, N uptake, management effects, effects, --------
International Collaborations International Collaborations Contd.Contd.
4. Developing next-generation, stand-alone modular process components
5. Sharing model components and databases
6. Possibly developing a common modular modeling framework, with parameterization, visualization, and analysis tools
ASAASA--SSSASSSA--CSSA Initiatives to CSSA Initiatives to Advance Models and ApplicationsAdvance Models and Applications
A. A new book series on Advances in Ag Systems Modeling:
1. Response of crops to limited water2. Introducing system models in field research 3. Root-soil interactions4. Quantifying soil structure effects5. Quantifying soil carbon changes in cropping systems
ASAASA--SSSASSSA--CSSA Initiatives to CSSA Initiatives to Advance Models and ApplicationsAdvance Models and Applications
B. Ad-hoc Committee on Modules, Models, andDatabases:
1. E-publish well-documented code for process modules, after having them peer reviewed for quality of science, documentation and meta data, industry-standard code structure, and pre-published validation.
2. E-publish the models built from peer-reviewed modules, after peer review of the model interfaces and pre-published validations.
AdAd--hoc Committee Contd.hoc Committee Contd.
3. E-publish experimental databases needed for modules and models after peer review for quality of measurements, spatial representation, and completeness for calibration/validation
4. Maintain a library of peer-reviewed modules, models, and databases on the CSA website or website of a contractor or collaborator
Purpose of the Proposed ActivitiesPurpose of the Proposed Activities
• Advance science through synthesis and quantification of processes and interactions in the form of code, which scientistscan use to build future models.
• Promote sharing of modules, models, and databases, which will reduce duplication and encourage the best science in the modulescontributed by the experts for each process.
• E-publishing and web storage of modules, models, and databases will provide publishing credit to the contributors, an essentialmotivation for time consuming tasks of using standard code structure and documentation.
• Important Note:The above proposed activities are not supposed to interfere with or impede further development and applications of existing models. Rather, they are meant to help future enhancement of these models by providing peer-reviewed components.
The Tri-Societies welcome your feedback on these initiativesAnd future collaborations