Transcript The model

“APEX-AGROTOOL” simulation environment
and its use for long-term analysis of different
crop rotation practices
Alex Topaj,
Sergey Medvedev1,
Vladimir Badenko,
Vitalij Terleev2
1Agrphysical
Research Institute
2Saint-Petersburg Polytechnical University
Workshop
“Simulation in den Umwelt- und
Geowissenschaften”
Müncheberg, 25-27 March, 2015
Who are we…
Agrophysical Research
Institute
Laboratory of Agroecosystem
Simulation
1967-2012
Ratmir A. Poluektov
1930-2012
1973 –2014
Your Majesty Model…
Model is a daily-step discrete recurrent operator describing
agroecosystem dynamics from planting to harvesting.
x1
2
x
x
...
x i (k  1)  f i ( x(k ), u (k ), w(k ), A)
x(0)  x0
k  1,2,...N
xn
x(k), x(k+1) –vector of state variables,
u(k) – controlling management (agronomic treatments),
w(k) – weather (daily data)
A – model
parameters.
Agricultural
crop as simulation object
i  1,2,..., n
Generic Crop Model AGROTOOL v 3.5.
Modeling domain
Limiting factors
competition, pests,
insects
P, K,
microelements
N, C
W, P
Q, T
Leaf Area Development & Light
Interception
Approach
D
LightVUtilization
P-R
(ecology)
Yield Formation
Y(PRT)
IVPhenology
Crop
f(T,O)
(mineral nutrition)
Root Distribution over Depth
EXP
Stresses
III Involved
W, N
(Nitrogen regime)
Water Dynamics
Evapo-transpiration
II
(Water
regime)
Soil CN-model
I
(Photosynthesis, phenological development)
R
PM
CN, P(5)
C:N Sub-model in AGROTOOL
Model verification







Leningrad region:
Spring barley, summer wheat, winter rye, oat, potato, perennial
grasses. Rotational and specialized field test.
Saratov region (middle Volga):
Summer wheat. Water stress field tests.
Krasnodar region:
Summer wheat, maize.
Altai region (West Siberia):
Alfalfa, summer wheat
Muncheberg, Badlauchstadt, Germany:
Summer wheat, spring barley, sugar beat.
Kaliningrad region:
Summer wheat, perennial grasses.
Tver region:
Summer wheat, spring barley, perennial grasses, rape, potato, oat.
Landscape field test.
Simulation software infrastructure & components
Project AGROTOOL
Stationary
DataBase SIAM
(MS Access,
PostgreSQL…)
Selector
Adapter
Agrotool
Weather
Generator
Parameter
Identification
Desktop
GUI
ODB
(MS Excel)
System
of
Polyvariant
Analysis
APEX
Weather
Generator
Model
AGROTOOL
Results
Single
Running
Interface
Input
data
Web-Interface
Adapter
Third-party
model
Project SIAM
Results
- data flows
- control
Project APEX
Third-party model
Frameworks of Crop Modeling Software
Issues:



Polyvariant analysis – multiple running of the model in
package mode with various input datasets (multivariate
computer experiment).
Generic shell – versatile user interface for different thirdparty crop models.
Structural identification (adaptation) – ability to
design the dynamic algorithm of model proceeding from
the set of alternative pre-developed modules
Framework
APEX
APSIM /
PMF
Polyvariant
analysis


Generic
shell



DSSAT
BioMA
GUICS
STICS /
RECORD












ROMUL /
DLES
Features
Structural
identification

Challenges for polyvariant analysis
Problem
Source of multivariance
Sensitivity analysis and parametric
identification
Parameter value variability
Statistical analysis and productivity
assessment
Actual
Weather
Climate change influence on crop
productivity
Future Weather
Scenarios
Optimization of Agrotechnologiy
Variants (dates and rates) of
technological treatments
Operative information support of
field experiments
Variants of technological
treatments and future weather
to the end of vegetation period
Precision agriculture and GIS
integration
Spatial heterogeneity of
agricultural field
Long-term analysis of Crop
Rotation
Fields, Seasons and Cultures of
Rotation under Investigation
Factor 1
Multivariate running. Full factor experiment
PROJECT
Factor 2
RESULTS
RUN
Scenario
Дата
BLEAF WSOIL EPLANT
13/04/11 0.12
22.1 324
14/04/11 0.14
24.1 345
15/04/11 0.16
23.4 355
…
Factors
Gradations
А – Project Creation Dialog; В – Scenario Table
Factors in APEX
Predefined qualitative factors:






