The Earth system is made up of interacting processes that connect the atmosphere, ocean, cryosphere, biosphere and lithosphere, spanning timescales and length scales. Humans are altering the Earth system by polluting the air, water and ground and modifying landscapes at an unprecedented rate. As a result, the Earth system is changing, with important and urgent consequences for present and future populations. Understanding exactly how the Earth system is changing is a key motivation for the broader climate and Earth system science community, many of whom, including myself, were gathered at the CMIP Community Workshop 2026 in Kyoto, Japan. The CMIP workshop provides a forum for the community to come together to evaluate recent progress in modelling efforts and also plan for the next phase of CMIP.
I was extremely fortunate to receive a travel award from the workshop organizers which helped me attend the workshop in-person, allowing me to have the opportunity to give a plenary oral presentation on my research on the role of future climate change, air pollution reductions and methane mitigation in driving hydroxyl radical (OH) and methane lifetime. I also had a fruitful time learning from and networking with other scientists.
What are Earth system models, and their importance in CMIP
Computational models of the Earth system are critical tools that scientists have developed and employed to better understand human influences on the Earth system. From the earliest versions of models in the 60s through today, models have provided unequivocal evidence on the warming impacts of human-emitted greenhouse gases. Compared to the earliest versions of models, the models of today represent a wider variety of processes, allowing for a more comprehensive understanding of Earth system changes. Crucially, these computational models allow scientists to understand Earth system changes and their drivers in the historical period as well as in future projections. One of CMIP’s key roles is facilitating the coordination of standardized model experiments which are then conducted across multiple models to provide results that eventually inform the Intergovernmental Panel for Climate Change (IPCC), a group of scientists that synthesizes the state of climate science into reports that inform policymakers at the United Nations level.
Poster session – Advancing Earth System Modelling: Key Datasets, Observations and Processes
I attended this poster session hoping to learn more about the latest research pertaining to Earth system modelling. I attempt to summarize my experience in this blog post, and I have also included some research highlights from other Early Career Researchers.
Earth system models help us to understand how human emissions are affecting the climate
A key strength of Earth system models is their ability to provide mechanistic insights into how human emissions of aerosols, which are tiny particles which could cool or warm the climate depending on their optical properties, and climate-warming greenhouse gases contribute to changes in the Earth system. Posters showed the connection between human emissions and atmospheric circulation changes, long-range pollutant transport changes, mid-20th century Arctic cooling, record-low Antarctic sea ice in 2023, and temperature and precipitation trends in the Western Mediterranean.
Another class of climate-altering human emissions is ozone-depleting substances, so-called because of their ability to deplete upper atmosphere (stratospheric) ozone. Stratospheric ozone helps to reduce incoming harmful radiation, and the human health threat posed by its depletion brought the world together to regulate these ozone-depleting substances under the Montreal Protocol. Dr. Vichawan (Print) Sakulsupich, a postdoctoral researcher at the University of Cambridge, showed how the stratospheric ozone reductions due to historical emissions of ozone-depleting substances was linked to increased cloudiness over the Southern Ocean via induced changes in atmospheric chemistry in the lower atmosphere (troposphere). By allowing scientists to draw the link between human emissions and various climate and Earth system changes, models not only are a powerful tool helping to increase scientific understanding of the Earth system, but also help to provide the evidence base that motivate policies designed to regulate human emissions in order to minimize their harmful effects.
Earth system models need to be evaluated against real-world observations
While models strive to emulate the real world as much as possible, a key measure of model performance is how well they reproduce observed phenomena. Thus, a big theme in this poster session was quantifying biases in Earth system model results and projections. From in-situ and local site measurements to satellite data, many observational datasets of a whole range of quantities exist today that can be used to validate and evaluate models. Various posters in this session evaluated models on how well they represented historical changes in various important climate-relevant quantities, such as radiative imbalance, sea surface temperatures, soil moisture and precipitation, as well as other Earth system features, such as ocean biogeochemistry, and various model biases were identified. Understanding what causes these biases can help to improve the representation of various processes in models, ultimately leading to increased confidence in model results and projections.
Better quantification and representation of uncertainties in Earth system model projections
Another important area of research in relation to model projections is improving the quantification of uncertainties in Earth system model results. Models can vary subtly from one to another, and, by analyzing multi-model projections, scientists can identify features that are robust across multiple models and also quantify the inter-model spread. Another source of uncertainty in future projections arises from the uncertainty in future socioeconomic and political changes, exemplified by the diverging nature of future scenarios developed as part of CMIP. Dr. May Chim, a postdoctoral researcher at the University of Exeter, identifies another major source of uncertainty: volcanic emissions, a powerful natural climate forcer that tends to cool the climate, which could contribute up to half of the total uncertainty in near-future projections. Notably, this uncertainty is not represented in current models due to the way that volcanic emissions are represented currently, and Dr. Chim’s work highlights the importance of improving model representation of volcanic emissions to account for their sporadic nature. Ultimately, a better quantification of uncertainties will help to inform the confidence in modelled future projections.
Earth system modelling for CMIP7
One of the main goals of the CMIP workshop is to plan for the future. CMIP is currently gearing up for its next phase (CMIP7) to provide analysis for the next IPCC synthesis report (AR7). For this next exercise, various scenarios and datasets have been prepared, and there were some posters in this session that discussed the new human population and volcanic emission datasets for CMIP7. Also, for CMIP7, new experiments guided by new science questions are being coordinated. An example of new science questions is an effort to quantify the net effects of afforestation, the act of growing trees in an effort to soak up carbon dioxide from the atmosphere, on the climate by considering other changes in the Earth system that arise from increasing trees in a multi-model setting, spearheaded by Dr. James Weber, lecturer at the University of Reading, as part of Phase 2 of the Aerosol and Chemistry Model Intercomparison Project (AerChemMIP2). Going forward, as models increase their representation of climate-responsive emissions and improve their representation of the carbon cycle, such research questions, which aim to link biogeochemical processes with atmospheric composition and climate, can increasingly be answered with Earth system models.
In conclusion, Earth system models are more sophisticated than ever before, allowing for new scientific insights on various Earth system interactions to be gleaned. With greater model complexity comes a greater need to critically evaluate model biases and uncertainties. The variety of topics explored in this poster session, which featured 40 submissions from all over the world, is a testament to the sheer breadth and depth of Earth system modelling science today as well as its global nature. As we enter the next phase of CMIP, the research community is well-poised to continue to contribute to policy-relevant science and address societal challenges. I left the poster session and the workshop feeling newly inspired to continue to pursue climate science.
