LAUSR

The Letter to the Editor (1) suggests that our paper (2) missed a link between topography and ocean circulation and failed to cite the relevant literature. We want to begin by clarifying that we acknowledged the link between topography and ocean circulation and discussed recent relevant literature. In particular, in ref. 2 we noted that “A recent study showed that flattening topography in coupled models leads to a weaker ocean circulation, but the impact on storminess is consistent with our atmospheric model results (storminess asymmetry is still significant with flattened topography).” We were referring to the study by Stouffer et al. (3) who showed the response to flattened topography, in an ensemble of comprehensive coupled climate models, which involves weaker ocean overturning circulation in the Atlantic, equatorward shift of surface zonal wind stress in the Southern Hemisphere, and intensification in the Northern Hemisphere extratropics (figures 7A, 8D–F and 12 in ref. 3). One can infer from figures 7A and 8 in ref. 3 that coupled models with flattened topography exhibit hemispheric asymmetries in extratropical surface zonal wind stress, especially over the ocean. The surface zonal wind stress response to flattened topography in our atmosphere-only simulations is similar to the coupled model responses (compare figures 1–8 D–F in ref. 3). We think that the regional hemispheric differences in coupled models with flattened topography would need to be smaller, cf. figure 2D in ref. 2, to conclude “that topography is likely the dominating factor in controlling the asymmetry of westerly winds and thus storminess between the hemispheres.” Given that PNAS limits the number of references, we chose to focus on ref. 3 since it used an ensemble of comprehensive coupled climate models. We apologize for not citing ref. 4 and other studies that demonstrated an impact of flattened topography on the ocean, e.g., refs. 5 and 6. A comparison of previous results suggests that flattened topography can switch the ocean overturning circulation from the Atlantic to the Pacific basin in some models, e.g., refs. 4–6. However, in other models, it does not switch (3). The impact of topography in the Atlantic basin is sensitive to ocean model details (figure 12 in ref. 3). In all models with flattened topography, the ocean circulation still transports energy across the equator breaking hemispheric symmetry. Thus, previous coupled model results provide evidence that topography affects the ocean circulation but does not fully FLAT ALL