This review explains how weather patterns, guided by the polar jet stream, influence the Baltic Sea’s climate and ecosystem. It covers the North Atlantic Oscillation, blocking events and other processes and discusses how they affect temperature, rainfall, and storms from days to decades. These shifts then impact oxygen levels, productivity, and acidification in the Baltic Sea. Physical links are fairly well known, but biogeochemical pathways remain uncertain.

Global mean surface temperature anomalies in 2023 and 2024 were the highest on the modern record. We quantify contributions to these anomalies from human activity, including the steep reduction in sulfur emissions from ships that began in 2020, and numerous natural factors. We attribute about 92% of the observed anomalies to a combination of various human-induced and natural factors, including important contributions from the reduction in marine sulfur emissions and the Indian Ocean Dipole.

There is a dangerous amount of uncertainty in our predictions of climate change in polar regions because some of feedbacks that might lead to changes that are too rapid for us to adapt to, or that cannot be reversed. We have run a set of simulations with a state-of-the-art Earth System Model that helps improve our understanding of how climate in these regions might change. Some of the aspects we investigate are reversible but many are not, especially those affecting ice sheets and sea level.

Multi-model ensembles are a key approach in climate model analysis, but their use involves many complex considerations. In this work, we review relevant literature and synthesize existing studies to contribute to the development of guidelines for designing and conducting ensemble analyses. This is complemented by a collection of useful resources and a discussion of emerging trends, supported by statistics tracing the number of publications.

In this work, we apply a novel quantifier, the spatial permutation entropy, to sea surface temperatures obtained from two commonly used products: ERA5 and NOAA OI v2 (NOAA Optimal Interpolation version 2). We report small scale differences between these products, as well as persistent trends at the large scale, which could be a consequence of global warming. We also report sudden changes that were not uncovered before, which correlate with different changes in the methodology or data sources of the products.

This study revisits how critical thresholds for tipping of the Atlantic Meridional Overturning Circulation (AMOC) are derived from comprehensive climate models. It is shown that the method for keeping the global salinity constant – by compensating salinity either at the surface or throughout the ocean volume – matters for AMOC tipping: with surface compensation, the tipping point shifts to larger freshwater forcing values, mainly due to (artificial) salt input in the North Atlantic and Arctic.

The Alps host glaciers, distinct ecosystems, socio-economic interests and water resources that are being impacted by climate change. In this study, we aim at understanding how warming occurs in the Alps in projected scenarios and what physical processes drive it. We find under these scenarios that elevations around the snowline will warm faster than elsewhere, because snow retreats to higher elevations. Indeed, snow slows down warming due to its high albedo and the energy consumed to melt it.

We studied how plants respond to long droughts using model simulations and satellite data. The model reproduces drought impacts fairly well but tends to show plants recovering too quickly. Improving how the model represents plant stress and recovery will help predict how ecosystems respond to more frequent and severe droughts in the future.

Southern Ocean winds drive global climate and have strengthened since 1980 due to Antarctic ozone depletion. We assessed which climate reconstructions best capture these changes using sea level pressure observations. We then used an Earth system model to attribute these changes between ozone and greenhouse gas emissions. Ozone depletion dominated past wind acceleration, but greenhouse gases will drive future changes after 2050.

This study shows how machine learning combined with global Earth observations can improve simulations of the land carbon cycle. Optimizing key model parameters enhances the accuracy of historical carbon fluxes, while machine-learning tools help assess the robustness of these results in the presence of compensating parameter effects. The findings demonstrate that parameter optimization strongly influences simulated carbon fluxes, highlighting its importance for improving climate models.

We investigate how resilient Earth’s climate system is to disturbances by analysing glacial-interglacial cycles using a simple climate model. By simulating small and large shocks to ice volume, we show that Earth’s climate more easily returns to its natural path during warm periods. These results improve our understanding of how Earth resilience varies over time in systems driven by external forces.

Atmospheric rivers (ARs) move vast amounts of water through the atmosphere and often cause weather extremes, yet they are usually studied as regional events. Using 84 years of mapped AR trajectories, we reveal the global "roadmap" of ARs, a transport network of high-activity hubs, sparse atmospheric highways & hierarchical basins. Our approach shows how water vapor is systematically channelled through an atmospheric transport network, offering new ways to study changes in the global water cycle.

