Socio-Economic Modeling for Large-scale Quantitative Climate Change Analysis

Group “B6”Summary Presented by

Ian Foster

2

Context

Impact of climate change on society and the ultimate effectiveness of specific responses depends on human actions, which will determine, for example, how energy supply and demand evolve over time, and how and when different responses are deployed and applied

We must model human responses if we are to understand likely impacts and the effectiveness of responses, and thus help to sustain a prosperous and secure society

3

For Example: Bioenergy

Will biofuels push out food crops, raise food prices, and impact food security?

Will biofuels create unexpected negative rather than positive external environmental effects?

Could biofuels even exacerbate the impact on climate when the entire production chain is taken into account?

How will increased investment in biofuels affect trade patterns?

What would a sustainable approach to bioenergy look like?

(Source: Sustainable Bioenergy, UN Energy)

4

State of the Art in Climate Change Impact/Response Modeling and Analysis Multiple decades of efforts in energy system modeling,

integrated modeling, etc. But limitations in computational capabilities have led to

gross simplifications in models and analyses Current models neglect, for example:

– Integration of diverse complex systems and their relationships

– Incorporation of nonlinearities and thresholds– Full representation of feedback effects– Treatment of risks & uncertainties

Exascale computers and advanced algorithms offer the potential to overcome these limitations

5

Socio-Economic Modeling Challenges

Quantitative exploration of specific subsystems– E.g., energy markets, international agreements,

technology innovation, human & social behaviors– Some are highly computationally demanding

Data estimation, integration, analysis– Data access, cleaning, management, visualization– Empirical estimation of parameters (much computation)

Integrated modeling of natural system & human system– Climate change, carbon cycle, water cycle, ecosystem– Multi-scale and multi-sector analysis– Goal: integrated natural system/human system model

Uncertainty analysis of subsystems and systems

6

From Terascale to Exascale

Terascale (i.e., today, almost)– Economic models with ~10 countries & ~10 sectors– Limited coupling with climate models– No treatment of uncertainty and business cycle risk– Simple impact analysis for a limited set of scenarios– Limited ability to provide quantitative policy advice

7

Petascale– Economic models with more countries, sectors,

income groups– Limited treatment of uncertainty, business cycle risk– Stronger coupling with climate models

From Terascale to Exascale

8

Tera

Peta

From Terascale to Exascale Exascale

– Economic models with all countries, many sectors, many income groups

– Many policy instruments (taxes, tariffs, quotas, CAFE, CO2 taxes), nonlinear policies, etc.

– High spatial resolution in land use, etc.– Detailed coupling & feedbacks with climate models– Optimization of policy instruments & technology

choices over time and with respect to uncertainty– Detailed model validation & careful data analysis– Treatment of technological innovation, industrial

competition, population changes, migration, etc.

9

From Terascale to Exascale

Incr

ease

d cl

imat

e de

tail

Increased socio-economic detail

Tera

Peta

Peta

Exa

10

A Potential Timeline

2007: Socio-economic models on small clusters 2009: First Petascale computer (we are told) 2011: First socio-economic-environmental models on

Petascale computers assuming major new program (otherwise 2016 or later?)

2015: First Exascale computer (we assume) 2016: Better socio-economic-environmental models

running on Exascale computers assuming major new program (otherwise 2021 or later?)

2025: Socio-economic-environmental Tokomak/ Stellarator

11

Requirements (1)

Comprehensive suite of socio-economic models of unprecedented sectoral, regional, and social detail, with comprehensive error analysis on the representation

State-of-the-art climate modeling with detailed and accurate regional resolution

Basic research into foundational issues such as spatial statistics, modeling of social processes, relevant micro-activity and biosphere coupling issues, and relevant mathematical challenges, such as multiscale modeling

Assembly and quality control of extensive data sets Comprehensive and detailed validation of both individual

models and large model systems

12

Requirements (2)

Development of novel, robust numerical techniques and high-performance computing approaches to deal with the expected orders-of-magnitude increase in model complexity; also visualization methods for high-dimensional data

A wide range of application studies aimed at both validation and application

Education programs aimed at training the next generation of computational economists and other social scientists

13

Thoughts on Program Structure

Matrix organization involving a six team approach– The teams are one dimension of the matrix; they

focus on solving problems and issues in their broad areas

– Along the other dimension of the matrix, teams interact, share information and progress

As progress is made in all six areas, and as the various modeling teams better understand how to consistently couple their modeling areas, we move to a common large-scale integrated model system

14

A Potential Structure: Six Subteams

1. Modern economic computation approaches involving greatly expanded micro data bases to capture diversity in decision-making agents & strategic incentives that different country players may exercise

2. Consistent socio-economic modeling with many regions/sectors & with technology rich specifications that capture technology system efficiencies, synergies, & diversities & represent least-cost dynamic transition pathways to a low-carbon, adaptive global economy

3. Social science modeling attempting to capture the human dimension of climate-induced changes in greenhouse gas emissions, mitigation, & adaptation

15

A Potential Structure: Six Subteams

4. Integrated assessment team bringing in the climate and natural resource models (water, agriculture, land use, forest changes, biodiversity, health and disease) and their coupling to socio-economic models.

5. Advanced optimization & computation methods needed for integrated assessment, solving large-scale nonlinear systems, and addressing approaches for accounting for uncertainty

6. Data access, estimation, analysis, and visualization; model validation

+ Ongoing analysis of hardware requirements

16

A Couple of Questions

Scope– Potential applications for exascale socio-economic

modeling reach far beyond “climate change” Ability to deal with what may be radically different

hardware—unknown

17

Synopsis

Integrated modeling of the social, economic, and environmental system with an extensive treatment of couplings among these different elements is a scientific and computational grand challenge

By allowing for far more detailed treatments of the various components and feedbacks among these different components, and issues of uncertainty and risk, Exascale computers have the potential to transform our understanding of the science of socio-economic-environmental interactions