Facing the climate challenge: Implications of the 2 degree limit

1,973 views

Published on

This is a lecture I gave for Leslie Field's class on Engineering and Climate Change at Stanford on September 24, 2013. It describes an alternative to the traditional benefit-cost framing of the climate problem, called "working forward toward a goal". It's one that relies on our best understanding of the climate system as well as the lessons from business planners about facing big strategic challenges. See the discussion in my book Cold Cash, Cool Climate: Science-based Advice for Ecological Entrepreneurs http://amzn.to/Av0O9O for details.

Published in: Technology
0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
1,973
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
5
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

Facing the climate challenge: Implications of the 2 degree limit

  1. 1. Facing  the  climate  challenge:     Implica0ons  of  the  2  degree  limit   Jonathan  Koomey   Research  Fellow,  Steyer-­‐Taylor  Center  for  Energy  Policy   and  Finance,  Stanford  University   jgkoomey@stanford.edu   hJp://www.koomey.com   Engineering  and  Climate  Change,  Professor  Leslie  Field   Stanford  University   September  24,  2013   1  Copyright  Jonathan  Koomey  2013  
  2. 2. The  world  is  warming  and  humans  are   responsible    “A  strong,  credible  body  of  scien0fic  evidence  shows  that  climate   change  is  occurring,  is  caused  largely  by  human  ac0vi0es,  and  poses   significant  risks  for  a  broad  range  of  human  and  natural   systems.  .  .  .    Some  scien0fic  conclusions  or  theories  have  been  so  thoroughly   examined  and  tested,  and  supported  by  so  many  independent   observa0ons  and  results,  that  their  likelihood  of  subsequently   being  found  to  be  wrong  is  vanishingly  small.  Such  conclusions  and   theories  are  then  regarded  as  seJled  facts.  This  is  the  case  for  the   conclusions  that  the  Earth  system  is  warming  and  that  much  of  this   warming  is  very  likely  due  to  human  ac0vi0es.”      US  Na0onal  Academy  of  Sciences.  2010.  Advancing  the  Science  of   Climate  Change   2  Copyright  Jonathan  Koomey  2013  
  3. 3. Historical  global  C  emissions   Sources: Carbon Dioxide Information Analysis Center (CDIAC), plotted in Cold Cash, Cool Climate. Copyright  Jonathan  Koomey  2013   3  
  4. 4. Big  jump  in  CO2  concentra0ons  from   fossil  fuels  and  land  use  changes   Sources: Vostok and Lawdome ice core data, plus measured concentrations from the Carbon Dioxide Information Analysis Center, plotted in Cold Cash, Cool Climate Copyright  Jonathan  Koomey  2013   4  
  5. 5. A  closer  look  at  the  last  12,000  years   Sources: Vostok and Lawdome ice core data, plus measured concentrations from the Carbon Dioxide Information Analysis Center, plotted in Cold Cash, Cool Climate Copyright  Jonathan  Koomey  2013   5  
  6. 6. Global  surface  temperatures  have   risen  in  the  last  century   Source: The Copenhagen Diagnosis 2009 Copyright  Jonathan  Koomey  2013   6  
  7. 7. Increasing  temperatures  “load  the   dice”   Source:    Adapted  from  a  graph  made  originally  by  the  University  of     Arizona,  Southwest  Climate  Change  Network   Copyright  Jonathan  Koomey  2013   7  
  8. 8. What  the  data  show   8   Source:    The  New  Climate  Dice:  Public  Percep=on  of   Climate  Change.  James  Hansen,  Makiko  Sato,  and   Reto  Ruedy.  August  2012.   hJp://www.giss.nasa.gov/research/briefs/ hansen_17/.    Data  are  for  Northern  Hemisphere.   X-­‐axes  in  graphs  below  are  in  standard  devia0ons,  not   degrees  C.   Copyright  Jonathan  Koomey  2013  
  9. 9. Percent  of  US  land  area  subject  to  1   day  precipita0on  extremes   Source: NCDC/NOAA 2011 Copyright  Jonathan  Koomey  2013   9  
  10. 10. No-­‐policy  case  carbon  dioxide   concentra0ons  to  2100   Source: Sokolov et al. 2009 for projected concentrations and ice core and directly measured data for historical numbers. Copyright  Jonathan  Koomey  2013   10  
  11. 11. No-­‐policy  case  greenhouse  gas   concentra0ons  to  2100  (all  gases)   Copyright  Jonathan  Koomey  2013   11   Source: Sokolov et al. 2009 for projected concentrations and ice core and directly measured data for historical numbers.
