Science/Modeling Organizations to Bridge the Science-Policy Gap - - PowerPoint PPT Presentation

Decision Support Systems: Science/Modeling Organizations to Bridge the Science-Policy Gap

Denise Lach, Director School of Public Policy

Wicked Problems

• Solution depends on how problem is

framed

• Stakeholders have radically different

world views for understanding the problem

• Problem constraints and resources

needed change over time

• Problem is never solved definitively

Super Wicked Problems

• Time is running out
• No central authority
• Those seeking to solve the

problem are also causing it

• Policies discount the future

non-rationally

Complications: Uncertain Futures

Role of Science in Wicked Problems

Decision Stakes System Uncertainties

High High Low

Decision Stakes System Uncertainties

High High Low Normal Science

Decision Stakes System Uncertainties

High High Low Normal Science Professional Consultancy

Decision Stakes System Uncertainties

High High Low Normal Science Professional Consultancy Post-Normal Science

Post-Normal Science

• Facts are uncertain, values in

dispute, stakes high, and decisions urgent

• Less than desired information

available

• Not all factors are necessarily

knowable

• Always faced with uncertainties
• Mistakes can be costly or lethal

Coping with Wicked Problems

• Authority
• Competition
• Collaboration

Can we substitute process for certainty in resolving wicked problems?

Post-normal Boundary Organizations for Integrating Science and Policy

Form a research agenda around the needs of stakeholders Assemble needed expertise to address key questions Design decision support tools to translate the research answers into practical applications

Produce useable knowledge about climate impacts in the PNW

Some Recent PNW Study Areas

Skagit 2060 Kitsap Futures Tillamook Coastal Futures Willamette Water 2100 Forest People Fire Treasure Valley Big Wood Basin

Envision – Conceptual Structure

Landscape Performance Models

Generating Landscape Metrics Reflecting “Stuff People Care About”, e.g. Water Scarcity, Habitat, Jobs

Multiagent Decision Models

Actors selecting policies and generate land management decision affecting landscape pattern Landscape Feedbacks

Landscape Temporal GIS Landscape Process Models

Biophysical/Social/Economic Models (e.g. Climate, Hydrology, Population Growth, Veg Dynamics, Fire, …)

Visualizations

Stakeholder Engagement and Understanding Dynamic Maps, Charts, Flyovers/ Flythroughs…

Policies and Scenarios (From Stakeholder Process)

Scenario Planning Process

Identify System, Develop Initial Datasets Develop System Models Create Scenarios Evaluate Scenarios Develop Preferred Scenario Implement Plan

Scientists Stakeholders

Endpoints as Starting Points for fModeling

Alternative Scenarios: Economic base, management approach

Highly Managed / Agricultural Economy Highly Managed / Tourism Economy Less Managed / Agricultural Economy Less Managed / Tourism Economy

Economic Base

Ag Economy Tourism Economy

Management

Less Managed Highly Managed

Big Wood Climate Model Selection

12 Alternative Scenarios: economic base, management approach, climate scenario

ENVISION Model Framework

Thinking About Complicated Information: What’s Important?

Types of Information from Model: High Elevation April 1 SWE

1980-2009 Interquartile Range

2 out of 3 modeled simulations indicate a consistent reduction in April 1 SWE.

Types of Information from Model: SWE

Types of Information: Frost Free Periods

Big Wood Data Atlas

Lessons Learned: Modeling Challenges

Empirical Basis Level

• f Detail

Mechanism (Processes) It’s a Balancing Act! Computation Data Availability Stakeholder Relevance Uncertainty

Lessons Learned: Project Design

• Projects are both challenging and interesting
• Integration should come first, not last
• Systems approach essential – we need more

systems thinkers

• Multidisciplinary approach is critical
• Place Matters – be clear about what is general

and what is specific

Lessons Learned: Collaboration

• Team dynamics determines success or failure
• The “Culture of Science” can be a plus and a

minus -

+ Solid scientific footing to be useful, credible – “Out of box” thinking critical – disciplinary boundaries can limit thinking

• Stakeholders are generally pretty interesting

people who know a heck of a lot – engage the thought leaders early and often

• Make assumptions, choices transparent
• Address important issues/questions
• Create simple visuals
• Provide options for individual exploration
• Develop intuitive interface – stories?
• Provide meta data and data access

Lessons Learned: Communicating Usable Knowledge

Questions?

“Standard” Envision Plug-ins

Plug-in Function Target models growth of a surface based on total and available capacities and existing densities – very useful for population growth and spatial allocation models Modeler a high-level, XML-based model specification and execution tool for relatively simple models Spatial Allocator Allows definition of global allocations, constraints and preferences, useful for a broad variety of applications, eg. Fire spread, insect infestation, crop rotations, management choices Sync a tool for synchronizing changes to related columns Trigger a tool for triggering a set of outcomes when a specified field change – similar to Sync, but more flexible, slightly slower Flow a hydrological modeling framework SppHabMatrix A flexible Habitat Suitability modeling framework Developer A tool for specifying urbanization dynamics, can be used in conjunction with Target for modeling population growth and develop processes

Envision “Adapter” Plug-ins

Plug-in Function

VDDT/ DynamicVeg Dynamic vegetation models (state-transition) for running VDDT-based vegetation models FlamMap Detailed Process-based fire model MAPPS Global biogeography model Geospatial Data Reader Dynamic spatial data object for reading a variety geospatial formats e.g. NetCDF MC2 Global biogeochemistry model Century V5 Biogeochemistry model

ENVISION

Biofuel Production Carbon Forest Products Extraction Fire Risk (Habitat) Habitat Suitability Resource Lands Protection

Evaluative Models Data Sources Autonomous Process Models

Parcels (IDU’s) Population Growth and Residential Expansion Policy Set(s) Agent Descriptors VDDT Vegetative Succession (Spatialized and Climatized) Climate Change

Envision Central Oregon

FLAMMAP Fire Spread Fire Risk (Structures) Social Networks Landscape Amenities Terrestrial Biodiversity

Integrated Decision Units (IDUs)

A spatial geometry to model both human decisions and successional processes Each IDU described in GIS by a set of attributes used to model climate effects, succession, wildfire and decisions