Poor use of science jeopardises climate lawsuits, finds Oxford University

LaRae Long Reforming Global Finance, SRI/ESG News, Latest Headlines

Press release

Existing obstacles to the success of climate-related lawsuits can be overcome with the use of recently-developed lines of scientific evidence, according to a new study from the Oxford Sustainable Law Programme and Environmental Change Institute, University of Oxford.

Filling the evidentiary gap in climate litigation will publish on 28.06.21 in Nature Climate Change, a leading interdisciplinary science journal, and is the first global study on the use and interpretation of climate-science evidence in lawsuits.

It assesses 73 lawsuits across 14 jurisdictions and finds that the evidence submitted by the litigants lags significantly behind the state-of-the-art in climate science, impeding claims that greenhouse-gas emissions have caused the impacts suffered by plaintiffs.

The majority of cases did not quantify the extent to which climate change was responsible for the climate-related events causing the impacts affecting plaintiffs, an important line of evidence since not all events occur due to climate change. Fewer still provided quantitative evidence linking defendants’ emissions with the plaintiffs’ injuries. 73% of cases did not refer to peer-reviewed evidence. And 26 claimed that weather events occurred due to climate change, without providing any evidence.

The findings make clear the vital importance of cutting-edge, peer-reviewed attribution science, which can provide exactly this kind of evidence and thus help prove causation. This would allow lawyers ascertain the prospects of successful litigation before cases reach court.

Plaintiffs have brought over 1,500 climate-related lawsuits worldwide, with the rate of claims being made increasing. High-profile cases such as Native Village of Kivalina v. ExxonMobil Corp, which was dismissed at the U.S. Court of Appeals, have shown how strong evidence of causation is critical to successful litigation.

Attribution science has recently been used to prove the impact of man-made climate change on extreme weather events like Hurricane Harvey.

As well as providing better evidence, attribution science can also inform the decision to pursue climate litigation cases, with uncertainties around some types of events (like droughts) being much higher than others (e.g. large scale extreme rainfall).

The authors call for greater awareness and use of cutting-edge attribution science when bringing climate litigation cases for court.

“Effective use of climate-science evidence in the courts could overcome existing obstacles to causality, set legal precedent for demonstrating causality with climate-science evidence, and make successful litigation on climate-change impacts feasible,” they state.

Rupert Stuart-Smith, lead author of the study says, ‘In recent weeks, successful lawsuits in the Netherlands, Germany, and elsewhere have seen courts demand countries and companies dramatically strengthen their climate targets. The power of climate litigation is increasingly clear.

‘However, many climate-related lawsuits have been unsuccessful. If litigation seeking compensation for losses suffered due to climate change is to have the best chance of success, lawyers must make more effective use of scientific evidence. Climate science can answer questions raised by the courts in past cases and overcome hurdles to the success of these lawsuits.’

Dr Friederike Otto, Associate Director of Oxford’s Environmental Change Institute, says, “in order to change the fate of the vast majority of climate litigation, courts and plaintiffs alike have to realise that science has moved on from ascertaining that climate change is potentially dangerous to providing causal evidence linking emissions to concrete damages.’

Professor Thom Wetzer, Founding Director of the Oxford Sustainable Law Programme, says, ‘Holding high-emission companies accountable for their contribution to climate change is key to driving systemic change and to protecting those most vulnerable to climate change impacts. Climate litigation aimed at generating that accountability is on the rise, but the results have been mixed.

Our research provides reason for optimism: with rigorous use of scientific evidence, litigators have room to be more effective than they currently are. It is now up to litigators to translate state-of-the-art science into high-impact legal arguments.’

Lead author Rupert Stuart-Smith is available for further comment and interview.

Media enquires:

[email protected]

[email protected]

About the Oxford Sustainable Law Programme

The Smith School of Enterprise and the Environment (SSEE) at the University of Oxford has recently established the Oxford Sustainable Law Programme (SLP) in close collaboration with the Faculty of Law and the Environmental Change Institute. This new multi-disciplinary research programme examines the use of the law in addressing the most pressing global sustainability challenges that humanity faces.

About the Environmental Change Institute

The Environmental Change Institute at the University of Oxford was established in 1991 to organise and promote interdisciplinary research on the nature, causes and impact of environmental change and to contribute to the development of management strategies for coping with future environmental change.

Climate litigation background practioner paper

Filling the evidentiary gap in climate litigation

Rupert F. Stuart-Smith1,2,3, Friederike E.L. Otto1,2, Aisha I. Saad1,4, Gaia Lisi1,5, Petra Minnerop6, Kristian Cedervall Lauta7, Kristin van Zwieten1,5, & Thom Wetzer1,3,5

1 Oxford Sustainable Law Programme, University of Oxford, Oxford, OX1 3QY, UK

2 Environmental Change Institute, University of Oxford, Oxford, OX1 3QY, UK

3 Institute for New Economic Thinking at the Oxford Martin School, University of Oxford, Oxford, OX1 3UQ, UK

4 Harvard Law School, Cambridge, MA, 02138, USA

5 Faculty of Law, University of Oxford, Oxford, OX1 3UL, UK

6 Durham Law School, Durham University, Durham, DH1 3LE, UK

7 Faculty of Law, University of Copenhagen, Copenhagen, DK-2300, Denmark

Lawsuits concerning the impacts of climate change make causal claims about the effect of defendants’ greenhouse gas (GHG) emissions on plaintiffs and have proliferated around the world. Plaintiffs have sought, inter alia, compensation for climate-related losses and to compel governments to reduce their GHG emissions. To date, most of these claims have been unsuccessful. Here, we assess the scientific and legal bases for establishing causation and evaluate judicial treatment of scientific evidence in 73 lawsuits. We find that the evidence submitted and referenced in these cases lags considerably behind the state-of-the-art in climate science, impeding causation claims. We conclude that greater appreciation and exploitation of existing methodologies in attribution science could address obstacles to causation and improve the prospects of litigation as a route to compensation for losses, regulatory action, and emission reductions by defendants seeking to limit legal liability.