  12. 12. No-­‐policy  case  greenhouse  gas   concentra0ons  to  2100   Source: Sokolov et al. 2009, plotted in Cold Cash, Cool Climate Copyright  Jonathan  Koomey  2013   12  
  13. 13. Current  trends  =  5  C  degrees  by  2100,   with  no  end  in  sight   Adapted from Copenhagen Diagnosis 2009, with MIT data taken from Sokolov et al. 2009. MIT climate sensitivity is 2.9 degrees C, but warming by 2100 doesn’t reflect the full warming impact because full equilibration takes centuries. 13  Copyright  Jonathan  Koomey  2013  
  14. 14. Fossil  fuel  scarcity  will  not  constrain   carbon  emissions   Source: Lower bound resource estimates from the IIASA Global Energy Assessment 2012 + Sokolov et al. 2009 (fossil emissions only). 14  Copyright  Jonathan  Koomey  2013  
  15. 15. What can we do? 15  Copyright  Jonathan  Koomey  2013  
  16. 16. Our options •  Adapt–modify human systems to make them more flexible and resilient •  Suffer–accept what comes (but what comes is likely to be costly in lives, ecosystem damage, and economic disruption) •  Mitigate–reduce emissions 16  Copyright  Jonathan  Koomey  2013  
  17. 17. Questions about mitigation options •  How much carbon will they save? •  How much will they cost? •  Are they feasible – technically? (science and technology) – logistically? (implementation and policy) – politically? (social will and equity) 17  Copyright  Jonathan  Koomey  2013  
  18. 18. 18   Cost-­‐benefit  analysis:    the  standard   approach   Copyright  Jonathan  Koomey  2013  
  19. 19. The  forecas0ng  quandary   •  Economics  ≠  physics:    we  need  to  act,  but  it’s   impossible  to  calculate  costs  and  benefits  in   an  accurate  way   •  Implica0on:    the  conven0onal  model  of  full   benefit-­‐cost  analysis  before  ac0ng  is  not   adequate  to  address  this  problem   19  Copyright  Jonathan  Koomey  2013  
  20. 20. An  evolu0onary,  path-­‐dependent  view   •  There  is  no  “op0mal  path”,  but  there  are  many   possible  alterna0ve  paths   –  We  can’t  plan  or  know  everything  about  the  path   ahead  but  the  warming  limit  defines  the  broad   outlines  of  success   •  Our  choices  now  affect  our  op0ons  later   •  Need  to     –  invest  in  a  broad  pornolio  of  op0ons   –  fail  fast   –  modify  plans  dynamically   –  learn  as  fast  as  we  can   20  Copyright  Jonathan  Koomey  2013  
  21. 21. Forecasts  ooen  underes0mate  the   possibili0es  for  change   •  Economic  models  (with  very  few  excep0ons)   –  assume  current  rigidi0es  will  con0nue  forward  in  the   forecast  (“The  Big  Mistake”,  related  to  Ascher’s   “assump0on  drag”)   –  assume  structure  of  property  rights  is  constant   –  ignore  increasing  returns  to  scale   –  rely  on  incomplete  technology  and  policy  pornolios   –  ignore  “no-­‐regrets”  op0ons   •  All  but  last  issue  true  for  top-­‐down  AND  boJom-­‐ up  models   21  Copyright  Jonathan  Koomey  2013  
  22. 22. An  alterna0ve  approach   •  Define  a  warming  limit  (e.g.  2  C  degrees    above   preindustrial  levels)   •  Determine  the  total  greenhouse  gases  we  can   then  emit  to  stay  under  that  limit   •  Define  pathways  that  meet  that  constraint   •  Assess  what  we’d  need  to  do  achieve  that   pathway  (#  of  power  plants,  rate  of  improvement   in  energy  efficiency,  etc)   •  Try  op0ons,  fail  fast,  alter  course  as  needed   22  Copyright  Jonathan  Koomey  2013  
