Decision support system characteristics Ecosystem Management Decision Support (EMDS) components Unified planning process

The Role of Decision Support Systems in Natural Resource Management:

Overview of the Ecosystem Management Decision Support Framework

Kevin James, Heartland Inventory & Monitoring Network

Resource Information Management Conference, Fort Collins, ColoradoApril 19-23, 2010

1) Decision support system characteristics

2) Ecosystem Management Decision Support (EMDS) components

3) Unified planning process

Outline

What is a decision support system?

A computer-based system composed of a language system, knowledge system, and problem-processing system whose collective purpose is the support of decision-making activities.

- Holsapple (2003)

Knowledge-based system

A knowledge base is a formal, typically logical, specification for the interpretation of information.

-Walters and Nielson (1998)

Decision making process

A method that guides an individual or group through a series of tasks from problem identification and analysis to design of alternatives and selection of an alternative.

- Mintzberg et al. (1976)

• Improve the efficiency with which assessments are conducted

• Optimize use of information

• Better integration of diverse topics

• More comprehensive analysis

• Improve the quality of assessment products

Decision support system objectives

What is EMDS?

• Ecosystem Management Decision Support (EMDS)

• Spatial decision support system

• Application framework in the Windows environment

• Knowledge-based decision support of ecological assessments at any geographic scale

• EMDS Consortiumhttp://www.institute.redlands.edu/emds

• EMDS 4.1 standalone edition (ArcGIS engine)

• EMDS 4.1 ArcGIS integrated edition (ArcGIS extension)

• Full compatibility with ArcGIS 9.2, 9.3, and 9.3.1

EMDS 4.1

Two-phase system:

1) Evaluation phase performed with logic engine:

fuzzy logic and dependency networks

2) Planning phase performed with decision engine:

rate alternatives against a set of criteria

EMDS integrated approach

• integrates software environments

• GIS / spatial modeling

• Knowledge-based processing

• Netweaver (logic engine)

• Data acquisition manager

• Hotlink browser

• Priority Analyst (decision engine)

Image from The Redlands Institutewww.institute.redlands.edu/emds

EMDS framework

EMDS stand alone edition

Spatial input: study area

Multiple spatial scales:parkmanagement unitcommunity typepatch

Tabular input: logic structureResource Attribute Indicator Indicator-2 Proposition ReferenceDO Dry oak type is in acceptable condition(union) Canopy cover Forest canopy cover is acceptable ≥ 60

OS Structure Overstory structure is acceptable(union) OS volume Overstory volume of all tree species is acceptable

(+) SC 2 BA Basal area of all tree species in size class 2 is acceptable .90 - 1.21SC 3 BA Basal area of all tree species in size class 3 is acceptable 1.25 - 2.48 - 7.02 - 9.5SC 4 BA Basal area of all tree species in size class 4 is acceptable 3.54 - 11.56SC 5 BA Basal area of all tree species in size class 5 is acceptable 2.33 - 27.52

OS density Overstory density of all tree species is acceptable(+) SC 2 density Density of all tree species in size class 2 is acceptable 27 - 40 - 145 - 173

SC 3 density Density of all tree species in size class 3 is acceptable 17 - 33 - 100 - 146SC 4 density Density of all tree species in size class 4 is acceptable 27 - 93SC 5 density Density of all tree species in size class 5 is acceptable 10 - 110

OS composition(+) AcerSacc

TiliAmerCarya spp Relative basal area of Carya species is acceptable .001 - .20QuerAlbaQuerRubrQuerMacr

Oak regeneration Oak regeneration is acceptable(union) Oak seedling RD Relative density of oak seedlings is acceptable .001 - .111

Oak SC 1 RD Relative density of oak trees in size class 1 is acceptable .001 - .514US structure Understory guild representation is acceptable(union)

native grass Foliar cover of native grass is acceptable .05 - 3.3native forbs Foliar cover of native forbs is acceptable 4.7 - 55.9native grass-like Foliar cover of native grass-like species is acceptable .05 - 17.5fern Foliar cover of fern species is acceptable .05 - 3.1native woody Foliar cover of native woody species is acceptable 12.4 - .4non-native Foliar cover of non-native species is acceptable 7.2 - .1

SC5 density 10 110

Netweaver logic model

Spatial evaluation

SC5 density 10 110

Hotlink browser: explaining results

• Evaluated logic model

• Derivation of conclusions

• Explore the dependency network

• Increases transparency of results

• Multi-Criteria Decision Analysis component• Uses output from the logic evaluation and a decision model. • Rates landscape elements with respect to

– their condition, and– factors related to the feasibility and efficacy of management.

Priority Analyst (PA)

Aspatial output

Spatial output

Priority analyst output

Ecosystem management: philosophical

Adaptive management: operational

Strategic planning: methodology

Natural resource management

Adaptive, coherent planning

Evaluation withsingle model

Generatealternatives

Implementselected alternative

ID key questionsand data needs

Reviseimplementation

Unified approach to evaluation adds coherence through institutional memory.

Applies to initial condition, alternatives, andplan performance.

A single unified model for planning

• Assess current condition– Context for planning (where are we starting from?)

• Evaluate alternative strategies– A framework for synthesizing results of scenarios– Harvest scheduling, vegetation modeling, etc.

• Priority analysis for more tactical decisions– Which activities to do where?

• Evaluate plan implementation– How well is the plan working?

EMDS Consortium

• Pacific Northwest Research Station, USFS– System design & project oversight

• ESRI, primary contractor – ArcGIS implementation

• Rules of Thumb, Inc.– Netweaver logic engine

• InfoHarvest, Inc.– Priority Analyst engine

http://www.institute.redlands.edu/emds