Transcript Simplot_Introduction Presentation

Analyzing Precision Ag Data
…a mini-workshop on instructional materials for moving
precision agriculture beyond mapping
AgKnowledge GIS Faculty Development Workshop
August 12-15, 2002
Kirkwood Community College, Cedar Rapids, Iowa
Joseph K. Berry
Berry & Associates
2000 South College, Suite 300
Fort Collins, CO 80525
Email: [email protected]
Web Site: www.innovativegis.com/basis
What Is Precision Agriculture? (new technology)
…about doing the right thing at the right
place and time
…identifies and responds appropriately
to the variability within a field
…augments (not replaces) indigenous
knowledge
(Circa 1992)
…stackable livestock
(Berry)
What Is Precision Agriculture? (Whole-field)
Whole-Field Management is based on broad averages of field
data with management actions directed by “typical” conditions.
Whole-field assumes the “average”
conditions are the same everywhere
Weigh-wagon, or grain elevator measurements, established a
field’s yield performance. Soil sampling determined the typical
nutrient levels within a field. From these and other data the best
overall seed variety was chosen and a constant rate of fertilizer
applied, as well as a bushel of other decisions— treating the
entire field as uniform within its boundaries.
(Berry)
What Is Precision Agriculture? (Site-specific)
Site-Specific Management recognizes the variability within a field
and changes management actions throughout a field.
Management-zones breaks the field
into areas of similar conditions
Z2
Z1
Z1
Z3
Z2
Surface-maps breaks the field into
consistent pieces that track the
specific conditions at each location
It involves assessing and reacting to field variability by tailoring
management actions, such as fertilization levels, seeding rates
and variety selection, to match changing field conditions. It
assumes that managing field variability leads to both cost savings
and production increases, as well as improved stewardship and
environmental benefits.
(Berry)
What Is Precision Agriculture? (Technologies)
A new application of existing
technologies to assist in managing field
variability…
• Global Positioning System (GPS) – Establishes
position in a field
• Data Collection Devices (IDI- monitors) – Collects
data “on-the-fly”
• Geographic Information Systems (GIS) – Used for
data visualization and analysis
• Intelligent Implements (IDI- controls) – Provides
variable rate control “on-the-fly”
(Berry)
What Is Precision Agriculture? (Processing Steps)
Utilizes spatial relationships in a field for site-specific
management…
• Continuous Data Logging 1) Yield Map
“What you see is what you get”
• Discrete Point Sampling 2) Condition Maps
“Guessing over the whole field”
Where is What
• Mapped Data Analysis 3) Relationships
“Now what could’ve caused that?”
• Spatial Modeling 4) Prescription Map
“Do this here, but not over there”
Why and So What
Do What
and Where
(Berry)
Precision Agriculture’s Big Picture
Enabling technologies
A new application of the
Spatial Technologies…
www.innovativegis.com/basis/pfprimer/
Where is What
Why and So What
…that utilizes spatial
relationships in a field for sitespecific management of fields
Do What and Where
Data processing approach
(Berry)
Overview of the Case Study
Workshop objectives are to ) introduce the
concepts, procedures and issues
surrounding precision agriculture data and
2) provide hands-on experience
Fall 2002
in analysis of these data
Example Applications–
several annotated examples of
grid-based map analysis
Install MapCalc– software
system for hands-on experience
Ex#1
(Berry)
Map Data Visualization and Summary
Table 1. Workbook Topics
Overview of the Case Study
Mapped Data Visualization and Summary
Comparing Mapped Data
Spatial Interpolation
Characterizing Data Groups
Developing Predictive Models
Analyzing Spatial Context
(Berry)
Map Data Visualization and Summary
Numerical statistics …min, max, range,
mean, median standard deviation, variance
Geographic statistics …total area, area
by class, drill-down
Ex#2
Descriptive statistics
Drill-Down
(Berry)
Comparing Map Data
Table 1. Workbook Topics
Overview of the Case Study
Mapped Data Visualization and Summary
Comparing Mapped Data
Spatial Interpolation
Characterizing Data Groups
Developing Predictive Models
Analyzing Spatial Context
(Berry)
Comparing Discrete Maps (Joint coincidence)
What differences do you see? –
“How different are the maps?”
“How are they different?”
“Where are they different?”
…but map patterns can be
quantitatively compared
A Coincidence Table reports the
number of cells for each joint
condition with diagonal cells
identifying agreement.