  23. 23. 2  C  degree  warming  limit   •  Keeps  global  T  within  humanity’s  experience   •  Likely  avoids  the  worst  of  the  posi0ve  feedbacks   •  Implies  cumula0ve  GHG  emissions  “budget”   •  Limit  itself  now  widely  accepted  (e.g.,  G8  in  2009),  but   implica0ons  s0ll  not  well  known   –  Global  emissions  must  turn  down  this  decade,  down  50%   by  2050,  more  soon  aoerwards   –  Wai0ng  has  a  real  cost   –  We  must  act  quickly  on  many  fronts   •  It’s  Sputnik,  not  Apollo   –  We  can’t  burn  it  all   •  C  Storage  not  prac0cally  relevant  for  decades,  if  ever   23  Copyright  Jonathan  Koomey  2013  
  24. 24. Delaying  makes  no  sense  in  the   warming  limit  context   •  When  we  act  makes  a  difference   •  Delaying  ac0on  on  climate   – eats  up  the  budget   – makes  required  reduc0ons  more  difficult  and   costly  later   – sacrifices  learning  and  reduces  possibili0es  for   future  ac0on   •  Remember,  energy  techs  don’t  ∆  fast   24  Copyright  Jonathan  Koomey  2013  
  25. 25. There’s  no  0me  to  waste   25   Source:    The  Copenhagen  Diagnosis,  2009   Copyright  Jonathan  Koomey  2013  
  26. 26. Most  2050  infrastructure  built   between  now  and  2050   26  Copyright  Jonathan  Koomey  2013  
  27. 27. Working  toward  the  limit   •  Like  strategic  planning,  not  forecas0ng   •  e.g.,  to  meet  some  frac0on  of  the  target   –  how  many  emission-­‐free  power  plants  would  we   have  to  build  and  how  much  capital  would  that   require?   –  how  fast  would  efficiency  need  to  improve  given   expected  rates  of  economic  growth?   –  what  ins0tu0onal  changes  would  be  needed  to   accelerate  the  rate  of  implementa0on?   •  A  way  to  organize  our  thinking  about  solu0ons   to  the  problem   27  Copyright  Jonathan  Koomey  2013  
  28. 28. Mee0ng  constraints  of  the  safer  climate   case  won’t  be  easy   Source: Lower bound resource estimates from the IIASA Global Energy Assessment 2012 + calcs in Cold Cash, Cool Climate (fossil emissions only). 28  Copyright  Jonathan  Koomey  2013  
  29. 29. Summary   •  Warming  limit  approach  is  similar  to  how  businesses   make  big  strategic  decisions   •  Focus  is  on  risk  reduc0on,  experimenta0on,  evalua0on,   innova0on  and  cost  effec0veness,  not  on  knowing   “op0mal”  path  in  advance  (impossible!)   •  Science  points  to  2  deg  C  limit  but  ul0mate  choice  is  a   poli0cal  judgment   –  Declare  value  judgment  up  front  (not  buried  in  black  box   models,  as  is  customary)   •  Implies  rapid  reduc0ons  and  keeping  most  fossil  fuels  in   the  ground  (requires  rapid  innova0ons  in  technologies   AND  behavior/ins0tu0ons)   29  Copyright  Jonathan  Koomey  2013  
  30. 30. Summary  (con0nued)   •  Immediate  implementa0on  is  essen0al  (can’t   just  wait  and  see  while  doing  R&D)   – Learning  by  doing  only  happens  if  we  do!   •  Exis0ng  low  carbon  resources  are  plen0ful  but   we’ll  need  new  innova0ons  in  later  decades  to   keep  reduc0ons  on  track   •  Start  small.    Think  big.    Get  going!   30  Copyright  Jonathan  Koomey  2013  
  31. 31.      “The  best  way  to  predict  the  future  is  to   invent  it.”    –Alan  Kay   31  Copyright  Jonathan  Koomey  2013  