Plaintiffs have brought over 1,500 climate-related lawsuits worldwide, and the number of claims filed continues to increase1. Without effective non-judicial mechanisms providing compensation for climate-related loss and damage, plaintiffs have filed lawsuits seeking financial remedies from high-emitting corporations for losses suffered due to climate change2. Robust scientific evidence is critical to the success of such claims3–5. For example, claims for compensatory damages must demonstrate a causal link between the defendant’s behaviour and the plaintiff’s injury. Recognising this, recent literature has drawn on analogous categories of case, such as toxic torts, in which modified causation tests reconcile legal causal analysis with scientific evidence that demonstrates multiple entities’ contribution to the risk of harm5. However, despite developments in scientific6 and legal theory5 on causation and attribution of climate change damages, compensatory-damages claims have been unsuccessful. Other lawsuits challenge inadequate state and corporate climate change mitigation targets and policies7. To establish admissibility, these claims may also rely on courts finding that emissions resulting from defendants’ policies led to impacts affecting the plaintiffs.

A claim must first meet procedural requirements that render it admissible, including ‘standing’, which establishes that plaintiffs have legally protected interests that entitle them to bring the claim. Demonstrating a connection between defendants’ actions and plaintiffs’ injuries may contribute to meeting the standing requirement. To date, admissibility has been the primary outcome-determinative obstacle for climate litigation8. If claims pass procedural hurdles, more stringent causation standards for attributing losses to defendants’ actions are applied to determine plaintiffs’ entitlement to relief (remedies).

To establish causation, the court needs to receive scientific evidence, including through expert witness testimony, bounded by rules on the expert’s independence and duties to the court, and amicus briefs submitted by third parties. Courts interpret this evidence through the lens of legal reasoning5. Such legal rules, encompassing both “normative correctives” applied to evidence5 and the standard of proof, vary between jurisdictions (Supplementary Information, ‘Standards of proof in law’). Across jurisdictions, however, courts ascribe value to authoritative sources of evidence such as Intergovernmental Panel on Climate Change reports or peer-reviewed publications9. The a priori question motivating this research is whether the scientific evidence courts were provided and have access to reflects the ‘state of the art’.

Challenges to scientifically demonstrating causation

The fields of event attribution and trend detection and attribution evaluate the causal relationships at issue in climate litigation. Attribution science comprises methods which generally use counterfactual inquiry to link observed trends or changes in probability or intensity of climate-related events to human influence. Existing methods can quantify the contribution of GHG emissions to specific events, including (1) extreme events, including storms10, droughts11, heatwaves12, or floods13, (2) long-term trends in glacier lengths14 or sea levels, and (3) persistent changes, for instance in mean temperatures or precipitation15. Despite the clear relevance of attribution-science evidence, there is limited precedent for courts to base findings of causation on such evidence, partly due to its relative novelty. In common with most forms of scientific evidence, including regarding the health consequences of exposure to pharmaceutical substances, courts’ use of climate-change attribution evidence to assess causation is subject to several considerations.

First, GHG emissions are fungible and climate-change impacts result from the cumulative emissions of multiple parties. This underlies use of ‘market share theory’5, an approach which, following precedent in pharmaceutical and tobacco litigation, allocates damages among defendants according to the portion of emissions for which they are responsible3. Market share theory has been used to allocate damages in cases where losses derive linearly or non-linearly from multiple entities’ actions, to reduce evidentiary challenges in calculating defendants’ exact contributions to losses. It is recognised that this simplistic allocation basis may not exactly reflect defendants’ contributions to losses. Attribution-science evidence quantifying individual defendants’ contributions to plaintiffs’ losses directly could be used instead of the market-share approach. This may be helpful for impacts with non-linear emissions-impact relationships16.

Second, extreme weather events generally occur due to the combination of human influence and the underlying chance that they would occur naturally. Climate lawsuits aim to follow past science-based tort litigation such as product liability cases, which sought compensation for medical conditions arising from exposure to pharmaceutical substances, tobacco smoke and asbestos3. Such conditions may have occurred in the absence of this exposure. Similarly, climate change may increase the likelihood or intensity of an extreme event so attribution statements are typically probabilistic. Legal frameworks exist for holding defendants responsible for changes in an event’s probability even if they make only a partial contribution to this change3,5. Standards of  legal proof for probabilistic evidence may differ from the likelihoods evaluated in scientific assessments3 (Supplementary Information, ‘Standards of proof in law’).

Finally, in common with many types of expert evidence, attribution findings are question-dependant17 and influenced by the event definition. Contrasting approaches to framing attribution questions produce quantitatively differing results, while remaining scientifically valid18. These approaches vary in their alignment with the logic of legal causation (Supplementary Information, ‘Approaches to conducting attribution studies’). Further, attribution assessments apply statistical methods to assess confidence and uncertainty in results19. Sources of uncertainty include limitations in model representations of the climate system and climatic observations for model evaluation. Uncertainty is assessed based on physical understanding of atmospheric processes and their representation in models, agreement between models and observations, and the quality of available evidence17.

