Abstract. While carbon dioxide is the main cause for global warming, modeling short-lived climate forcers (SLCFs) such as methane, ozone, and particles in the Arctic allows us to simulate near-term climate and health impacts for a sensitive, pristine region that is warming at 3 times the global rate. Atmospheric modeling is critical for understanding the long-range transport of pollutants to the Arctic, as well as the abundance and distribution of SLCFs throughout the Arctic atmosphere. In this study, we evaluate 18 state-of-the-art atmospheric and Earth system models by assessing their representation of Arctic and Northern Hemisphere atmospheric SLCF distributions, considering a wide range of chemical species and multiple observational datasets. The results show a large range in model performance, with no one particular model performing well for all regions and species. The multi-model mean (mmm) was able to represent the general features of SLCFs in the Arctic and had the best overall performance. For the SLCFs with the greatest radiative impact ($CH_4$, $O_3$, BC, and $SO_4^{2-}$), the mmm was within ±25% of the measurements across the Northern Hemisphere. Therefore, we recommend a multi-model ensemble be used for simulating climate and health impacts of SLCFs.
Introduction
The Arctic is the "canary in the coal mine" for climate change, but it remains notoriously difficult to model. Beyond carbon dioxide, SLCFs—including methane ($CH_4$), tropospheric ozone ($O_3$), black carbon (BC/soot), and various aerosols—exert a powerful but transient influence on both regional warming and human health. The challenge lies in the Arctic's geography: it is far from major pollution sources, meaning models must accurately simulate long-range transport, complex vertical mixing in a stable atmosphere, and removal processes like "wet deposition" (rain/snow). This paper provides a comprehensive "stress test" for 18 of the world’s leading models to identify which approaches are currently most effective.
Methodology
As part of the Arctic Monitoring and Assessment Programme (AMAP), researchers compared 18 different models, including Chemical Transport Models (CTMs), Chemistry-Climate Models (CCMs), and full Earth System Models (ESMs). To isolate model performance from emission uncertainties, all models utilized the same anthropogenic emissions dataset (ECLIPSE v6b). Simulations were analyzed across two 2-year windows: 2008–2009 and 2014–2015.
Overview of the 18 atmospheric and Earth system models evaluated in the study.
The study utilized a diverse array of observational data to validate model performance, including surface networks (WDCGG, EMEP, IMPROVE, NAPS), satellite instruments (TES, ACE-FTS, MOPITT), and various aircraft and ship-based field campaigns. These datasets provided critical information on the annual means, seasonal cycles, and 3D atmospheric distributions of pollutants.
Map of Arctic observation locations for ozone, black carbon, and sulfate used for model evaluation.
The "Wisdom of the Crowd": The Multi-Model Mean (MMM) consistently outperformed any individual model. For the most critical climate forcers, the MMM was within ±25% of actual measurements across the Northern Hemisphere.
Latitudinal Bias Flip: A notable trend emerged where model biases often shifted from positive to negative as latitude increased. This suggests that while models may accurately represent source regions, they lose precision when simulating the transport of pollutants toward the Pole.
Wildfire and Vertical Mixing: Models generally overpredicted ozone and aerosols in Alaska during the summer, pointing to systemic issues in handling wildfire plume injection heights. Furthermore, almost all models struggled to simulate the rapid drop-off of pollutants above the Arctic tropopause, highlighting vertical mixing as a significant weak link.
Species-Specific Performance: Methane was simulated most accurately (+1.3% bias in the Arctic), while Organic Aerosol (OA) showed the highest uncertainty and largest biases. Interestingly, while atmospheric concentrations of BC were often underestimated, deposition fluxes in ice cores were overestimated by a factor of three, suggesting models may "rain out" pollutants too quickly.
Conclusion
Vertical mixing, long-range transport, deposition, and wildfire parameterization remain highly uncertain processes in current atmospheric models. Until these individual processes are better represented, researchers and policy-makers should rely on multi-model ensembles rather than single-model results for Arctic climate and health assessments. Future work must focus on evaluating the 3D distribution of SLCFs to improve these uncertain model components.
Summary updated
Abstract
<strong class="journal-contentHeaderColor">Abstract.</strong> While carbon dioxide is the main cause for global warming, modeling short-lived climate forcers (SLCFs) such as methane, ozone, and particles in the Arctic allows us to simulate near-term climate and health impacts for a sensitive, pristine region that is warming at 3 times the global rate. Atmospheric modeling is critical for understanding the long-range transport of pollutants to the Arctic, as well as the abundance and distribution of SLCFs throughout the Arctic atmosphere. Modeling is also used as a tool to determine SLCF impacts on climate and health in the present and in future emissions scenarios. In this study, we evaluate 18 state-of-the-art atmospheric and Earth system models by assessing their representation of Arctic and Northern Hemisphere atmospheric SLCF distributions, considering a wide range of different chemical species (methane, tropospheric ozone and its precursors, black carbon, sulfate, organic aerosol, and particulate matter) and multiple observational datasets. Model simulations over 4 years (2008â2009 and 2014â2015) conducted for the 2022 Arctic Monitoring and Assessment Programme (AMAP) SLCF assessment report are thoroughly evaluated against satellite, ground, ship, and aircraft-based observations. The annual means, seasonal cycles, and 3-D distributions of SLCFs were evaluated using several metrics, such as absolute and percent model biases and correlation coefficients. The results show a large range in model performance, with no one particular model or model type performing well for all regions and all SLCF species. The multi-model mean (mmm) was able to represent the general features of SLCFs in the Arctic and had the best overall performance. For the SLCFs with the greatest radiative impact (CH<span class="inline-formula"><sub>4</sub></span>, O<span class="inline-formula"><sub>3</sub></span>, BC, and SO<span class="inline-formula"><math xmlns="http://www.w3.org/1998/Math/MathML" id="M3" display="inline" overflow="scroll" dspmath="mathml"><mrow><msubsup><mi/><mn mathvariant="normal">4</mn><mrow><mn mathvariant="normal">2</mn><mo>-</mo></mrow></msubsup></mrow></math><span><svg:svg xmlns:svg="http://www.w3.org/2000/svg" width="13pt" height="17pt" class="svg-formula" dspmath="mathimg" md5hash="ae2ede098a116a7d88438e298a87d51c"><svg:image xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="acp-22-5775-2022-ie00001.svg" width="13pt" height="17pt" src="acp-22-5775-2022-ie00001.png"/></svg:svg></span></span>), the mmm was within <span class="inline-formula">±</span>25â% of the measurements across the Northern Hemisphere. Therefore, we recommend a multi-model ensemble be used for simulating climate and health impacts of SLCFs. Of the SLCFs in our study, model biases were smallest for CH<span class="inline-formula"><sub>4</sub></span> and greatest for OA. For most SLCFs, model biases skewed from positive to negative with increasing latitude. Our analysis suggests that vertical mixing, long-range transport, deposition, and wildfires remain highly uncertain processes. These processes need better representation within atmospheric models to improve their simulation of SLCFs in the Arctic environment. As model development proceeds in these areas, we highly recommend that the vertical and 3-D distribution of SLCFs be evaluated, as that information is critical to improving the uncertain processes in models.