Successfully reported this slideshow.
Your SlideShare is downloading. ×

2014 10 rotman mecnhanism and climate models

More Related Content

Related Audiobooks

Free with a 30 day trial from Scribd

See all

2014 10 rotman mecnhanism and climate models

  2. 2. PREVIEW  Main issue: A transition from models to mechanisms in climate change  Argument for a mechanistic view in climate change  Feedback  Optimality  Control/ manipulation  Understanding  Arguments against mechanisms in climate science  Holism  Failed mechanisms in physical sciences  So what?  Not yet  Will never happen  Not needed 2
  3. 3. WHAT IS UNDER SCRUTINY HERE?  The internal structure of climate models  Feedback in climate models  Mapping models to mechanism (M2M)  Many to one?  Many to many?  Optimality and plurality of models  Communicating results, metadata and mechanisms 3
  4. 4. SOME TOPICS OF INTEREST IN CLIMATE SCIENCE  Social values (viz. Epistemic values) in creating models (Winsberg 2012)  Complexity of models and “analytic understanding” (Parker 2014)  Multiplicity/plurality of climate models, (Parker 2010a)  Uncertainty of models  Stability, reliability of models  Explanatory power and understanding of climate models  Modularity Adapted from (Knutti and Sedláček 2012) 4
  5. 5. ROLES OF MODELS  Climate change is mainly about building and assessing models  Climate models are mainly:  predicting tools  generate other models or hypotheses  Quantification of theories of climate change  “hybrid”: predict and explain  Do climate models really explain? How?  Do we have an IBE with climate models? 5
  6. 6. OVERLAPPING MODELS IN MECHANISMS  Differenct communities focus on different parts  They do not necessarily look at the “coupled model”  Climate scientists are specialized 6
  7. 7. SOME VIEWS ABOUT MODELS AND MECHANISMS  Models are not related to mechanisms  mathematical models exist in physics, without being related to any mechanism  some models summarize data (phenomenal models)  some other models predict (are phenomenally adequate) but do not explain  Models represent mechanisms  One task of model building is to represent the dynamics of mechanisms (Bechtel and Abrahamsen 2011)  Models needs mechanisms to be explanatory  Models are explanatory when they describe a mechanism (Craver 2006)  Models map to mechanisms (M2M)  Let us call these models “mechanistic models” 7
  8. 8. MODEL ASSESSMENT IN CLIMATE SCIENCE  Confirmation of “the truth” of existing models (Lloyd 2010)  Adequacy-for-a-purpose: (Parker 2009)  Realism: accurate description of the actual climate system  Bayesian view  Possibilism (Katzav 2014)  Present focus: mapping models to a mechanism  How does model X map on the mechanism Y? 8
  9. 9. THREE EXTREME PREDICTIONS A. Where do I need to look in the sky to find the moon in London ON at 16:30 on 25.10.2044? B. What will be Ioan’s state of health on 4:30 on 25.10.2044, given this and this constraints on the world and what we know of his diet, genes, etc.? C. How far can I drive a Honda Civic car from London ON with a full tank of gas, in this direction, in the weather conditions, all things being equal? A= one theory, simple simulation, simple data, perfect prediction B= no theory no model, some mechanisms C= one mechanism, no theory, some initial conditions, no need of models 9
  10. 10. QUESTIONS  1 Are some (all?) climate models mechanistic?  2 Why explanation?  3 Can climate models explain without being “mechanistic”?  What advantages does a mechanistic view bring to climate science?  4 So what? Why do we need mechanistic explanation anyway? 10  1 yes, those in which feedback plays a role  2 We want to understand the “causal story” of the climate system. The understanding of why a phenomenon occurs (Parker 2014).  Question to Parker: is a mechanistic explanation better than causal explanation in improving our understanding of a phenomenon and of its question “why?”  3 yes, they can, but still mechanistic explanations can do better  4 Because with explanation, control, understanding and manipulation come!  4’ we can hope for the optimal model
  11. 11. EXTRAPOLATING MECHANISMS  Universality: Model-building occurs anywhere in science  Neuroscience/cognitive science (empirical data and laws/equations)  Biology (empirical data)  Physics (laws, symmetries)  Life science, medicine  Scientific revolution can be read, charitably as a process building models, mechanisms, unifying, eliminating models, creating theories etc.  I think it makes sense to talk about:  “mechanisms & models (together) in climate science” 11
  12. 12. A SIMPLIFIED VIEW I. Convergence from a plurality of models to a limited number of models  Culling models  Coupling submodels  Constraint models II. Mapping from a limited number of models to a limited number of mechanisms III. Convergence of mechanisms to a theory (unification of mechanisms and models) I think II deserves attention in the light of CMIP5 I am quietist about III. And I is already discussed in the literature 12