Climate science evidence in past litigation

How well does the evidence submitted to climate-related lawsuits compare to the state-of-the-art in climate science? A growing body of literature explores the role attribution science can play in climate-related litigation, from legal and scientific perspectives3,5,20,21. This is the first global study of the use and interpretation of climate-science evidence in lawsuits. We identify the scientific evidence needed to make successful causal arguments and analyse the evidence provided by plaintiffs and defendants in 73 cases across 14 jurisdictions (Methods, Table 1, listed in full in Supplementary Table 1).

We consider cases that advance claims that defendants’ GHG emissions impacted plaintiffs (Methods). Our analysis finds that most cases did not quantify the extent to which alleged impacts are attributable to climate change and fewer still provided quantitative evidence linking defendants’ emissions and plaintiffs’ injuries. Although some recently-filed cases provided stronger evidence, e.g., refs. 22–24, 73% of cases did not refer to peer-reviewed attribution studies. Most of the cases that referred to attribution findings did so to establish general links between emissions and certain types of climate-change impacts, rather than to attribute losses sustained by plaintiffs. Moreover, despite substantial evidence that climate change increases the probability and intensity of a range of events, not all climate-related hazards are affected by climate change17. This underlines the importance of providing evidence specific to the impact for which a causal link is alleged3,5.

54 cases cited meteorological extremes as the cause of alleged impacts (Supplementary Table 2). Of these cases, 26 claimed that weather events occurred due to climate change, without providing evidence. A further 6 provided no quantitative estimate of the influence of climate change on the event’s magnitude or probability. Losses stemming from extreme weather events cannot be presumed attributable to climate change a priori25. Even for events where climate change plays a substantial role, GHG emissions will increase the event’s likelihood or intensity, rather than being its sole cause26. Further, many legal claims concerned events, such as tropical cyclones, for which evidence of human influence is limited27, or for which technical obstacles to conducting attribution assessments exist10. However, climate change substantially influences other events with considerable humanitarian consequences, such as heatwaves and some droughts28. Establishing causation is easiest for events where the influence of climate change is greatest: these are logical subjects of climate lawsuits.

38 cases made causal allegations related to sea-level-rise impacts. Human influence is very likely the dominant cause of global-mean sea-level change since 197029. However, regional sea levels may differ due to natural and human-influenced processes, modifying the anthropogenic contribution to impacts. These processes include regional variation in thermal expansion and ocean and atmospheric circulation, glacial-isostatic adjustment, and land subsidence29. Factors such as subsidence outweigh the effect of human-induced climate change in some regions30. 12 cases concern impacts of glacier retreat (Supplementary Table 2), an established consequence of climate change14. However, individual glaciers’ response to climate change depends on their geographic and climatic settings31. Linking emissions to glacial and sea-level-rise impacts at the appropriate level of specificity for establishing legal causation may require evidence on these local factors (Box 1). This is uncommon in the cases analysed.

The evidence needed for legal causation

Many of the lawsuits assessed here are at the early stages of adjudication or were dismissed before substantive consideration of causation arguments. Our analysis of courts’ interpretation of scientific evidence considers all lawsuits in which courts evaluate this evidence, including cases that were ultimately dismissed for reasons unrelated to issues of causation.

Courts evaluate scientific evidence by applying tests which set evidentiary thresholds for establishing causation. Across jurisdictions, causal analyses focus on the relationship between defendants’ conduct and plaintiffs’ losses, but the tests and standards of proof applied vary (Supplementary Information, ‘Causation tests used in climate litigation’). Our analysis finds that some courts required defendants to be the sole or material contributor to the alleged harms32,33; others have applied causality tests more flexibly. However, plaintiffs have been unable to overcome even the more flexible causation tests applied in several jurisdictions which ask if damages are ‘fairly traceable’ to defendants’ actions34–36. This is typically due to courts’ finding that the evidence provided does not substantiate the connection between individual emitters’ actions and plaintiffs’ losses. In our view, in most cases concerning impacts for which the causal link to climate change does exist, existing scientific methodologies could fill the evidentiary gaps identified by courts.

Our analysis shows that when courts considered evidence on causation, they typically found that plaintiffs failed to demonstrate that defendants’ emissions caused the alleged impacts. Plaintiffs have claimed, with or without supporting evidence, that climate change caused certain harms and therefore that defendants’ GHG emissions render them liable for a portion of these losses. In claims seeking compensation for climate-change impacts brought in the US, Germany and New Zealand, courts found that plaintiffs failed to demonstrate that their injuries would not have occurred in the absence of defendants’ emissions. These findings were due to defendants’ small contribution to climate change37, the high number of emitters responsible33, the absence of a method for discerning entities responsible for impacts38, and the lack of a direct causal link between defendants’ actions and losses32,34,35,39–41. In some cases, courts have incorrectly stated the infeasibility of attributing climate impacts to individual emitters33. Similarly, in claims seeking emission reductions, courts found that plaintiffs failed to demonstrate how a positive judgement would reduce climate-change impacts33.

In our view, these findings are not a product of technical limitations preventing the production of required evidence. Scientific evidence on the influence of individual actors’ GHG emissions is available for existing42–44 and projected impacts45.