(Berry)
Comparing Map Surfaces (Difference map)
1997_Yield_Volume
- 1998_Yield_Volume
Yield_Diff
Map Variables… map values within an analysis grid
can be mathematically and statistically analyzed
Ex#3
…green indicates
areas of increased
production
…yellow indicates
minimal change
…red indicates
decreased production
(Berry)
Spatial Interpolation
Table 1. Workbook Topics
Overview of the Case Study
Mapped Data Visualization and Summary
Comparing Mapped Data
Spatial Interpolation
Characterizing Data Groups
Developing Predictive Models
Analyzing Spatial Context
(Berry)
Spatial Expression of Arithmetic Average
“Mapping the Variance”
…Soil Samples are collected with GPS coordinates
…the location and nutrient levels of the sample points
(Discrete Data) are used to estimate the nutrient
pattern throughout the field (Continuous Data)
(Berry)
Spatial Interpolation (Mapping spatial variability)
Ex#4
…the geo-registered
soil samples form a pattern
of “spikes” throughout the
field. Spatial Interpolation
is similar to throwing a
blanket over the spikes that
conforms to the pattern.
…all interpolation algorithms assume that
1) “nearby things are more alike than
distant things” (spatial autocorrelation),
2) appropriate sampling intensity, and
3) suitable sampling pattern.
…the continuous surfaces produced “map
the spatial variation in the data samples.
(Berry)
Spatial Interpolation (Average vs. IDW)
Comparison of the interpolated surface to the whole field average
shows large differences in localized estimates
Difference Map
(Berry)
Spatial Interpolation (Compare maps)
Comparison of the IDW and Krig interpolated surfaces shows small
differences in in localized estimates
Difference Map
(Berry)
Spatial Interpolation Techniques
Characterizes the spatial distribution by fitting a mathematical
equation to localized portions of the data (roving window)
(Berry)
Spatial Interpolation
Assessing Interpolation Results (Residual Analysis)
…the best map is the
one that has the best
“guesses”
(Berry)
Spatial Interpolation
A Map of Error (Residual Map)
…shows you where your estimates are likely good/bad
(Berry)
Characterizing Data Groups
Table 1. Workbook Topics
Overview of the Case Study
Mapped Data Visualization and Summary
Comparing Mapped Data
Spatial Interpolation
Characterizing Data Groups
Developing Predictive Models
Analyzing Spatial Context
(Berry)
Visualizing Spatial Relationships
Interpolated Spatial Distribution
Phosphorous (P)
What spatial
relationships do you
see?
…do relatively high levels
of P often occur with high
levels of K and N?
…how often?
…where?
(Berry)
Calculating Data Distance
…an n-dimensional plot depicts the multivariate distribution; the distance
between points determines the relative similarity in data patterns
…the closest floating ball is the least similar (largest data distance) from the comparison point
(Berry)
Identifying Map Similarity
…the relative data distance between the comparison point’s data pattern
and those of all other map locations form a Similarity Index
The green tones indicate field locations with fairly similar P, K and N levels; red tones indicate dissimilar areas.
(Berry)
Clustering Maps for Data Zones
…a map stack is a spatially organized set of numbers
…groups of “floating balls” in data space
identify locations in the field with similar data
patterns– data zones
…fertilization rates vary for the different
clusters “on-the-fly”
Ex#5
(Cyber-Farmer, Circa 1992)
Variable Rate Application
(Berry)
Evaluating Clustering Results
…if the boxes do not overlap (much), the clusters are distinct.
(Berry)
Developing Predictive Models
Table 1. Workbook Topics
Overview of the Case Study
Mapped Data Visualization and Summary
Comparing Mapped Data
Spatial Interpolation
Characterizing Data Groups
Developing Predictive Models
Analyzing Spatial Context
(Berry)
RS Imagery as GIS Data Layers
A RS image is just a “shishkebab
of numbers” like any other
grid map (raster)
Image
52
148
46
26
(Beyond our sight)
Color Infrared
34
44
Remote sensing
images are composed
of numbers, just like
any other map in a
grid-based GIS…
“Map-ematical
Processing”
43
NIR (R)
Red (G)
Green (B)
57
P
312
K
257
7.5
7.2
ph
etc.
(Berry)
Creating Prediction Models (Scatter Plot)
…a Scatter Plot identifies the “joint condition” at each map
location; the trend in the plot forms a prediction equation
Map Set New Graph Scatter Plot
(Berry)
Deriving a Predictive Index (NDVI)
…an index combining the Red and NIR maps can be used to generate
a better predictive model
Normalized Difference Vegetation Index
for the sample grid location
NDVI= ((NIR – Red) / (NIR + Red))
NDVI= ((121-14.7) / (121 + 14.7))= 106.3 / 135.7= .783
(Berry)
Evaluating Prediction Maps (Spatial error analysis)
…the regression equation is evaluated and the predicted map is compared
to the actual measurements to generate an error map
Error = Predicted - Actual
for the sample grid location
Yest = 55 + (180 * .783) = 196 …error is 196 – 218 = 22 bu/ac
Error = Predicted - Actual
Note that the average error is 2.62 and 67% of the predictions are within +/- 20 bu/ac
Also, most of the error is concentrated along the edge of the field
(Berry)
Stratifying Maps for Better Predictions
Stratifying by Error Zones
Other ways to stratify mapped data—
1) Geographic Zones, such as proximity to the field
edge; 2) Dependent Map Zones, such as areas of low,
medium and high yield; 3) Data Zones, such as areas
of similar soil nutrient levels; and 4) Correlated Map
Zones, such as micro terrain features identifying
small ridges and depressions.
