Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Climate and Earth system modelling is the use of mathematical representations of key components and properties of the atmosphere, ocean and biosphere to construct computer models. These models – which can range significantly in their complexity, depending on their intended application – are used to simulate important aspects of the Earth system and indicate how they may change in the future.
Projections of Arctic warming have large uncertainties. Here the authors consider ocean heat transport and its contribution to Arctic warming; high-resolution model results show increased Bering Strait transport compared with lower-resolution results, with implications for projected warming rates.
The impact of forest loss on land surface temperature in the tropics is five times greater than the response to forest gain, according to satellite observations of temperature and land cover.
The summary of Common Era temperature reconstructions in the 2021 Sixth Assessment Report of the Intergovernmental Panel on Climate Change insufficiently characterizes reconstruction uncertainties associated with estimating global mean temperatures.
Hueholt et al. find that considering how the rate of temperature change contributes to ecosystem risk helps inform future hypothetical design of climate intervention scenarios
Earth system model projections of vegetation–climate feedback frequently depend on inaccurate values of evaporation sensitivity to vegetation changes, potentially resulting in misleading conclusions. A promising avenue involves improving the transpiration partitioning parameterizations and incorporating groundwater connections to refine the modelled sensitivity.
Methane concentrations are rising faster than ever in the atmosphere. Now, a compilation of observations points towards increased methane emissions from Arctic wetlands as being partly responsible.
Large language models can summarize, aggregate, and convey localized climate-related data to people in a cost-effective and expeditious manner. This Comment introduces a simple, proof-of-concept prototype and argues that the approach holds the potential to truly democratize climate information.
Climate models have evolved from research tools to underpin decision-making across the globe. To provide optimal value for society in the future, the models need to be made operational.