The abstract claims the land-ocean asymmetry results in a “six-month delay” in the global TAE annual cycle relative to the total SR received by Earth. However, in Figure 2b and the text, the global TAE peak is said to occur around 24 July, which is about 1.5 months after the NH ISEF peak (around June) and roughly 6 months after the global mean solar radiation minimum, not the annual mean cycle. The use of “six-month delay” is ambiguous and could be misinterpreted as a 180° phase shift. Can you clarify the exact reference point for this delay?
In Section 4.1, you explicitly state that hemispheric differences in surface albedo (α) and atmospheric emissivity (ε) are ignored to “isolate the role of land-ocean distribution.” However, observational data show that the SH has a higher albedo due to greater cloud cover and Antarctic ice, and that greenhouse gas concentrations and thus emissivity are not identical between hemispheres. Does this simplification not potentially overstate the role of mixed-layer depth by neglecting a known negative feedback (higher SH albedo) that would dampen the SH energy cycle and perhaps enhance the NH dominance, thereby confounding your attribution?
In Equation (3), the surface energy balance for Ts,j includes a term −Fj, implying Fj is an upward flux from the surface. In Equation (4) for the atmosphere, it appears as +Fj. However, in the standard setting (Table 2), γ=20 W m−2K−1, which is an order of magnitude larger than typical bulk aerodynamic coefficients (~10-20 W m⁻² K⁻¹ for sensible heat alone, but here it seems to represent total turbulent flux). For a typical surface-air temperature difference of ~5 K, this yields Fj=100 W m−2, which is an unrealistically large flux for the global mean. Does this high value artificially amplify the coupling and affect the phase sensitivity?
You claim that the asynchronous response requires H0>12 m. However, Figure 5a shows that for high albedo (α=0.5), the ratio TAE_SH,max/TAE_NH,max remains near 1.0 even for H0=35 m, indicating no asymmetric response. This contradicts your statement that the response appears “regardless of the value of α.” Can you explain this discrepancy, and does this not suggest that albedo is a more critical factor than claimed?
In the NO_LOD experiment, you replaced all land with a 10-m-deep ocean layer. However, a 10-m depth globally increases the effective heat capacity of the surface dramatically compared to land (which has almost no thermal inertia). This not only removes the hemispheric asymmetry but also adds significant thermal inertia to the NH, which would change its phase. How do you separate the effect of removing the asymmetry from the effect of increasing the absolute thermal inertia of the NH surface? A globally uniform 0.1-m “swamp” ocean would have been a more appropriate control.
In Figure 2b (ERA5), the global TAE peak is about 2.65×1023 J, while in Figure 3b (simple model), the global TAE peak is around 10.55×1023 J, a factor of 4 difference. Since the model is supposed to represent a hemispheric mean, this implies a major discrepancy in the atmospheric mass or temperature scaling. Can you explain this order-of-magnitude difference and confirm that the simple model’s Ta values are physically realistic?
In Equation (1), you define Cp=1004.7 kJ kg−1. This is incorrect—the specific heat of air at constant pressure is 1004.7 J kg−1K−1 (not kJ). This is likely a typo, but it propagates into the calculation of TAE values. Does correcting this factor of 1000 affect any of the magnitude comparisons in the paper?