Recent  

The climate system is strongly regulated by interactions between the ocean and atmosphere. However, many uncertainties remain in the understanding of these interactions. Our analysis uses a relatively novel approach to quantify causal links between the ocean surface and lower atmosphere based on satellite observations. We find that both the ocean and atmosphere influence each other, but with varying intensity depending on the region, demonstrating the power of causal methods.

There are lots of possibilities for future emissions, with different impacts in different places. Complex models can study these impacts, but take a long time to run even on powerful computers. Simple methods can be used to reduce this time by estimating the complex model output, but these aren’t perfect. This study looks at the accuracy of one of these techniques, showing that there are limitations in its use, especially for low emissions future scenarios.

Permafrost thaws and releases carbon (C) as the Arctic warms. Most Earth System Models (ESM) have poor estimates of C stored now and so their future C losses are much lower than using the PInc-PanTher permafrost C model with climate inputs from 6 ESM. Bias-corrected soil temperatures and vegetative productivity plus geoengineering lowering global temperatures from a “business as usual level to a moderate emissions level keep C in the soil worth about $20 trillion in climate damages by 2100.

A new statistical method combines climate projections and observations to reduce uncertainty in the projections of the sea surface temperature (SST) over the North Atlantic ocean. The method is able to reduce the climate model uncertainty in the SST over the historical period, but also in the future, with a decrease of 65 % in the short term, up to 50 % in the long term.

Emergent Constraints (ECs) reduce the spread of projections between climate models. ECs estimate changes to climate features impacting adaptation policy, and with this high profile, the method is under scrutiny. Asking “What is an EC?”, we suggest they are often the discovery of parameters that characterise hidden large-scale equations that climate models solve implicitly. We present this conceptually via two examples. Our analysis implies possible new paths to link ECs and physical processes.

Multi-instrument records with varying signal-to-noise ratios are becoming increasingly common as legacy sensors are upgraded and data sets are modernized. Induced changes in higher-order statistics such as the autocorrelation and variance are not always well-captured by cross-calibration schemes. Here we investigate using synthetic examples how strong resulting biases can be and how they can be avoided in order to make reliable statements about changes in the resilience of a system.

The Earth has nearly the same mean albedo in both hemispheres, a feature not well replicated by climate models. Global warming causes changes in surface and cloud properties that affect albedo, in turn feeding back into the warming. We show that models predict more darkening due to ice loss in the northern than the southern hemisphere in response to increasing CO₂ concentrations. This is to varying degrees counteracted by changes in cloud cover, with implications for cloud feedback on climate.

Phytoplankton absorbs the solar radiation entering the ocean surface, and contributes to keep the associated energy in surface waters. This natural effect is not commonly represented in the oceanic part of climate models, or often suffers simplifications. We show that an incomplete representation of this biophysical interaction affects the way climate models capture ocean warming, what in turn uncertains the forecast of oceanic emissions of an important greenhouse gas called the nitrous oxide.

The second law of thermodynamics has long intrigued scientists, but which role does it play in the Earth system? This review shows that its main role is that it shapes the conversion of sunlight into work. This work can then maintain the dynamics of the physical climate system, the biosphere, as well as human societies. The relevance of it is that apparently many processes work at their limits, so that they become predictable by simple means.

We test the reversibility the of the Earth system using idealized model simulations of overshoot pathways, where carbon dioxide is removed from the atmosphere to compensate for too large past and near term emissions. On a centennial time scale, we find a high degree of reversibility if the overshoot size remains small, and we do not find tipping points even for intense overshoots. We caution, however, that current ESMs are most likely not able to skillfully model tipping points in ecosystems.

The Vb-cyclones simulated with a coupled regional climate model with two different driving data sets are compared against each other in historical period, thereafter the future climate predictions were analyzed. The Vb-cyclones in two simulations agree well in terms of their occurrence, intensity and track in two simulations, though there are discrepancies in seasonal cycles and their process linking Mediterranean Sea in historical period. So significant changes were observed in the future.

This study evaluates five bias correction methods (four univariate and one multivariate) for correcting multivariate heat-stress indices. We show that the multivariate method can benefit the indirect correction that first adjusts individual components before indices calculation, and its advantage is more evident for index relying equally on multiple drivers. Meanwhile, the direct correction of heat-stress indices by univariate Quantile Delta Mapping approach also has a comparable performance.

Understanding whether the El Niño–Southern Oscillation (ENSO) is likely to change in the future is important due to its widespread impacts. By using large ensembles, we can robustly isolate the time evolving response of ENSO variability in 14 climate models. We find that ENSO variability evolves in a non-linear fashion in many models and that there are large differences between models. These non-linear changes imply that ENSO impacts may vary dramatically throughout the 21st century.

Ocean surface temperature is strongly affected by vertical mixing processes induced by waves. The correct estimation of a thermal state of the ocean upper layer is crucial for climate modeling and weather forecasting. The accuracy of predictions of exchange of heat between oceanic waters and the atmosphere depends on a reliable input from wave models admitting mixing processes. The study discusses the importance of higher-order approximation of vertical mixing due to high and steep waves.

Climate change is caused by the accumulated heat in the Earth system, with the land storing the second largest amount of this extra heat. Here, new estimates of continental heat storage are obtained, including changes in inland water heat storage and permafrost heat storage in addition to changes in ground heat storage. We also argue that heat gains in all three components should be monitored independently of their magnitude due to heat-dependent processes affecting society and ecosystems.

We examine how geoengineering using aerosols in the atmosphere might impact urban climate in the greater Beijing region containing over 50 million people. Climate models have too coarse resolutions to resolve regional variations well, so we compare two work arounds for this – an expensive physical model, and a cheaper statistical method. The statistical method generally gives a reasonable representation of climate, has limited resolution, and a different seasonality from the physical model.

One limit of planetary habitability is defined by the threshold of global glaciation: If Earth cools, growing ice cover makes it brighter, leading to further cooling since more sunlight is reflected, eventually leading to global ice cover ("Snowball Earth"). We study how much carbon dioxide is needed to prevent global glaciation in Earth's history given the slow increase in the Sun's brightness. We find an unexpected change in the characteristics of climate states close to the Snowball limit.

We carry out an assessment of an ensemble of general climate models (CMIP6), based on the ability of the models to represent the key physical process that are important for representing the European climate. Filtering the models with the assessment leads to more of the models with less global warming being removed, and this shifts the lower part of the projected temperature range towards greater warming. This is in contrast to the affect of weighting the ensemble using global temperature trends.