| 17 Jun 2026
Poleward shift of the North Pacific storm track driven by springtime East Asian dust heating
Abstract. The North Pacific storm track shapes precipitation and temperature patterns over the Arctic and western North America, yet how its sensitivity to springtime East Asian dust remains poorly understood. Based on the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2) reanalysis data from 1980 to 2022, we find that anomalously high springtime East Asian dust loading is robustly associated with a systematic poleward shift of the North Pacific storm track on interannual timescales. The physical mechanism proceeds through a clear causal chain. In details, shortwave absorption by the trans-Pacific dust plume warms the mid-troposphere between 850 and 400 hPa, exciting an anomalous anticyclonic circulation over the North Pacific. This thermal perturbation restructures the meridional temperature gradient and shifts the zone of maximum Eady growth rate poleward, thereby relocating the preferred region for baroclinic eddy development to higher latitudes. The Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) sensitivity experiments reproduce both the spatial pattern and sign of the storm track response, confirming that dust shortwave absorption is sufficient to drive the observed displacement. These findings demonstrate that natural dust aerosols can modulate large-scale North Pacific atmospheric dynamics and suggest that springtime dust variability should be considered in regional climate assessments.
The authors investigated how the East Asian dust activity impacts the North Pacific storm track during springtime. The authors first investigate a physical connection between the dust activity and the poleward shift of the storm track using reanalysis and observational datasets. They find that shortwave absorption due to dust aerosols results in mid-tropospheric heating, affects the baroclinicity in the storm track region, and shifts the storm tracks. This physical explanation is supported using a WRF simulation, with and without dust, for the year 2018.
The manuscript is well written, and I enjoyed reading it. The physical mechanisms are clearly explained. However, I do have some remaining concerns for the paper. As the authors mention, interannual variability of dust is tightly connected to circulation, such as jets and storm tracks (dust must have been transported to the Pacific by winds). There could be a mode of circulation across East Asia and the Pacific that favors both dust import into the Pacific and a poleward shift of the storm track. Hence, from the observations, the causal direction is unclear. The authors do show model experiments which have clear causality; however, the model experiment and the observational analysis are not well connected. Therefore, I suggest a major revision is needed to clarify this causality. Below are my detailed comments and suggestions to better connect the two sections.
Major Comments
1. Figures 3 to 6 show results from regression analysis, where atmospheric variables are regressed onto EADI. I expected the units to be in `per EADI'. How is the regression being calculated?
2. Figures 8 and 9 show results from the WRF simulation. As the EADI in 2018 is around 300, the responses here are for an EADI difference of 300. On interannual time scales, the EADI variability is an order of 10. If I multiply 1/30 to Figure 8 and compare it with Figure 3, the results from Figure 8 are very small. This suggested to me that the impact of dust-induced heating on the poleward shift, while the mechanism is there, is very small, and there could be another factor that affects both dust transport and storm tracks. However, because I'm not clear how the regression was calculated, this comparison could be inaccurate. Could the authors clarify and show that 'responses per EADI' are comparable in the regression analysis and the modeled responses? In short, it would be helpful if the results are presented per EADI.
3. The North Pacific storm track location shifts on interannual timescales due to processes other than dust activity. What is the relative magnitude of the dust-related storm track variability, inferred from the WRF simulations, compared to the full variability? I can imagine doing this by rescaling the responses in Figure 8 per EADI, multiplying it by the interannual variability of EADI, and then comparing with the full interannual variability. The authors mention that the regressed results are about 15% of climatology, but as I mentioned before, it is difficult to get a grasp of this number without clearly knowing how regression has been done.
4. Discussions around Figure 4 can be improved. Firstly, the clear-sky TOA and SFC SW responses seem weak. They are statistically significant; however, their magnitudes are smaller than the noise outside (e.g., the Bering Strait). Second, the difference between clear-sky and all-sky SW responses for both TOA and SFC shows the impacts of clouds (Figs. 4 a,c,d f). There is a clear north-south dipole, indicating increased clouds to the north and decreased clouds to the south (Figs. 4d,f). This suggests that it corresponds to the poleward shift of the storm tracks. Therefore, the results in Figs. 4d,f do not help identify the thermal driver but show the storm track shift response from different variables. In general, I think the TOA and SFC SW results are not central to the physical mechanism. I would suggest removing them for clarity or having them in the supplementary materials. Panels b and e are very clear.
5. Can authors plot climatological MTG and EGR in Figure 6? It would be important to know if the dipole response is also centred around the climatological maximum, as for the storm track metrics.
Minor Comments
1. Line 13. yet how -> yet
2. Line 53-54. It would be helpful to clarify which season these studies focused on. Many are for winter.
3. Line 107. explicitly isolate -> remove?
4. Line 152. It would be helpful to mention that the storm track metrics correspond to red and yellow contours since there are many contents in the figure.
5. Line 258. 'through thermal wind balance' here reads as if it is a process. Please revise.
6. Figure 6. I was trying to match Figs. 3 and 6, but they had different latitudinal bounds. It would be nice to keep them consistent for comparison.