Search

Flash droughts that are accompanied by extreme heat drive more severe and prolonged impacts on global ecosystems, according to analysis of global reanalysis data and satellite observations.

This study investigates uncertainties in impact assessments when using climate projections. The uncertainties in health-related metrics combining temperature and humidity are much smaller than if the uncertainties in the two variables were independent. The finding reveals the potential for joint assessment of projection uncertainties in other variables used in impact studies.

Owing to the turbulent nature of the ocean, mesoscale eddies are omnipresent. An analysis of atmospheric conditions associated with several hundred thousand eddies in the Southern Ocean suggests that the transitory sea surface temperature fronts associated with these eddies alter near-surface winds, clouds and rainfall.

There are large uncertainties associated with the projection of climate extremes. This study shows that the uncertainties are mainly due to internal climate variability. However, model projections are consistent when averaged across regions, allowing robust projection of future extremes.

Decisions about how soon, how quickly and by how much carbon dioxide emissions are reduced will determine whether the climate target of limiting warming to 2 °C can be met. Research reveals that it will probably only be possible if ambitious reductions are implemented within the next two decades and emissions eventually fall to zero.

The contribution of human-induced climate change to global heavy precipitation and hot extreme events is quantified. The results show that of the moderate extremes, 18% of precipitation and 75% of high-temperature events are attributable to warming.

Changes in extreme heat are often calculated as anomalies above a reference climatology. A different definition—week-day heatwaves surpassing the current record by large margins—shows that their occurrence probabilities depend on warming rate, not level, and are higher than during recent decades.

Since the Paris Agreement, the impacts of 1.5 and 2 °C global warming have been emphasized, but the rate of warming also has regional effects. A new framework of model experiments is needed to increase our understanding of climate stabilization and its impacts.

If CO₂ emissions after 2015 do not exceed 200 GtC, climate warming after 2015 will fall below 0.6 °C in 66% of CMIP5 models, according to an analysis based on combining a simple climate–carbon-cycle model with estimated ranges for key climate system properties.

Including passive CO₂ uptake as an anthropogenic removal in greenhouse gas accounting systems could undermine the Paris Agreement; measures to address this include acknowledging the need for Geological Net Zero and disaggregated accounting for carbon sinks.

Uncertainties and subjective choices affecting remaining carbon budgets should be fully considered when applying them to international and national climate policies.

The politically defined threshold of dangerous climate change is an increase of 2 degrees Celsius in the mean global temperature. Simulations here show that when carbon dioxide and a full suite of positive and negative radiative forcings are considered, total emissions from 2000 to 2050 of about 1,400 gigatonnes of carbon dioxide yield a 50% probability of exceeding this threshold by the end of the twenty-first century. 'Business as usual' emissions will probably meet or exceed this 50% probability.

Climate model ensemble boosting can yield physically coherent storylines for record-shattering climate extremes such as the 2021 Pacific Northwest heatwave. Combining information from storyline approaches with process understanding can inform planning for future extremes of unprecedented intensity.

In order to limit climate warming, CO₂ emissions must remain below fixed quota. An evaluation of past emissions suggests that at 2014 emissions rates, the total quota will probably be exhausted within the next 30 years.

Climate change detection is confounded by internal variability, but recent initial-condition large ensembles (LEs) have begun addressing this issue. This Perspective discusses the value of multi-model LEs, the challenges of providing them and their role in future climate change research.

It has been assumed that spatial patterns of warming are the same under transient and equilibrium scenarios. Analysis of a multi-model ensemble shows that this is not the case, with greater land warming for a transient state, increasing risks that need to be considered in adaptation planning.

It has been predicted, by theory and models, that heavy precipitation will increase with climate change and this is now being seen in observations. Emergence of signals such as this will enable testing of predictions, which should increase confidence in them.

Considerable confusion exists as to the most likely value of climate sensitivity; by proposing a consistent framework for analysing and synthesizing research into the palaeoclimate of the past 65 million years, a value of 2.2–4.8 °C warming in response to atmospheric CO₂ doubling is obtained, in agreement with IPCC estimates.