The admission is contained in a new paper published in Nature Geoscience, which says natural factors and unforeseen events were responsible for models overestimating the temperature rise in the troposphere.

Natural variability included El Nino and La Nina weather patterns and long cycle movements in the Pacific and Atlantic Oceans.

Unforeseen factors that contributed to cooling included volcanic eruptions, a weaker sun in the last solar cycle and a rise in pollution from coal-fired power plants in China.

The paper, “Causes of differences in model and satellite tropospheric warming rates”, is the latest shot in an ongoing scientific row over the pause or slowdown in the global temperature rise over the past two decades despite a big increase in carbon dioxide levels in the atmosphere.

Authors on the paper included Benjamin Santer from the Lawrence Livermore National Laboratory in the US, Michael Mann from Penn State University and Matthew England from the University of NSW.

Sceptics have claimed the paper as evidence to support the “pause”.

But authors said the paper ruled out claims the atmosphere was less sensitive to carbon dioxide or that future warming was not a concern.

“None of our findings call into question the reality of long-term warming of Earth’s troposphere and surface, or cast doubt on prevailing estimates of the amount of warming we can expect from future increases in GHG concentrations,” the authors said.

Researchers found that internal variability could explain differences between modelled and observed tropospheric temperature trends in the last two decades of the 20th century.

But it could not explain the divergence for the past two decades, the time of the “pause”.

“We conclude that model overestimation of tropospheric warming in the early twenty-first century is partly due to systematic deficiencies in some of the post-2000 external forcings used in the model simulations,” it said.

Unlike many high-profile papers, only the abstract was made publicly available and there was no announcement of its release.

However, in a question and answer paper published by Nature, the authors said one of the lessons learned was that “forcing matters”.

“If we systematically misrepresent these external influences in model simulations, we’ll see differences between modelled and observed warming rates,” they said.

“We need to do a better job understanding how these external influences actually changed in the real world, and we need to put our best estimates of these forcing factors into model simulations.”

Another lesson was that “natural internal variability matters”, particularly when comparing modelled and observed temperature changes with different sequences of internal variability, and over short periods.