Figure \ref{fig:WetProfiles} shows profiles during Cyclone Carlos, and figure \ref{fig:DryProfiles} during the dry season. During cyclone Carlos very depleted \(\delta\)^2H isotope signatures were observed near the surface and much higher water vapour high up in the atmosphere. There is no evidence of a planetary boundary layer. For the dry season profiles shown there is a clear bolundary layer. In this case, the rayeligh models are unlikely to be very accurate as the surface and the free troposphere are disconnected - a mixing model probably provides a better representation of the isotope profile (assumming the dry end member associated with the free troposphere can be defined). Complicating matters further is that the free troposphere RH shows a number of different layers indicating a single column model as used here is unlikely to accurately represent the full profile. \citet{Dyroff_2015} showed very complex vertical profiles in the subtropical Atlantic that could not be explained by a single process (ie rayleigh condensation processes or mixing). Instead within a single column the surface was close to following an open rayleigh model, whilst in the free troposphere mixing between 3 different air masses was required to explain the variability.