The paper assesses the impact of horizontal resolution of the EC-Earth climate model on its ability to predict El Nino Southern Oscillation (ENSO). The high-resolution simulations show better forecast skill linked to improved simulation of ENSO-related variability and ENSO teleconnections with the equatorial Atlantic. However, the remaining poor skill in the western Pacific highlights the importance of better understanding ENSO simulation errors and mean state biases to improve forecasts.

Burned area has declined globally since the late 1990s, especially in tropical savannas, yet many climate models miss this trend because they poorly represent human fire use and suppression. We tested whether adding a simple link to human development improves a fire model. This reduced large regional errors and better captured downward trends in several areas, though it increased global underestimation and reduced variability. Human influence is key, but simple assumptions have trade-offs.

This model study sets out to investigate how the collapse of the Greenland ice sheet to a virtually ice-free state due to increasing temperatures depends on the rate of this increase. We find oscillations in ice volume on millennial timescales that impact the time it takes before the ice sheet collapses, concluding that there is a mode of deterministic chaos internal to the ice sheet dynamics.

We developed a deep learning system to predict future ocean phytoplankton, the base of the marine food web. Using long-term records from climate model simulations and past ocean data, it provides skillful chlorophyll forecasts across global coastal regions, comparable to those from dynamic climate model forecasts. The predicted chlorophyll also explains historical changes in fish catch, offering a new tool to help communities prepare for climate-driven marine ecosystem changes.

Precipitation on land is split into different pathways, contributing to runoff (blue water) or to plant water use (green water). Climate change alters this balance and shapes how future precipitation is divided. We use Earth system models to study these changes and their controls. We find that stronger extreme five-day precipitation is the main control and routes more future precipitation into blue water, even where average precipitation decreases, with implications for freshwater availability.

Climate sensitivity quantifies how much Earth warms for a given radiative forcing, and is often described by the Transient Climate Response (TCR) value. Although the Intergovernmental Panel on Climate Change (IPCC) estimates the most likely TCR to be 1.8 K, some models predict values >2.4 K. Record warmth in 2023–2024 raises questions as to whether the TCR may be larger than previously suggested. Using up-to-date data, we estimate the TCR to be ~1.8 K. We also show that future warming falls within the low–mid model range, making the 2°C Paris target still feasible.

This study analyses simulations of regional climate over the Iberian Peninsula, with and without an explicit simulation of irrigation. It shows that the model matches observations much better with irrigation, particularly river discharge and evapotranspiration. The presence of simulated irrigation also makes the air cooler over irrigated areas and more humid over the whole Peninsula, leading to increases in rainfall, mostly located in the mountains that surround the highly irrigated Ebro Valley.

Observations show an increasing imbalance between how much energy the Earth absorbs from the Sun and emits back to space, leading to climate change. We evaluate how well climate models simulate both the magnitude and trend of the imbalance. We find that models capture the magnitude but underestimate the trend, which is not related to how models handle volcanic aerosols when switching to future scenarios. The models that best simulate the magnitude are the ones with moderate climate sensitivity.

We run two Earth System Models models in CO₂ emissions-driven mode to 2300 for three climate scenarios. For climate overshoot scenarios, there is a large residual warming in the 22nd century in one model, despite negative CO₂ emissions, related to Southern Ocean heat release. In both models, while global mean surface temperature is largely reversible, other global and regional climate models exhibit hysteresis and irreversibility.

The northern high latitudes mitigate climate change by land ecosystems absorbing more carbon from the atmosphere than they release. This carbon sink is sensitive to long-term CO₂ emissions. Here, we use a probabilistic framework to quantify the timing and magnitude of a transition to a carbon source in different future emission scenarios, including overshoot. The fate of this region is dependent on the balance between carbon loss from permafrost and gain from increased vegetation productivity.

Europe is warming faster than the global average, raising concerns about future wildfire risks. Using regional climate models, we find that extreme fire weather is projected to become more severe, more frequent, and more widespread across the continent, especially if global warming reaches 3 °C. The projected increase is mainly linked to a drier atmosphere. Our findings highlight the need to limit warming and strengthen adaptation practices to protect vulnerable ecosystems and communities.