Despite inter-jurisdictional variation in the standards of proof for causation, there are shared characteristics of evidence needed to support claims. Cases seeking compensation for adaptation costs must demonstrate that (1) risk mitigation measures are required, and (2) the cost sought is a consequence of climate change resulting, in part, from defendants’ actions. Evidence submitted in cases of this type generally demonstrated that the hazards threatening plaintiffs’ property and wellbeing were substantial and required hazard-mitigation measures. However, the evidence for causation was typically much weaker, despite availability of the scientific methodologies needed to generate it (see above). Cases claiming that adaptation policies are inadequate may argue that losses were foreseeable due to climate change (e.g., ref. 46) which may in turn be supported by attribution-science evidence.

Lawsuits seeking emission reductions considered in this analysis (Methods) make causal claims about the impact of a positive judgment on projected impacts, usually to establish legal standing. Our analysis shows that a case was dismissed by a court in the U.S.47 partly due to plaintiffs’ failure to establish standing because they did not demonstrate that emission reductions would reduce the impacts they experience. Scientific evidence linking defendants’ GHG emissions to future impacts would address this finding.

The evidence gap in climate litigation

The existing literature suggests that there are real challenges to satisfying causation tests in climate-related litigation3,5. In our view, at least some of these challenges can be addressed by better exploiting scientific developments, particularly in the field of climate-change attribution. It is therefore crucial that plaintiffs have access to this science, and that this evidence is brought to the attention of courts. Our analysis identifies three areas where scientific research could address existing evidentiary shortfalls: (1) attribution of climate-change impacts to individual emitters of GHGs42–45 (possible using existing attribution-science methods), (2) research on the foreseeability of climate-change impacts resulting from future emissions, and (3) research disentangling the legally-relevant social and physical drivers of climate risks and impacts. The latter recommendation is the most methodologically challenging, due to challenges in quantifying non-climatic contributions to impacts.

Establishing a defendant’s contribution to plaintiffs’ losses has presented a key challenge in cases in which causal claims have been adjudicated. However, current scientific methodologies enable quantifying individual emitters’ marginal contributions to extreme weather events and slow-onset changes42–44. Methods for modelling the response of the climate system to excluding certain GHG emissions exist48 and have been applied to evaluate the contribution of countries’ emissions to extreme-weather probabilities42. Evidence of this type appears capable of plugging evidentiary gaps identified in some prior cases32,34,35,37–41. In some lawsuits seeking emission reductions, plaintiffs argued that implementing emission-reduction measures would reduce otherwise anticipated impacts, typically to establish standing. In one case, the court found that defendants had not breached legal duties as plaintiffs failed to demonstrate how requested emission reductions would affect the magnitude of climate-change impacts47. Counterfactual analyses could support these claims by assessing individual emitters’ contributions to projected changes in impacts under future climate change (e.g., ref. 45).

Our analysis also suggests that there is scope to develop evidence which provides a more complete assessment of the drivers of losses on which claims are based. Climate-related losses result from physical events, about which attribution science provides evidence, mediated by communities’ exposure and vulnerability to these hazards49. Research that evaluates the relative importance of different drivers and which of these factors constitute the legally relevant cause(s) of losses may support establishing defendants’ liability for climate impacts.

Finally, although most attribution findings are presented in terms of changing event probability, 48% of cases analysed made claims about the impact of climate change on the magnitude, rather than probability, of events. Attribution evidence on event intensities can be provided using existing methods (e.g., refs. 10,12) and may be more informative for some impacts, like adverse effects on human health50. Such evidence may also strengthen legal claims by providing an evidentiary basis for asserting that a defendant’s conduct has made a plaintiff worse off. Attributing changes in event intensity to a defendant aligns with the logic of the ‘but-for’ test in law and may satisfy causation tests by showing how the magnitude of a harm was altered by an individual defendant’s conduct. Future research could determine whether probability or intensity-based attribution metrics best assess the contribution of climate change to different types of losses.

Individuals, communities and activist non-governmental organisations have increasingly turned to the courts for relief for costs associated with present-day climate change impacts, or to avoid future impacts. While the cases we examined may not have been dismissed solely on evidentiary grounds, existing case law provides insight into the challenges that need to be addressed if future climate-related lawsuits are to succeed. As plaintiffs overcome initial procedural hurdles, such as standing, evidence on causation will become more determinative of case outcomes. Attribution science is a fundamental source of evidence for informing and substantiating causal claims about climate-change impacts, hitherto underutilised in law21.

Our analysis of 73 cases from 14 jurisdictions finds that, in contrast to previous interpretations8, limitations in the scientific evidence provided to past lawsuits may have hindered the making of findings of causation. This conclusion is based on the finding that such evidence lags significantly behind the state-of-the-art in climate science. To maximise the chances of establishing causation in the courts, plaintiffs should ensure that (1) cases filed concern impacts that are demonstrably attributable to climate change, and (2) that evidence submitted to the courts clearly substantiates the alleged relationship between defendants’ emissions and plaintiffs’ losses.

Across jurisdictions, courts have found insufficient links between defendants’ emissions and plaintiffs’ injuries. In some cases courts have explicitly, and incorrectly, stated the infeasibility of scientifically attributing climate impacts to individual emitters33,34. This underlines the importance of dialogue between the legal and scientific communities on the factual basis for causal claims. Better dialogue would ensure that lawyers are aware of, and able to request, evidence that can be used to robustly evaluate causal claims. This could be achieved through selection of independent experts with relevant scientific expertise, such as Intergovernmental Panel on Climate Change authors, as well as (outside the context of any particular case) funding to support the continuing development of the science in this field. Effective use of climate-science evidence in the courts could overcome existing obstacles to causality, set precedent for demonstrating legal causality with climate-science evidence, and make successful litigation on climate-change impacts feasible.


