Perceptions of carbon dioxide emission reductions and future warming among climate experts

The Intergovernmental Panel on Climate Change (IPCC) utilizes emission scenarios to project future climate outcomes, ranging from scenarios with significant mitigation efforts to those with catastrophic warming. However, the IPCC refrains from assigning probabilities to these scenarios, leaving policymakers to interpret the likelihood of each outcome. This study investigates the beliefs of IPCC authors themselves, recognizing their expertise and influence on both scientific and public discourse. The authors' perceptions of likelihood are crucial because their opinions shape policy recommendations and public understanding of climate change. The study focuses on understanding the personal beliefs of IPCC authors about four key climate outcomes: maximum warming by 2100, the likelihood of exceeding 3°C warming, the year net-zero CO2 emissions will be reached, and the rate of carbon dioxide removal (CDR) in 2050. The researchers hypothesized a positive correlation between authors' personal estimates and their perceptions of their peers' beliefs (second-order beliefs), reflecting a potential 'false consensus effect'. They also hypothesized that researchers focusing on climate solutions (Working Group 3) would hold more optimistic views than those focused on impacts and adaptation (Working Group 2). This study's importance lies in understanding the beliefs that shape both the generation and communication of climate science, ultimately influencing policy decisions and public perception.

The paper references previous research highlighting the challenges of communicating uncertainty in IPCC reports and the benefits of providing likelihood information about future climate outcomes. It cites studies emphasizing the influence of IPCC assumptions and modeling choices on political decisions and technological pathways. The literature also touches upon the influence of peer beliefs ('second-order beliefs') on individual opinions within scientific communities and the potential for biases like the 'false consensus effect', where individuals overestimate the extent to which others share their views. Several prior studies are mentioned that surveyed IPCC authors' beliefs regarding future warming, indicating a general consensus on the likelihood of significant warming but lacking comprehensive exploration of other relevant climate outcomes and the influence of second-order beliefs.

A survey was distributed to 909 IPCC AR6 authors (from the SR1.5 report onwards). The survey solicited predictions for four future climate outcomes: maximum global warming by 2100, likelihood of exceeding 3°C warming, the year of net-zero CO2 emissions, and the rate of CDR in 2050. Participants were also asked to estimate their peers' beliefs for each outcome. 211 authors responded. Median responses were analyzed due to non-normal distribution of data. The relationship between personal beliefs (first-order beliefs) and perceived peer beliefs (second-order beliefs) was assessed using Spearman rank correlations. Differences in perceptions between Working Groups (WG1, WG2, WG3) were analyzed using Kruskal-Wallis tests. Wilcoxon signed-rank tests examined differences between first-order and second-order beliefs. Sample weights were calculated to account for potential demographic biases using inverse proportional fitting based on gender, continent of citizenship, and working group. Outliers were identified and handled through removal or recoding. Statistical analyses were performed in R.

The survey revealed significant skepticism among IPCC authors regarding the achievement of Paris Agreement temperature goals. A majority (86%) estimated maximum warming exceeding 2°C by 2100, with 58% estimating at least a 50% likelihood of exceeding 3°C. Despite this pessimism about temperature targets, most authors (median year 2055) expressed optimism about reaching net-zero CO2 emissions before 2050, although this was inconsistent with their warming predictions. Strong correlations were found between authors' personal beliefs and their perceptions of peer beliefs across all four outcomes (rs ranging from 0.55 to 0.78), indicating a pronounced false consensus effect. Limited differences were observed between working groups' estimations, except for the likelihood of exceeding 3°C warming. A slight tendency for participants to overestimate their peers' optimism regarding net-zero CO2 and CDR rates was noted.

The findings highlight a discrepancy between IPCC authors' belief in achieving net-zero CO2 and their expectation of high levels of warming, suggesting potential underestimation of climate sensitivity, misunderstanding of the relationship between net-zero and warming, or other factors. The strong correlation between personal and peer beliefs reinforces the impact of the false consensus effect within the climate science community. The limited differences between working groups suggest that biased perceptions are not primarily driven by group isolation, but rather by shared cognitive heuristics. The observed bias towards overestimating peer optimism might indicate a tendency to present a more optimistic outlook than warranted. This study provides insights into the beliefs that shape both the production and communication of climate science knowledge, underscoring the need for a broader community discussion to bridge potential gaps between individual perceptions and the overall scientific consensus.

This study demonstrates that while many IPCC authors are skeptical of meeting the Paris Agreement goals, they remain optimistic about achieving net-zero CO2 emissions. The false consensus effect is prevalent, highlighting a need for improved communication and understanding within the scientific community. Future research could explore the cognitive processes underlying these beliefs and develop strategies for bridging the gap between individual perceptions and overall scientific consensus. The inconsistent beliefs regarding warming levels and net-zero emissions warrant further investigation.

The study's reliance on self-reported data and potential response biases inherent in surveys could limit the generalizability of the findings. The survey response rate of approximately 23% could also affect representativeness. The specific wording of survey questions might have influenced responses. The analysis focused on median responses, potentially overlooking important variations within the data distribution. Despite attempts to weight for demographics, the sample may not perfectly represent the entire IPCC authorship.