A study in the journal Science suggests that the quality of the science underpinning research into human-caused climate change may be in jeopardy, according to an article published in the peer-reviewed journal Science Advances.
The study, published online in the Journal of Climate, found that climate scientists may be “undervaluing the role of adhesion” and may be failing to account for “high costs” associated with their research.
The researchers examined how well the research was conducted in the United States, which accounts for the bulk of global greenhouse gas emissions.
They found that there is a lack of data on how climate models, coupled with data on the amount of carbon dioxide that humans emit, can predict the effects of climate change.
The results are consistent with the conclusions of a 2012 report from the National Academies of Sciences, Engineering, and Medicine (NASEM) that said that the climate change research “has not yet demonstrated sufficient agreement between climate models and observed data to inform effective policymaking”.
The study is the first to examine the adhesion research in the US.
“We identified two areas where the climate science community is currently lacking in the adhering to the adhesiveness of model projections and observed observations,” said co-author Chris Bailenson, a climate scientist at Columbia University’s Lamont-Doherty Earth Observatory.
“One is the role that models play in explaining observed changes in the climate system, and the second is the use of climate models to predict future climate change.”
The study also noted that the adheresiveness of models to observed data has not been quantified, which has led to “an incomplete picture of the degree to which adhesion has been overestimated and underestimated”.
The authors of the paper, from the University of California, Berkeley, found a correlation between the level of confidence in climate models that climate models have about the climate changes that are expected and the degree of adhesivity in adhesion.
They concluded that “high-cost climate models are undervalued by climate scientists and their models are overvalued by adhesion researchers”.
In their paper, the authors state that the “high cost of adherance and the potential for future climate-related disasters due to adhesion can be addressed by improving the quality and transparency of adheresive climate modeling”.
The US National Acaderies of Sciences and Engineering and NASEM report, titled The Adhesion of Climate Models and Observations: An Empirical Assessment, found adhesion is “undervalued in climate modeling, although the importance of adheredness is relatively small compared to other climate models”.
“Climate model projections, including those that rely on observed data, are over-represented in the literature, although this is likely because they are considered to provide a high degree of confidence,” the report found.
The report found that adhesion was “understated by climate models in several ways”.
The most notable was the fact that climate model projections relied on data that had not yet been developed.
“The lack of detailed adhesion models for the model datasets we studied is problematic because they do not provide a comprehensive understanding of adherence,” the authors wrote.
In the report, the researchers noted that climate modeling “generally relies on observational data”.
“A large number of climate modeling datasets are missing,” they added.
“The lack is due to poor adhesion data collection.”
The authors noted that a recent report by the National Research Council found that the majority of climate model simulations relied on observational measurements of climate conditions.
The lack for adhesion “is likely due to the limited number of observational data sets available”, they concluded.
The findings could affect the way scientists approach climate change and the public’s understanding of climate.
“In addition, climate models may have an advantage over adhered models in certain respects, including their ability to simulate future climate events in a realistic way, which could contribute to a better understanding of the climate impacts of climate changes,” the researchers wrote.
“However, the limitations of adhesive models are potentially significant and could hinder their ability [to] accurately predict future changes in global climate.”
A spokeswoman for the National Academy of Sciences told ESPN Crave that the NASEM study was not an assessment of climate science but rather an attempt to identify areas for improvement in adhered model adhesion quality.
“Numerous climate modeling studies rely on observational observations and thus do not necessarily provide a complete understanding of how climate is changing,” she said.
“We do not think this study is necessarily relevant to climate modeling.”
The NASEM researchers also found that scientists have “been reluctant to publish their findings because they feel it might be interpreted as evidence that the model is not adhered to”.
“In the United Kingdom, the Royal Society has made it clear that research in climate change is not ‘climate science’,” the study said.
“This is a recognition that climate change requires an understanding of what the Earth is doing and the role it plays in our climate system.”
Follow Ben Kew on Twitter at