  1. Setzer, J. & Byrnes, R. Global trends in climate change litigation: 2020 snapshot. https://www.lse.ac.uk/granthaminstitute/wp-content/uploads/2020/07/Global-trends-in-climate-change-litigation_2020-snapshot.pdf (2020).
  2. Toussaint, P. Loss and damage and climate litigation: The case for greater interlinkage. Rev. Eur. Comp. Int. Environ. Law (2020) doi:10.1111/reel.12335.
  3. Marjanac, S. & Patton, L. Extreme weather event attribution science and climate change litigation: an essential step in the causal chain? J. Energy Nat. Resour. Law 36, 265–298 (2018).
  4. McCormick, S. et al. Strategies in and outcomes of climate change litigation in the United States. Nat. Clim. Chang. 8, 829–833 (2018).
  5. Minnerop, P. & Otto, F. E. L. Climate change and causation: joining law and climate science on the basis of formal logic. Buffalo Environ. Law J. 27, 49–86 (2020).
  6. Hannart, A., Pearl, J., Otto, F. E. L., Naveau, P. & Ghil, M. Causal Counterfactual Theory for the Attribution of Weather and Climate-Related Events. Bull. Am. Meteorol. Soc. 97, 99–110 (2016).
  7. Peel, J. & Osofsky, H. M. A Rights Turn in Climate Change Litigation? Transnatl. Environ. Law 7, 37–67 (2018).
  8. Burger, M., Horton, R. M. & Wentz, J. The Law and Science of Climate Change Attribution. Columbia J. Environ. Law 45, 57–241 (2020).
  9. Lee, M. The Sources and Challenges of Norm Generation in Tort Law. Eur. J. Risk Regul. 9, 34–47 (2018).
  10. van Oldenborgh, G. J. et al. Attribution of extreme rainfall from Hurricane Harvey, August 2017. Environ. Res. Lett. 12, 124009 (2017).
  11. Cowan, T., Undorf, S., Hegerl, G. C., Harrington, L. J. & Otto, F. E. L. Present-day greenhouse gases could cause more frequent and longer Dust Bowl heatwaves. Nat. Clim. Chang. 10, 505–510 (2020).
  12. Vautard, R. et al. Human contribution to the record-breaking June and July 2019 heatwaves in Western Europe. Environ. Res. Lett. 15, 094077 (2020).
  13. Schaller, N. et al. Human influence on climate in the 2014 southern England winter floods and their impacts. Nat. Clim. Chang. 6, 627–634 (2016).
  14. Stuart-Smith, R. F., Roe, G. H., Li, S. & Allen, M. R. Increased outburst flood hazard from Lake Palcacocha due to human-induced glacier retreat. Nat. Geosci. 14, 85–90 (2021).
  15. Hegerl, G. C. et al. Climate Change Detection and Attribution: Beyond Mean Temperature Signals. J. Clim. 19, 5058–5077 (2006).
  16. Harrington, L. J. & Otto, F. E. L. Adapting attribution science to the climate extremes of tomorrow. Environ. Res. Lett. 13, 123006 (2018).
  17. Otto, F. E. L. et al. Toward an Inventory of the Impacts of Human-Induced Climate Change. Bull. Am. Meteorol. Soc. 101, E1972–E1979 (2020).
  18. Otto, F. E. L., Massey, N., Van Oldenborgh, G. J., Jones, R. G. & Allen, M. R. Reconciling two approaches to attribution of the 2010 Russian heat wave. Geophys. Res. Lett. 39, (2012).
  19. Philip, S. et al. Attribution Analysis of the Ethiopian Drought of 2015. J. Clim. 31, 2465–2486 (2018).
  20. Allen, M. R. et al. Scientific challenges in the attribution of harm to human influence on climate. Univ. PA. Law Rev. 155, 1353–1400 (2007).
  21. Marjanac, S., Patton, L. & Thornton, J. Acts of God, human influence and litigation. Nat. Geosci. 10, 616–619 (2017).
  22. Sacchi et al. v. Argentina et al., Communication to the Committee on the Rights of the Child (Sept. 23, 2019).
  23. Bushfire Survivors for Climate Action Inc v Environment Protection Authority [2020] NSWLEC 152.
  24. In re Greenpeace Southeast Asia and Others, Case No. CHRNI-2016-0001 (Sept. 9, 2019).
  25. Otto, F. E. L. Attribution of Weather and Climate Events. Annu. Rev. Environ. Resour. 42, 627–646 (2017).
  26. van Oldenborgh, G. J. et al. Extreme heat in India and anthropogenic climate change. Nat. Hazards Earth Syst. Sci. 18, 365–381 (2018).
  27. Patricola, C. M. & Wehner, M. F. Anthropogenic influences on major tropical cyclone events. Nature 563, 339–346 (2018).
  28. Otto, F. E. L. et al. Anthropogenic influence on the drivers of the Western Cape drought 2015–2017. Environ. Res. Lett. 13, 124010 (2018).
  29. Oppenheimer, M. et al. Sea Level Rise and Implications for Low-Lying Islands, Coasts and Communities. in IPCC Special Report on the Ocean and Cryosphere in a Changing Climate (eds. Pörtner, H.-O. et al.) 321–446 (2019).
  30. Ward, P. J., Marfai, M. A., Yulianto, F., Hizbaron, D. R. & Aerts, J. C. J. H. Coastal inundation and damage exposure estimation: a case study for Jakarta. Nat. Hazards 56, 899–916 (2011).
  31. Sagredo, E. A., Rupper, S. & Lowell, T. V. Sensitivities of the equilibrium line altitude to temperature and precipitation changes along the Andes. Quat. Res. 81, 355–366 (2014).
  32. City of Oakland v BP p.l.c., 969 F.3d 895 (9th Cir. 2020).
  33. Lliuya v RWE AG, District Court of Essen, Judgment of 15 December 2016 – 2 O 285/15.
  34. Native Village of Kivalina v. ExxonMobil Corp., 696 F.3d 849 (9th Cir. 2012).
  35. Comer v Murphy Oil USA, Inc., 718 F.3d 460 (5th Cir. 2013).
  36. American Electric Power Co. v Connecticut, 564 U.S. 410 (2011).
  37. Smith v Fonterra Co-operative Group Ltd [2020] NZHC 419.
  38. California v General Motors Corp., No. C06-05755 MJJ, WL 2726871 (N.D. Cal. 2007).
  39. Board of County Commissioners of Boulder County v Suncor Energy (U.S.A.), Inc., 405 F. Supp. 3d 947 (D. Colo. 2019).
  40. Rhode Island v Chevron Corp., 393 F. Supp. 3d 142 (D.R.I. 2019).
  41. Mayor and City Council of Baltimore v BP p.l.c., 388 F. Supp. 3d 538 (D. Md. 2019).
  42. Otto, F. E. L., Skeie, R. B., Fuglestvedt, J. S., Berntsen, T. & Allen, M. R. Assigning historic responsibility for extreme weather events. Nat. Clim. Chang. 7, 757–759 (2017).
  43. Licker, R. et al. Attributing ocean acidification to major carbon producers. Environ. Res. Lett. 14, 124060 (2019).
  44. Ekwurzel, B. et al. The rise in global atmospheric CO2, surface temperature, and sea level from emissions traced to major carbon producers. Clim. Change 144, 579–590 (2017).
  45. Nauels, A. et al. Attributing long-term sea-level rise to Paris Agreement emission pledges. Proc. Natl. Acad. Sci. 116, 23487–23492 (2019).
  46. Illinois Farmers Insurance Co. v Metropolitan Water Reclamation District of Greater Chicago, No. 2014CH06608 (Ill. Cir. Ct. 2014).
  47. Sinnok et al. v State of Alaska et al., No. 3AN-17-09910 CI (Alaska Super. Ct. 2018).
  48. Skeie, R. B. et al. Perspective has a strong effect on the calculation of historical contributions to global warming. Environ. Res. Lett. 12, 024022 (2017).
  49. Clarke, B. J., E. L. Otto, F. & Jones, R. G. Inventories of extreme weather events and impacts: Implications for loss and damage from and adaptation to climate extremes. Clim. Risk Manag. 32, 100285 (2021).
  50. Ebi, K. L. et al. Using Detection And Attribution To Quantify How Climate Change Is Affecting Health. Health Aff. 39, 2168–2174 (2020).