  32. 32. References   •  Allison,  et  al.  2009.  The  Copenhagen  Diagnosis,  2009:  Upda=ng  the  World  on  the  Latest  Climate  Science.  Sydney,  Australia:  The   University  of  New  South  Wales  Climate  Change  Research  Centre  (CCRC).     •  Caldeira,  Ken,  Atul  K.  Jain,  and  Mar0n  I.  Hoffert.  2003.  "Climate  Sensi0vity  Uncertainty  and  the  Need  for  Energy  Without  CO2  Emission  "     Science.    vol.  299,  no.  5615.  pp.  2052-­‐2054.  <hJp://www.sciencemag.org/cgi/content/abstract/299/5615/2052>   •  DeCanio,  Stephen  J.  2003.  Economic  Models  of  Climate  Change:    A  Cri=que.  Basingstoke,  UK:  Palgrave-­‐Macmillan.     •  Brown,  Marilyn  A.,  Mark  D.  Levine,  Walter  Short,  and  Jonathan  G.  Koomey.  2001.  "Scenarios  for  a  Clean  Energy  Future."    Energy  Policy     (Also  LBNL-­‐48031).    vol.  29,  no.  14.  November.  pp.  1179-­‐1196.     •  Gritsevskyi,  Andrii,  and  Nebojsa  Nakicenovic.  2000.  "Modeling  uncertainty  of  induced  technological  change."    Energy  Policy.    vol.  28,  no.   13.  November.  pp.  907-­‐921.     •  Koomey,  Jonathan.    Tes0mony  of  Jonathan  Koomey,  Ph.D.  for  a  hearing  on  "Efficiency:    The  Hidden  Secret  to  Solving  Our  Energy  Crisis".     Joint  Economic  CommiJee  of  the  U.S.  Congress.    U.S.  Congress.  Washington,  DC:  U.S.  Congress.  July  30,  2008.  <hJp:// www.jec.senate.gov/index.cfm?FuseAc0on=Hearings.HearingsCalendar&ContentRecord_id=6fc51d63-­‐e7e2-­‐82b7-­‐10c3-­‐3faa2c150115>   •  Koomey,  Jonathan  G.  Cold  Cash,  Cool  Climate:    Science-­‐Based  Advice  for  Ecological  Entrepreneurs.  Burlingame,  CA:  Analy0cs  Press,   2012.   •  Krause,  Floren0n,  Wilfred  Bach,  and  Jonathan  G.  Koomey.  1992.  Energy  Policy  in  the  Greenhouse.  NY,  NY:  John  Wiley  and  Sons.  (1989   edi0on  of  this  book  downloadable  at  <hJp://files.me.com/jgkoomey/9jzwgj>)   •  Meinshausen,  Malte,  Nicolai  Meinshausen,  William  Hare,  Sarah  C.  B.  Raper,  Katja  Frieler,  Reto  Knu|,  David  J.  Frame,  and  Myles  R.   Allen.  2009.  "Greenhouse-­‐gas  emission  targets  for  limi0ng  global  warming  to  2  degrees  C."    Nature.    vol.  458,  April  30.  pp.  1158-­‐1162.   <hJp://www.nature.com/nature/journal/v458/n7242/full/nature08017.html>   •  Pacala,  S.,  and  Rob  Socolow.  2004.  "Stabiliza0on  Wedges:  Solving  the  Climate  Problem  for  the  Next  50  Years  with  Current  Technologies   "    Science.    vol.  305,  no.  5686.  August  13.  pp.  968-­‐972.  [hJp://www.sciencemag.org/cgi/content/abstract/305/5686/968]   •  Williams,  James  H.,  Andrew  DeBenedic0s,  Rebecca  Ghanadan,  Amber  Mahone,  Jack  Moore,  William  R.  Morrow,  Snuller  Price,  and   Margaret  S.  Torn.  2011.  "The  Technology  Path  to  Deep  Greenhouse  Gas  Emissions  Cuts  by  2050:  The  Pivotal  Role  of  Electricity."     Science.    November  24.  [hJp://www.sciencemag.org/content/early/2011/11/22/science.1208365.abstract]     32  Copyright  Jonathan  Koomey  2013  
  33. 33. Extra  slides   33  Copyright  Jonathan  Koomey  2013  
  34. 34. Contributors  to  climate  change   through  2005   Source: IPCC 2007 (Working Group 1, the Physical Science Basis)Copyright  Jonathan  Koomey  2013   34  