Soil
Culture/Variety
Weather
Technology
Location
Initial State
Disadvantages
• No metrics/order relationships
• Semantic constraints on the
models being connected
• Fixed set of factors for model
analysis
Advantages
• Limited and manageable
number of factors
• «Semantically rich» data
processing and result
analysis
Weather generator linking.Climate
change investigation
Parameters
Actual
Weather
Data
identification
Location:
1. Leningrad region
Cultures:
1. Spring barley
2. Potato
3. Winter rye
variation
Generated
Weather
Scenarios
Parameters
Weather
generator
generation
Scenarios:
1. A2
2. B2
GCMs:
1. HadCM3 (GB)
2. ECHAM (Germany)
Model
Terms:
1. 2020
2. 2050
3. 2080
Decision making levels in crop
management (yield programming)
Level
Decisions
Planning
horizon
Simulation
Tools
Use Cases
I
Strategic
(Project)
Years,
Decades
Soil fertility
simple
regression
models
Land-use
projects
Crop rotation
planning
II
Tactic
(Planned)
Vegetation
Periods
Crop
production
models
Management
and
Optimization of
Agrotechnologies
Operative
Days, Weeks,
Months
Crop
production
models
On-line support
and forecasting
III
Dynamic Crop Models for Crop Rotation
Analysis: Challenges and Requirements





Improving the accuracy and adequacy of simulation due to many
factors taking into account
Multi-variant computation caused by input data variability (e.g.
Weather vs. Climate)
Statistical interpretation of simulation results and risk analysis
Big number of characteristics controlled and/or monitored by the
model (productivity, physiology, ecology, fertility etc.)
Advanced management of model uncertainties
Simulation of several consequent vegetation periods
according to chosen rotation scheme


The model must simulate different cultures and take into
consideration agroecosystem dynamics during non-growing
season (wintering)
The runtime framework must support the calculation of
scenarios in a predetermined sequence and the transfer of data
from the previous scenario to the next one
«A-A» for Crop Rotation Analysis
Requirement
Crop model:
Generic simulator
Current state
«Inherited» model variables:
AGROTOOL:
Versatile algorithm for all maintained cultures. Calibrated
Shoot litter (aboveground biomass)
models for cereals (summer and winter wheat, winter rye,
Root litter (belowground biomass)
barley, oats), maize, potato, root vegetables, annual and
Humus
content in 1m. Layer
perennial forages,
legumes.
Total Mineral Nitrogen in 1m. Layer
«Wintering»
Predecessor influence
Simulation infrastructure:
Continuous calculation. Modified descriptions of snow
Nitrogen
melting and Nodule
soil thermal
regime. (for legumes)
Separated calculation of litter and root residues in the
module of carbon-nitrogen transfer and transformation in
soil. Sub-model of symbiotic nitrogen fixation and nodule
nitrogen dynamics.
APEX:
Multiple running
Validated and implemented integrated environment for
multivariate analysis and automation of computer
experiments with crop models.
Crop rotation support
Special plug-in for planning not full factorial experiments
and performing complex serial-parallel schemes of scenario
computation.
Forecasting
Built-in stochastic generator of daily weather variables that
takes into account possible climate changes.
APEX functionality for continuous
model computation



«Factor joining» (not-full
factor experiment).
Clutch of gradations of
several factors to the
tuple
Assigning of scenario
running order
Setting the transfer of
data from the results of
the previously performed
scenario to the input
data the following
scenario
APEX Crop Rotation Plugin
selection
tool
dividing
factor
ordering
tool
Crop Continuity in AGROTOOL
(“Groundhog Year” Test)
SPIN-UP
STEADY
STATE
http://agrotool.ru
http://www.rpoluektov.ru