We are grateful to P. Guzik for research assistance and to C.-F. Schleussner for valuable comments on the manuscript. We gratefully acknowledge support from the Foundation for International Law for the Environment, R.S.-S. acknowledges support from the Natural Environment Research Council grant NE/S007474/1, Climate Analytics, and the Oxford Martin Programme on the Post-Carbon Transition.

Author contributions

All authors planned the analyses which R.S.-S., A.S., and G.L. performed. All authors contributed to the interpretation of the results and to writing the manuscript.

Competing interests

The authors declare no competing financial or non-financial interests.

Box 1: Scientific evidence in past lawsuits
Case study 1: Lliuya v. RWE AG

In 2015, Saúl Luciano Lliuya filed a case against a German energy company, RWE, seeking compensation for the pro rata cost of measures taken to protect his property against climate change impacts, based on Article 1004 of the German Civil Code. This article applies the ‘conditio sine qua non’ test for causality: the plaintiff’s injury would not have occurred fully or partially if not for the defendant’s activity. The District Court of Essen dismissed the case, reasoning that RWE would not qualify as “disturber by conduct” under Section 1004 of the German Civil Code given that the number of contributors to climate change render attributing individual damages to specific actors impossible. On appeal, the Higher Regional Court of Hamm accepted that the case has been conclusively argued and initiated an evidentiary phase. The first stage of the evidentiary phase assesses whether the claimant’s property is indeed seriously threatened by a potential glacial lake outburst flood, including a court visit to Peru. Subsequently, the court will assess evidence around whether the heightened flood risk is attributable to RWE, and finally whether the damage contributed by the defendant is measurable and can be calculated.

Recent research has demonstrated that it is indeed possible to provide an attribution assessment spanning the full causal chain in this setting14.

Case study 2: Native Village of Kivalina v. ExxonMobil Corp.