  35. 35. Impacts  of  Uncertainty,  Learning,  and  Spillovers  (IPCC  AR4  ,  2007)   Figure  2.2.  Emissions  impacts  of  exploring  the  full  spectrum  of  technological  uncertainty  in  a  given  scenario  without  climate  policies.   Rela=ve  frequency  (percent)  of  130,000  scenarios  of  full  technological  uncertainty  regrouped  into  520  sets  of  technology  dynamics  with   their  corresponding  carbon  emissions  (GtC)  by  2100  obtained  through  numerical  model  simula=ons  for  a  given  scenario  of  intermediary   popula=on,  economic  output,  and  energy  demand  growth.  Also  shown  is  a  subset  of  13,000  scenarios  grouped  into  53  sets  of   technology  dynamics  that  are  all  "op=mal"  in  the  sense  of  sa=sfying  a  cost  minimiza=on  criterion  in  the  objec=ve  func=on.  The   corresponding  distribu=on  func=on  is  bi-­‐modal,  illustra=ng  "technological  lock-­‐in"  into  low  or  high  emissions  futures  respec=vely  that   arise  from  technological  interdependence  and  spillover  effects.  Baseline  emissions  are  an  important  determinant  for  the  feasibility  and   costs  of  achieving  par=cular  climate  targets  that  are  ceteris  paribus  cheaper  with  lower  baseline  emissions.  Adapted  from  Gritsevskyi   and  Nakicenovic,  2000.   35  Copyright  Jonathan  Koomey  2013  
  36. 36. Decanio  concludes…          “The  applica0on  of  general  equilibrium  analysis  to  climate   policy  has  produced  a  kind  of  specious  precision,  a  situa0on   in  which  the  assump0ons  of  the  analysts  masquerade  as   results  that  are  solidly  grounded  in  theory  and  the  data.   This  leads  to  a  tremendous  amount  of  confusion  and   mischief,  not  least  of  which  is  the  no0on  that  although  the   physical  science  of  the  climate  is  plagued  by  uncertain0es,  it   is  possible  to  know  with  a  high  degree  of  certainty  just  what   the  economic  consequences  of  alterna0ve  policy  ac0ons  will   be.”  (italics  in  original)   36  Copyright  Jonathan  Koomey  2013  
  37. 37. Fossil  fuel  resources  are  huge   37  Source:    Table  A-­‐1  from  Cold  Cash,  Cool  Climate,  mainly  using  GEA  data  2012   Note:    Current  annual     global  primary  energy     use  is  0.6  ZJ  (1  ZJ  =     1000  EJ  or  10e21  J),     which  is  about  30  TW.     Copyright  Jonathan  Koomey  2013  
  38. 38. MIT  level  1  case  =  1  doubling  of  CO2   equivalent  concentra0ons   38  Copyright  Jonathan  Koomey  2013  
  39. 39. Probability  of  <  2  C  increase  in  2100   rela0ve  to  preindustrial  0mes   39   Source:  Meinshausen  et  al.  2009  and  calcula0ons  in  Cold  Cash,  Cool  Climate.   Copyright  Jonathan  Koomey  2013  
  40. 40. Wedges:    A  useful  heuris0c   •  Originally  proposed  in  Pacala  and  Socolow   2004   •  One  wedge  =  25  GtC  over  50  years   •  Allows  simple  quan0ta0ve  characteriza0on  of   efforts  needed  to  reduce  emissions   •  Pile  up  wedges  to  reach  emissions  target   40  Copyright  Jonathan  Koomey  2013  
  41. 41. Stabiliza0on  wedges   41   Source:    Carbon  Mi0ga0on  Ini0a0ve  [hJp://cmi.princeton.edu/wedges/]   Copyright  Jonathan  Koomey  2013  
  42. 42. Individual  wedges  =  25  GtC  saved  over   50  years   42   Source:    Carbon  Mi0ga0on  Ini0a0ve  [hJp://cmi.princeton.edu/wedges/]   Copyright  Jonathan  Koomey  2013  
  43. 43. Simple  to  explain,  but  limited   •  Straighnorward  method   – Create  a  business-­‐as-­‐usual  baseline   – Analyze  how  much  each  op0on  will  save  rela0ve   to  the  baseline   •  Dependent  on  the  baseline  and  on  accurate   assessments  of  the  cost  effec0veness  of   op0ons  (as  are  all  conven0onal  forecas0ng   assessments)   •  Doesn’t  tell  you  how  many  wedges  you  need!   43  Copyright  Jonathan  Koomey  2013  

×