Model Documentation - WITNESS: Difference between revisions

From IAMC-Documentation
Jump to navigation Jump to search
No edit summary
mNo edit summary
 
(6 intermediate revisions by the same user not shown)
Line 9: Line 9:
WITNESS is an open source Integrated Assessment Model [[wikipedia:Software framework|software framework]] - which means it's can be used as an application ready to use as is, or a toolset you can use, complete, customize or overload.
WITNESS is an open source Integrated Assessment Model [[wikipedia:Software framework|software framework]] - which means it's can be used as an application ready to use as is, or a toolset you can use, complete, customize or overload.


It's built on a powerful open source cooperative simulation platform  called [https://www.witness4climate.org/sostrades-3/ SoSTrades], and uses [[wikipedia:Multidisciplinary design optimization#:~:text=Multi-disciplinary design optimization (MDO) is a field of,solve design problems incorporating a number of disciplines.|multi-disciplinary analysis and optimization (MDAO)]] library called [https://gemseo.readthedocs.io/en/stable// GEMSEO].  
It can be used to explore potential future pathways for greenhouse gas emissions under different policy scenarios, assess how different levels of global warming will affect natural and human systems, identify the most cost-effective strategies producing energy while reducing emissions, analyze how to transition from fossil fuels to renewable energy sources while ensuring energy security and economic stability, evaluate the potential impacts of specific climate policies (e.g., carbon pricing or renewable energy incentives impact on emissions or economic growth), determine the role of emerging technologies such as carbon capture and storage or advanced nuclear in achieving climate goals…
 
Its cooperative study setup and analysis capabilities also allow WITNESS to enhance cooperation to achieve global climate targets within a group of "coopetitors".
 
 
WITNESS is  built on a powerful open source cooperative simulation platform  called [https://www.witness4climate.org/sostrades-3/ SoSTrades], and uses [[wikipedia:Multidisciplinary design optimization#:~:text=Multi-disciplinary design optimization (MDO) is a field of,solve design problems incorporating a number of disciplines.|multi-disciplinary analysis and optimization (MDAO)]] library called [https://gemseo.readthedocs.io/en/stable// GEMSEO].  


This leading edge simulation base allow using the framework to set up and solve a large range of problems, be it in the pure IAM ecosystem area or deriving IAM scenarios/results for various applications such as evaluation of physical risks, optimization of ecosystem energy migration, or assets portfolio valuation…
This leading edge simulation base allow using the framework to set up and solve a large range of problems, be it in the pure IAM ecosystem area or deriving IAM scenarios/results for various applications such as evaluation of physical risks, optimization of ecosystem energy migration, or assets portfolio valuation…
Line 16: Line 21:
The initial motivations to code "Yet Another IAM", a were - in that order :
The initial motivations to code "Yet Another IAM", a were - in that order :


# to properly take into account energy-macroeconomy coupling, which was very well highlighted as problematic in current scenarios in a [https://www.carbone4.com/lancement-iris-initiative/ document from Carbon4], and well synthetized in the World GDP vs World energy production diagram it contains showing how most IAM predictions represent a sudden and abrupt change from decades of correlation between these 2 quantities
# to properly take into account energy-macroeconomy coupling, which was very well highlighted as problematic in current scenarios in a [https://www.carbone4.com/lancement-iris-initiative/ document from Carbon4], and well synthetized in the World GDP vs World energy production diagram it contains showing how most IAM predictions represent a sudden and abrupt change from decades of correlation between these 2 quantities.
# to properly add and couple an endogenous population model. Population is exogeneous in most models whereas climate impact on population and increase of pandemic rates with pressure on land could have significant economic impact- as demonstrated by impact of SARS-CoV-1, or measured productivity losses during recent heatwaves
# to properly add and couple an endogenous population model. Population is exogeneous in most models whereas climate impact on population and increase of pandemic rates with pressure on land could have significant economic impact- as demonstrated by impact of SARS-CoV-1, or measured productivity losses during recent heatwaves
# to build an open and cooperative basis for a modular IAM that would be easy to extend and scale, thanks to re-using in the IAM area the leading edge complex system of systems simulation techniques that initial authors were  experts at using in aerospace sector (including but not limited to MBSE and MDAO) - with the ability to reliably manage very large amount of models and automating handling of interactions and feedback loops and avoid the simulation constraints hampering modeling intend.
# to build an open and cooperative basis for a modular IAM that would be easy to extend and scale, thanks to re-using in the IAM area the leading edge complex system of systems simulation techniques that initial authors were  experts at using in aerospace sector (including but not limited to MBSE and MDAO) - with the ability to reliably manage very large amount of models and automating handling of interactions and feedback loops and avoid the simulation constraints hampering modeling intend.


We decided to develop a Stock Flow Coherent model, as it seems they are the only models that succeeds in back-testing evaluations (cf [https://chair-energy-prosperity.org/wp-content/uploads/2024/10/WP2024-les-modeles-IAMs-et-leurs-limites.pdf/ "Les modèles IAMs et leurs limites" A.Grandjean 2024]).
We decided to develop a Stock Flow Coherent model, as it seems they are the only models that succeeds in back-testing evaluations (cf [https://chair-energy-prosperity.org/wp-content/uploads/2024/10/WP2024-les-modeles-IAMs-et-leurs-limites.pdf "Les modèles IAMs et leurs limites" A.Grandjean 2024]).
 




