In the context of the axisymmetric theory of Emanuel (1986), we have shown that the RMW is very nearly an M surface, because of the constraints of thermal wind balance and slantwise moist neutrality. Observations confirm this prediction to a certain extent, although there is a small but systematic tendency for M to decrease upwards along the RMW of real storms. In contrast, simulated hurricanes evolve towards a state where M increases substantially upwards along the RMW.
Theoretically, the outward slope of M surfaces and the RMW should increase approximately linearly with radius (at constant height), and should be essentially independent of intensity. Observations agree quite well with this prediction, while the simulations yield mixed results.
The degree to which slantwise moist neutrality is a valid approximation for tropical cyclones is investigated in the simulations. Within the eyewall, the vortex is actually conditionally unstable below about 4.5 km, while it is stable above. In spite of this, the rate at which the maximum winds decay with height is often well predicted by theory.