In 2008, the Alaskan village of Kivalina filed a public nuisance claim seeking to hold 24 energy companies liable for coastal erosion, alleged to be due to climate change, which threatened the village, requiring it to be relocated inland. The Court applied the ‘fair traceability’ test, which requires the plaintiff to demonstrate a substantial likelihood that the defendant’s conduct (emissions) was the seed of the plaintiff’s injury. The U.S. Court of Appeals for the Ninth Circuit dismissed the lawsuit, in part due to the plaintiff’s failure to establish standing through demonstrating causation. The Court ruled that the causal link between the defendant’s activities and the plaintiff’s damage was too indirect since “it is not plausible to state which emissions – emitted by whom … and at what place in the world – ‘caused’ Plaintiffs’ alleged global-warming related injuries [because there are] a multitude of ‘alternative culprit[s]’ allegedly responsible for the various chain [sic] of events allegedly leading to the erosion of Kivalina”34.

While attribution-science research specifically addressing the causal link alleged in this case (i.e., specific to the impacts at this location) does not yet exist, it would be scientifically possible to conduct a study to determine the extent to which defendants’ emissions led to the losses experienced by Kivalina.




Table 1

Country Number of cases Successful Dismissed Pending
Australia 3 0 1 2
Belgium 1 0 0 1
Canada 3 0 0 3
Colombia 1 1 0 0
Germany 2 0 1 1
Ireland 1 1 0 0
Netherlands 2 1 0 1
New Zealand 4 1 2 1
Pakistan 2 1 0 1
Philippines 1 0 0 1
South Korea 1 0 0 1
Switzerland 1 0 1 0
Uganda 1 0 0 1
USA 45 3 21 21
International jurisdictions 5 0 2 3

Table 1: Number of cases and case outcome (as of November 2020) for lawsuits considered in this analysis, categorised by jurisdiction. Cases are listed in full in Supplementary Table 1.


Scope of case-law analysis

Our case law analysis covers cases that make claims about the past, present and future impacts of climate change and the relationship between GHG emissions and these impacts. We sought to analyse all lawsuits, brought using a variety of legal bases, where courts considered scientific evidence on causation. Cases were selected through a systematic review of the Sabin Center for Climate Change Law at Columbia Law School’s climate litigation database (http://climatecasechart.com/) conducted in May 2020. The database categorises cases according to criteria such as the case objective (for instance, ‘actions seeking adaptation measures) or the principal laws under which the claim was brought. Cases within relevant ‘case categories’ were identified and assessed for alignment with the project scope, including by searching for the presence of keywords pertaining to climate-change impacts in complaints. Claims in the following categories of the Sabin Center database were assessed for their relevance to the project scope: Adaptation (including ‘Actions seeking money damages for losses’), Public Trust Claims, Common Law Claims, Climate Change Protestors and Scientists, and claims brought under the National Environmental Policy Act (U.S.); Suits against Corporations, Public Trust, Human Rights, GHG Emissions Reduction and Trading, Climate Adaptation (non-U.S.). The review of the Sabin Center database was supplemented by a review of cases cited in relevant academic literature, e.g., refs. 3,8,51. A review of news articles was used to identify cases filed during the period over which we conducted the analysis, in September 2020.

We identified a list of 73 cases from 14 national jurisdictions which either refer directly to attribution science or use evidence to claim that climate change is or will become responsible for particular damages. The cases determined to be within the project scope do not include those that refer only to general impacts of climate change. Rather, we only consider cases that make causal claims about impacts in specific locations, for which attribution-science evidence is relevant.

Cases identified within this scope fell into the categories of tort, human rights, public trust, takings, contract, judicial review, and federal statutory claims. The tort and human rights claims typically sought remedies in the form of financial compensation for climate change impacts or emission reductions. All ‘damage liability’ suits that were filed relied on causal argumentation about the relationship between emissions and impacts. By contrast, only a subset of pro-regulatory claims for emission reductions – typically brought as human rights claims, or, for U.S. cases, under the public trust doctrine – have sought to link the actions of defendants to specific climate-that losses. Those that did were included within the project scope (Supplementary Table 1). Similar lawsuits that did not make claims that defendant emissions caused specific climate-related losses were excluded from the analysis as attribution science evidence was not relevant to the legal arguments made in these cases. We note that, in future, attribution-science evidence may be used by a wider range of lawsuits3,21.

In our evaluation of the scientific evidence provided by parties to the case, we considered for inclusion publicly available case documents submitted by plaintiffs, defendants, and other parties. Such documents included complaints, petitions, responses, motions, and amicus briefs. A list of key documents which were used to inform our findings, for each case, is included in Supplementary Table 1. We included cases in early stages of litigation (for instance, for which only the complaint was available) but excluded two cases believed to be within the scope of the research for which either no court documents were available (Petition of Torres Strait Islanders to the United Nations Human Rights Committee), or for which court documents, or high-quality translations of them, were only available in languages not spoken by the research team (Federal Environmental Agency (IBAMA) v. Siderúrgica São Luiz Ltd. and Martins). Documents in English, Spanish, and French were included. In a small number of cases, such as Friends of the Irish Environment v. Ireland, the Petition or complaint could not be made public and so only the legal analysis of the judgement and appeal was possible.

We then identified court documents which introduced evidence or claims regarding the potential causal relationship between greenhouse-gas emissions and climate change impacts. As far as possible this was approached systematically through the use of a series of relevant search terms. We reviewed non-searchable documents manually.

Regarding the judicial assessment of the scientific evidence provided to cases, we considered final decisions and opinions, as well as court documents released during the pre-trial stage, such as orders to grant or dismiss motions to remand, orders to accept amicus briefs, or questions directed to expert witnesses.