The global structure of WITNESS is often presented using this picture
The global structure of WITNESS is often presented using this picture (Water not well implemented yet despite shown here) :
 
[[File:WITNESS Overview Small.png|alt=WITNESS overview]]
[[File:WITNESS Overview Small.png|alt=WITNESS overview]]




but the detailed coupling structure is depending of the problem you solve, coupling being handled automatically.
but the detailed coupling structure is depending of the problem you solve, coupling being handled automatically.


As an example, [https:///www.witness4climate.org/files/WITNESSCouplings/WITNESS Val Full Optim new x0 converged Eval interface diagram.sozi.html here] is a map of a typical WITNESS full couplings.
As an example, [https:///www.witness4climate.org/files/WITNESSCouplings/WITNESS_Val_Full_Optim_new_x0_converged_Eval_interface_diagram.sozi.html here] is a map of a typical WITNESS full couplings.





Latest revision as of 10:44, 26 December 2024

Alert-warning.png Note: The documentation of WITNESS is 'in preparation' and is not yet 'published'!

Model Documentation - WITNESS

Corresponding documentation
Previous versions
No previous version available
Model information
Model link
Institution Open-Source for Climate (OS-C), N/A, https://os-climate.org/transition-analysis/., Linux Foundation (LF), N/A, https://www.linuxfoundation.org/.
Solution concept Systems dynamics based approach
Solution method OptimizationSimulation-based optimization
Anticipation


Model presentation

WITNESS is an open source Integrated Assessment Model software framework - which means it's can be used as an application ready to use as is, or a toolset you can use, complete, customize or overload.

It can be used to explore potential future pathways for greenhouse gas emissions under different policy scenarios, assess how different levels of global warming will affect natural and human systems, identify the most cost-effective strategies producing energy while reducing emissions, analyze how to transition from fossil fuels to renewable energy sources while ensuring energy security and economic stability, evaluate the potential impacts of specific climate policies (e.g., carbon pricing or renewable energy incentives impact on emissions or economic growth), determine the role of emerging technologies such as carbon capture and storage or advanced nuclear in achieving climate goals…

Its cooperative study setup and analysis capabilities also allow WITNESS to enhance cooperation to achieve global climate targets within a group of "coopetitors".


WITNESS is built on a powerful open source cooperative simulation platform  called SoSTrades, and uses multi-disciplinary analysis and optimization (MDAO) library called GEMSEO.

This leading edge simulation base allow using the framework to set up and solve a large range of problems, be it in the pure IAM ecosystem area or deriving IAM scenarios/results for various applications such as evaluation of physical risks, optimization of ecosystem energy migration, or assets portfolio valuation…


The initial motivations to code "Yet Another IAM", a were - in that order :

  1. to properly take into account energy-macroeconomy coupling, which was very well highlighted as problematic in current scenarios in a document from Carbon4, and well synthetized in the World GDP vs World energy production diagram it contains showing how most IAM predictions represent a sudden and abrupt change from decades of correlation between these 2 quantities.
  2. to properly add and couple an endogenous population model. Population is exogeneous in most models whereas climate impact on population and increase of pandemic rates with pressure on land could have significant economic impact- as demonstrated by impact of SARS-CoV-1, or measured productivity losses during recent heatwaves
  3. to build an open and cooperative basis for a modular IAM that would be easy to extend and scale, thanks to re-using in the IAM area the leading edge complex system of systems simulation techniques that initial authors were  experts at using in aerospace sector (including but not limited to MBSE and MDAO) - with the ability to reliably manage very large amount of models and automating handling of interactions and feedback loops and avoid the simulation constraints hampering modeling intend.

We decided to develop a Stock Flow Coherent model, as it seems they are the only models that succeeds in back-testing evaluations (cf "Les modèles IAMs et leurs limites" A.Grandjean 2024).


The global structure of WITNESS is often presented using this picture (Water not well implemented yet despite shown here) :

WITNESS overview


but the detailed coupling structure is depending of the problem you solve, coupling being handled automatically.

As an example, here is a map of a typical WITNESS full couplings.


You can find some videos of live presentations of WITNESS model are available on witness4climate.org website.

Reference documentation

WITNESS reference documentation is located directly in the source GitHub repositories, in a "documentation" folder next to the source code for each discipline.

Let's for example, consider the population model folder in GitHub:

Population github folder snapshot

The documentation markup file can be displayed directly in the tool in the "Documentation tab" for each discipline:

Population documentation in GitHub
Population documentation in GitHub

and will also be available in the tool in the "Documentation" tab for the discipline:

Population documentation in SoSTrades
Population documentation in SoSTrades


References and links to academic papers / reference dataset used are listed at the end of these disciplines documentation pages.

Installation documentation

Installation documentation for standalone laptop installation can be found here.

Local classical installation or as a Docker component is available for Linux, Mac and Windows.

For a full cloud installation with DevSecOps of models, please contact the project here :-)

Developer's documentation

A draft developer's documentation is available here.

User and developer's MOOC are planned to be developed and will be hosted by Linux Foundation.

Support

Support sessions are organised for users ("Come as you are" sessions) and for developers ("Code as you are" sessions).

Please check here for next sessions and registration.