Hurricane's maximum potential intensity and surface heat fluxes

Emanuel's concept of Maximum Potential Intensity (E-PI) relates the maximum velocity $V_{\rm max}$ of tropical storms, assumed to be in gradient wind balance, to environmental parameters. Several studies suggested that the unbalanced flow is responsible for E-PI sometimes significantly underpredicting $V_{\rm max}$. Additionally, two major modifications generated a considerable range of E-PI predictions: the dissipative heating and the power expended to lift water were respectively suggested to increase and reduce E-PI $V_{\rm max}$ by about 20%. Here we re-derive the E-PI concept separating its dynamic and thermodynamic assumptions and lifting the gradient wind balance limitation. Our analysis reveals that E-PI formulations for a balanced and a radially unbalanced flow are similar, while the systematic underestimate of $V_{\rm max}$ reflects instead an incompatibility between several E-PI assumptions. We discuss how these assumptions can be modified. We further show that irrespective of whether dissipative heating occurs or not, E-PI uniquely relates $V_{\rm max}$ to the latent heat flux (not to the total oceanic heat flux as originally proposed). We clarify that, in contrast to previous suggestions, lifting water has little impact on E-PI. We demonstrate that in E-PI the negative work of the pressure gradient in the upper atmosphere consumes all the kinetic energy generated in the boundary layer. This key dynamic constraint is independent of other E-PI assumptions and thus can apply to diverse circulation patterns. Finally, we show that the E-PI maximum kinetic energy per unit volume equals the local partial pressure of water vapor and discuss the implications of this finding for predicting $V_{\rm max}$.