Transcript NADSSoverview - University of Nebraska–Lincoln

National Agricultural
Decision Support System
(NADSS)
Principal Investigator: S. Goddard,
Co-Principals: J. Deogun, M.J. Hayes, K.G. Hubbard, S.E.
Reichenbach, P.Z. Revesz, W.J. Waltman, and D.A. Wilhite
Co-Investigators: M.E. Tooze, S.K. Harms, J.S. Peake, and
T. Tadesse
A Geospatial Decision Support System for
Agricultural Risk Management
The Partnership
National Science Foundation’s Digital Government Program
National Drought Mitigation Center, University of Nebraska--Lincoln
High Plains Regional Climate Center, UNL
USDA Risk Management Agency, Natural Resources Conservation
Service, National Agricultural Statistics Service, and the Farm
Service Agency
USGS EROS Data Center
Nebraska Research Initiative on Geospatial Decision Support Systems
GIS Workshop
Funding
Source: NSF: $1 Million, 7/01—6/04
Title: DIGITAL GOVERNMENT: A Geospatial Decision
Support System for Drought Risk Management
Principal Investigators: Steve Goddard, Jitender Deogun, Michael J.
Hayes, Kenneth G. Hubbard, Stephen Reichenbach, Peter Revesz, W.J.
Waltman, Donald A. Wilhite, and Mark D. Svoboda, University of
Nebraska-Lincoln (UNL), Lincoln, Nebraska 68588-0115.
([email protected])
Co-Investigators: Sheri K. Harms, University of Nebraska-Kearney;
J.S. Peake, University of Nebraska-Omaha; Ray Sinclair and Sharon
Waltman, USDA Natural Resources Conservation Service, National
Soil Survey Center, Lincoln, NE; and Marcus Tooze, GIS Workshop,
Lincoln, NE.
Funding
Source: USDA RMA/FCIC: $1.3 Million, 10/02—9/04
Title: RISK ASSESSMENT AND EXPOSURE ANALYSIS
ON THE AGRICULTURAL LANDSCAPE: A Holistic
Approach to Spatio-Temporal Models and Tools for
Agricultural Risk Assessment and Exposure Analysis
Principal Investigators: Steve Goddard, Jitender Deogun, Michael J.
Hayes, Kenneth G. Hubbard, H. Douglas Jose, Stephen Reichenbach,
W.J. Waltman, Donald A. Wilhite, and Mark D. Svoboda, University of
Nebraska-Lincoln (UNL), Lincoln, Nebraska 68588-0115.
([email protected])
Co-Investigators: Norman Bliss, EROS Data Center; Sioux Falls, SD:
Sheri K. Harms, University of Nebraska-Kearney; and J.S. Peake,
University of Nebraska-Omaha; Ray Sinclair and Sharon Waltman,
USDA Natural Resources Conservation Service, National Soil Survey
Center, Lincoln, NE; and Marcus Tooze, GIS Workshop, Lincoln, NE.
Project Goals

Develop a support system of geospatial analyses
that will enhance agricultural risk assessment and
exposure analysis. Initial emphasis is on drought.




Compute and map drought indices at increased spatial
and temporal resolutions.
Provide transparent access to distributed geospatial,
relational, and constraint databases.
Develop new algorithms (using data mining and
knowledge discovery techniques) that seek out patterns
between weather stations, crop yields, and ENSO
events.
Develop new geospatial analyses to better visualize the
emergence, evolution, and movement of drought.
National Agricultural Decision
Support System (NADSS)

http://nadss.unl.edu/
Current Tools


Our current tools apply risk
analysis methodologies to
the study of drought
 Integration of basic
models with data
generates “information”
for analysis by decision
makers
 Information can be
gathered at any
resolution for which we
have data
http://nadss.unl.edu
Current NADSS Tools
Current NADSS
Tools
Prototype planting date guide tool with
climograph, date sliders, numerical
information, and navigation buttons.
Sample Climograph and Soil Moisture Regime
probability analysis map
Building a Spatial View

Data from information and knowledge layers are
translated spatially and interpolated to provide a “risk
view” for a defined area
Risk Indicators
Drought Indices
Soil Data
Climate Data
Reported Yields
Other Data Type
Surfacing
Raster interpolation of data points
within various windows
Inverse Distance Weighting
Spline
Kriging
Display
Re-summarization of
raster data
Generation of displayable
images
Risk Assessment in Practice

By combining several domain specific factors from our
“information layer” we are able to create maps displaying
the risk for states, regions or countries
The user adjusts
weight factors for
each variable
Variables are
spatially rendered
The result is a
“spatial” view
of risk
Risk Assessment
Total Market Value
Projecting Potential
Impacts for DecisionMakers as County
Dairy Farms
Profiles
Beef Farms
Congressional Delegation
State Legislature
USDA and State Agencies
Commodity Groups and
Agribusiness
Data Mining and
Knowledge Discovery
NIR Corn Yields In Nebraska Through Time
Annual
Milk
Production
Non-Irrigated Corn
Yields of
Nebraska
Through Time
Corn Yield (Bushels/Acre)
140
Clinton County
120
y = 1.6509x - 3195
100
R2 = 0.7646
80
60
1991?
1973-1974
40
1988
1966
20
0
1940
1950
1960
1970
Year
Year
1980
1990
2000
ENSO events and other El
Nino/La Nina processes
can serve as a trigger
mechanism for drought.
Mining the patterns
between crop yields and
ENSO signals may provide
new insights to risk
management and forecasting potential impacts
on cropping systems.
Distributed Geospatial Decision
Support System Architecture
Presentation (User Interface)
e.g., Web Interface, Java applet
Data cache
Knowledge Layer
e.g., Data
e.g.,
Mining,
Exposure
Exposure
Analysis,
Analysis,
Risk Assessment
Risk Assessment
Data cache
Information Layer
e.g., Drought Indices, Regional Crop Losses
Data cache
Distributed Spatial and Relational Data
e.g., Climatic Variables, Agricultural Statistics
HTTP
IIOP
RMI
TCP
4-Layer Architecture for
NADSS
User Interface
Web Client
Java Client
Local Database Cache
Knowledge Layer
Data Mining Tools
Drought Exposure Analysis Tools
Drought Vu lnerability and Soil
Climate Reg ime Analysis Tools
Local Database Cache
Information Layer
Index Producing Service
(SPI, PDSI, NSM)
Wrapped legacy Mapping Service
(ArcPlot, GRASS)
Interpolation Service
(Spline, Kriging, IDW )
Local Database Cache
Data Layer
(Local)
Attribute Data Cache
CGI data access (client)
Data Layer
(Distributed)
Relat ional DB
UCA N client
Geodata Cache
DiscoveryLink (Wrapper)
UCA N server
Structure flat files
OGDI (driver stub)
OGDI driver
MLPQ (Constrained DB)
Geospatial DB
HTTP
IIOP
RMI
NADSS Benefits and Impacts


Improving spatial and temporal analysis
for drought risk management
 State level to County level to Field level
 Monthly Index to Weekly Index (SPI,
PDSI, Newhall Simulation Model)
Responding to drought events more
effectively
Conclusion

We are addressing Data Interpretation and
Data Integration problems by creating a
Distributed Geospatial Decision Support
System architecture

The Distributed Geospatial Decision
Support System architecture is applicable
to many other distributed geospatial
decision support systems