  13. 13. ADVANTAGES OF MECHANISMS  Introduce new explanations  Integrate causal stories  Introduce levels  Facilitate communication between submodels and between subroutines  Can map elements of models to mechanisms and give them materiality  Cluster of different models into mechanisms based on the M2M  Move from statistical explanations/arguments close to what the layman wants to hear (not probabilities, but conditionals) 13
  14. 14. “BOTTOMING OUT” MECHANISMS  Ignore the fundamental and fundamentality (deep physics)  Work at scales  Relative to a scale (time space energy)  Multiscale modeling 14
  15. 15. FROM MODELS TO MECHANISMS  Why do we need mechanisms? A Kantian innuendo:  “Dynamical models without mechanistic grounding are empty, while mechanistic models without complex dynamics are blind.” (Bechtel and Abrahamsen 2011)  This suggests a relation among models and mechanisms.  Normatively: models and mechanism should be mapped one onto the other. 15
  16. 16. DO MODELS EXPLAIN?  The Craver-Kaplan hypothesis (Kaplan and Craver 2011):  Models explain only when there is a model-to-mechanism mapping. M2M  Models needs to be modular in order to explain (Weber 2008) 16
  17. 17. THE MECHANISM-MODEL MAPPING  Biologists discover mechanisms  Models resemble the mechanism  Some models are better, some are worse, in representing the mechanism 17
  18. 18. MECHANISMS IN MODELS’ CLOTHING Are mechanisms already in the climate science? Try to identify in CMIP-5 the mechanistic mindset (not language) Unveil their explanatory role Explain the M2M mapping. 18
  19. 19. SYNTHETIC MODELING  Mechanism complements the computational modeling  It is not a question of reinterpretation of what climate scientists already do  It is more or less a reconstruction based on M2M  It does bring in a clearly stated language of levels  Cycles of amplification are called amplifying mechanisms 19
  20. 20. MECHANISTIC OPERATIONS IN THE MODELS  Decomposition is a procedure that happens in mechanisms  Switch on and off various components:  Inhibition  Stimulation  Recomposition of the operation of the mechanisms (Bechtel, 2011) 20
  21. 21. CLIMATE FEEDBACK 101  Feedback is never linear  Apply a forcing (CO2)  Temperature raise  Feedback changes  Look for mechanisms that are not switched off al low temperature  Once these processes go on, there is amplification or reducing of the temperature 21
  22. 22. FEEDBACK  Feedback can be positive or negative  The net feedback from the combined effect of changes in water vapor, and differences between atmospheric and surface warming is almost surely positive.  The net radiative feedback due to all cloud types combined is positive. 22
  23. 23. CMIP-5: A LOLLAPALOOZA OF FEEDBACK  AOGCM are not enough!  Earth System Models  Earth System Models of Intermediate Complexity  Includes cycles  Since AR4, the understanding of mechanisms and feedbacks of extreme in temperature improved 23
  24. 24. 24
  25. 25. FEEDBACK: M2M  Feedback can be captured by:  Non-linear equations  A cycle in a mechanism Simple mechanisms are serial: start to finish. They contain only linear causal chains Feedback loops complicate mechanisms. They are non-sequential Introduce timescale Synchronization of feedback (makes them positive or negative, depending on phase factor) 25
  26. 26. PRINCIPAL FEEDBACKS  The water vapor/lapse  Albedo  Cloud 26
  27. 27. CARBON CYCLE IN CMIP5 AND FEEDBACK  Increased atmospheric CO2 increases land and ocean uptake  Limitations on plant growth imposed by nitrogen availability 27
  28. 28. VAPOUR-CO2-CLIMATE  Vapour is a feedback not a Forcing of climate change  It is a fast and strong feedback (see Ch 8) 28
  29. 29. HOW DO WE REACH OPTIMALITY?  Optimality does not belong to a model  Through mechanisms (Machamer, Darden, and Craver 2000)  Optimal mappings between models and mechanisms  Reduce uncertainty 29
  30. 30. TIMESCALE MATTERS!  The effect of feedbacks is clear for longer timespans  Some feedbacks are delayed by centuries or millennia 30
  31. 31. Lifetime (years) GWP20 GWP100 GTP20 GTP100 31 CH b 4 12.4a Nocc fb 84 28 67 4 With cc fb 86 34 70 11 HFC-134a 13.4 Nocc fb 3710 1300 3050 201 With cc fb 3790 1550 3170 530 CFC-11 45.0 Nocc fb 6900 4660 6890 2340 With cc fb 7020 5350 7080 3490 N2O 121.0a Nocc fb 264 265 277 234 With cc fb 268 298 284 297 CF4 50,000.0 Nocc fb 4880 6630 5270 8040 With cc fb 4950 7350 5400 9560
  32. 32. ARGUMENTS AGAINST M2M IN CLIMATE SCIENCE  Climate science is a physical science in which mechanisms do not play the same role as in neuroscience/life science/  Some “disastrous” examples of mechanism thought in physics (ether, phlogiston, Cartesian physics)  Climate models are mathematical models, unlike models in neuroscience  Climate science is holistic, in pursue of complexity, not reductionist. Emergence looms large  Climate science is more about statistical reasoning, not about discovering reality/mechanisms.  Climate modelers are partially blackboxing, or probably grey-boxing their object of study 32