Scientific analysis

The scientific analysis seeks to understand the nature of the attribution-science evidence used in climate-related legal cases. We identify key components of the evidence in each case within the project scope, through a review of complaints, expert testimony, and defendants’ responses. These factors include (i) the type of attribution evidence provided in the case, (ii) the strength of attribution evidence as presented in the case (and as compared with that available in the published literature), (iii) the methodologies used to quantify attributable damages, and (iv) the content of expert testimony.

For each lawsuit, we evaluated the type of evidence provided on causation according to the source of the evidence used for alleging the existence of a causal link between GHG emissions and impacts. Evidence from Intergovernmental Panel on Climate Change (IPCC) reports has commonly been submitted by plaintiffs to substantiate claims, including in unsuccessful cases. IPCC attribution assessments of climate change impacts are typically regional or global in scale, and therefore do not provide evidence specific to damages alleged in most cases. To understand how different forms of attribution knowledge affect legal interpretation, we differentiate between cases that extrapolate from regional / global attribution statements, those where attribution is presumed based on consistency of impacts with published and peer-reviewed climate projections, those for which attribution is presumed with no supporting evidence, and those that use an attribution study into the causes of the impacts sustained by plaintiffs, and whether or not it has been peer-reviewed.

Further to our assessment of cited attribution evidence, where relevant, our analysis considers the expert testimony admitted to cases. This includes considering which experts are called upon and whether they have published in the field of attribution science.

Our analysis of the strength of attribution evidence used in within-scope cases considers the magnitude of human influence on the climatic event relevant to the litigation, and the confidence in and uncertainty associated with these findings. We can therefore assess the extent to which the magnitude of human influence found affects the legal determination of causality, in the context of jurisdiction-specific standards, such as the ‘doubling of the risk’ test in England and Wales3. Further, we considered any evidence provided about plaintiffs’ pre-existing (independent of climate change) vulnerability to climate impacts, including defendants’ use of pre-existing vulnerabilities to counter plaintiffs’ arguments that there is a causal link between GHG emissions, and the impacts experienced.

Finally, we consider how assessment of the attribution of physical impacts corresponds to the damages sought in the case. For each lawsuit, we identify whether plaintiffs sought to claim that the entire cost of the damages was attributable to anthropogenic GHG emissions, or whether damages were quantified according to attributable changes in event intensity or probability. We also consider whether economic and/or non-economic losses are quantified, what relief was requested, and how a defendant’s contribution to losses was quantified in determining the requested relief.

Our assessment of the scientific evidence provided in the cases analysed includes the collection of qualitative – such as whether the type of evidence provided on causation was a peer-reviewed attribution study or a regional attribution statement from an IPCC report – and quantitative – for instance on the magnitude of economic relief requested – data, supplemented by expert judgements.

Legal analysis

For each lawsuit within the scope of the analysis, for which courts’ opinions, decisions, orders, or judgements were available (Supplementary Table 1), we evaluated how the courts have assessed and interpreted the scientific evidence on causation brought before them. The immediate obstacles to the success of litigation have largely been on admissibility and procedural grounds8, with causation evidence generally not being outcome determinative. Our analysis, therefore, evaluates judicial opinion on causal arguments irrespective of the immediate reason for the case outcome and identifies how attribution science evidence can best support litigation if non-evidentiary, procedural obstacles are overcome. We identify trends across jurisdictions and time in how attribution science has been used by litigants and interpreted by courts to overcome procedural and evidentiary hurdles to successful litigation.

Our legal analysis considers courts’ use of scientific evidence in establishing the admissibility of a case (as used under English law: whether a court has jurisdiction to adjudicate on a matter) or in satisfying procedural and substantial requirements of the law. For a case to be admissible, plaintiffs must have standing to bring the lawsuit, and defendants must have legal duties to plaintiffs which they may have breached. We consider whether a court finds that a sufficient link exists between the losses experienced by plaintiffs and climate change, as well as the link between defendants’ emissions and these losses. We also assess the extent to which courts find that plaintiffs’ losses were foreseeable to defendants, and the admissibility of expert testimony. We further identify whether courts find defendants to be liable for losses. Our analysis focuses on the extent to which deficits in scientific evidence played a role in a courts’ assessments of these issues. We also consider the approaches courts have taken in responding to scientific arguments concerning defendants’ contributions to alleged impacts, including the apportioning of liability based on market share of emissions or on defendants’ marginal contribution to impacts.

In cases in which courts have directly engaged with causation arguments, we identify the tests applied by the courts, how these have varied between jurisdictions and over time, and the differences between these tests. We then consider the implications of these causation tests for the courts’ evaluation of scientific evidence and the standards of proof that plaintiffs need to meet to bring successful climate change lawsuits.

Finally, we provide an overall assessment of the role of science in case outcomes and an analysis of the required changes in scientific evidence that would improve the likelihood of establishing causation in future cases. We contextualise this analysis in the challenges faced when applying existing legal concepts to novel, global challenges for which they may not be well suited52.

Data availability: Case documents were sourced primarily from the Sabin Center for Climate Change Law at Columbia Law School’s Climate Change Litigation database (http://climatecasechart.com/). Where relevant case documents were unavailable on this database, they were sourced from individual courts’ public databases or from Westlaw.


  1. Banda, M. L. Climate Science in the Courts A Review of U.S. and International Judicial Pronouncements (2020); https://www.eli.org/sites/default/files/eli-pubs/banda-final-4-21-2020.pdf.
  2. Fisher, E., Scotford, E. & Barritt, E. The Legally Disruptive Nature of Climate Change. Mod. Law Rev. 80, 173–201 (2017).