The Error Zones map is used
as a template to identify the
NDVI and Yield values used to
calculate three separate
prediction equations.
A Composite Prediction map is
created by applying the
equations to the NDVI data
respecting the template map
zones.
(Berry)
Assessing Prediction Results
Stratified
Prediction
Error Map
Actual
Yield
Error Map for
Stratified Prediction
none
80%
Whole Field
Prediction
none
(Berry)
The Precision Ag Process (Fertility example)
As a combine moves through a field 1) it uses GPS to check its location then 2) checks
the yield at that location to 3) create a continuous map of the yield variation every few
feet. This map 4) is combined with soil, terrain and other
maps to derive a 5) “Prescription Map” that is used to
Steps 1)–3)
6) adjust fertilization levels every few feet in the field.
Prescription Map
Step 5)
45.00
On-the-Fly
Yield Map
40.00
Map Analysis
Step 4)
Farm dB
Step 6)
35.00
Cyber-Farmer, Circa 1992
…you’ve come a long ways baby
Zone 3
30.00
25.00
Zone 2
20.00
15.00
10.00
5.00
Zone 1
5.00
10.00
Variable Rate Application
15.00
20.00
25.00
30.00
(Berry)
Spatial Data Mining
…making sense out of a map stack
Mapped data that
exhibits high spatial
dependency create
strong prediction
functions. As in
traditional statistical
analysis, spatial
relationships can be
used to predict
outcomes
…the difference is
that spatial statistics
predicts where
responses will be
high or low
(Berry)
Analyzing Spatial Context
Table 1. Workbook Topics
Overview of the Case Study
Mapped Data Visualization and Summary
Comparing Mapped Data
Spatial Interpolation
Characterizing Data Groups
Developing Predictive Models
Analyzing Spatial Context
(Berry)
Micro Terrain Analysis (Deriving Slope and Flow)
Characterizing Slope
A digital terrain surface is formed by
assigning an elevation value to each
cell in an analysis grid. The “slant” of
the terrain at any location can be
calculated— inclination of a plane
fitted to the elevation values of the
immediate vicinity
Slope and Flow maps draped over
vertically exaggerated terrain surface
Characterizing Surface Flow
A map of surface flow is simulated by
aggregating the “steepest downhill
paths” from each cell— confluence
(Berry)
Micro Terrain Analysis (Slope and Flow Classes)
Calibrating Slope and Flow Classes:
Areas of Gentle, Moderate, and
Steep slopes are identified; areas of light, moderate and heavy flows are identified
(Berry)
Micro Terrain Analysis (a simple erosion model)
Determining Erosion Potential: The slope and flow classes are
combined into a single map identifying erosion potential
(Berry)
Map Analysis Macros (Fat buttons)
Assembly language
Programming Languages (Visual C++)
Programming Objects (ActiveX Controls)
Ex#6
Input
Parameter Specification
Output
General Software System
Application Languages (MapCalc scripting)
Fat Buttons (Embedded Macros)
Application-Specific System
Execution Environment (Visual Basic)
Fat Buttons (Embedded Controls)
(Berry)
Gaps in Our Thinking
• Limited Approach –
Mapping vs. Data
Analysis; Tools vs. Science
• Science Link –
“Scientific Method” Doctrine,
The “Random” Thing, Appropriate Driving
Variables, Correlation vs. Causation
• Market Confusion –
Empirical Verification,
Economic Validation, Rationalization (Productivity
vs. Stewardship)
…Environmental Trump Card
…Education/Training is Key
Education, Enlightenment, Economics, Environment
(Berry)
Analyzing Precision Ag Data
…a mini-workshop on instructional materials for moving precision agriculture beyond mapping
Presented by Joseph K. Berry
"Precision farming isn't just a
bunch of pretty maps, but mapped
data and a set of procedures
linking these data to appropriate
management actions."
See http://www.innovativegis.com/basis/pfprimer/Default.html
…to access the online Precision Farming Primer
See http://www.innovativegis.com/basis/MapAnalysis/Default.html
…to access the online Map Analysis book
Berry & Associates // Spatial Information Systems
2000 South College, Suite 300, Fort Collins, CO 80525
Email: [email protected]
Web Site: www.innovativegis.com/basis