John MolinariState University of New York, Albany, New York USA



The dramatic and unexpected intensification of Hurricane Opal in the Gulf of Mexico, followed by an equally unexpected filling before landfall, have focused interest on the mechanisms of hurricane-trough interactions. A great deal of published work exists on this topic, but it still has not proven possible to accurately predict hurricane intensity change during trough interactions. This suggests that many unanswered questions remain. In this brief writeup I will make a list of what I think are unresolved questions. It is my hope that during the poster sessions some of these questions will be addressed. I am deliberately omitting almost all references, because so many could be provided for each of the questions below. For the purposes of this writeup, "trough" refers to any sort of positive potential vorticity (PV) anomaly in the upper troposphere.


2.1 What is meant by trough interaction?

When a synoptic scale trough is, for instance, 1500 km northwest of a hurricane, it almost always will influence hurricane motion, as has been shown by PV inversion. Is the trough then interacting with the hurricane? A theoretician might say yes, a synoptician no. For our purposes, the key question is whether the trough interaction is having an impact on tropical cyclone intensity. It is not trivial to determine this! What is needed is an objective way to define a trough interaction. Eddy momentum flux convergence has been used with some success, as have PV fluxes and nonlinear balanced vertical motions near the core. But no formal way yet exists of defining a trough interaction.

2.2 How does a synoptic scale trough interact with a mesoscale hurricane?

Upward motion ahead of a synoptic scale trough is on a scale much larger than that of a hurricane. If large-scale upward motion occurs over a hurricane, will it not induce heating over a large scale as well? If so, it will produce weakening, not strengthening of the storm. This has been established from theory, modeling, and, to a lesser extent, observation.

Our view of this is that synoptic scale troughs do NOT interact favorably with tropical cyclones. Rather, some process must occur to reduce the scale of the upper trough before it approaches the tropical cyclone. We have argued this using PV principles. How does the scale reduction occur? One means is by synoptic-scale wave breaking induced in part by the hurricane outflow anticyclone. In the absence of that, we have seen a rather complex mix of vertical shear and diabatic heating influences that together act to pinch off a small scale "piece" of the upper PV maximum and bring it over the storm. This scale matching of upper PV and tropical cyclone is essential in our view.

2.3 What is the role of vertical shear?

We all accept the idea that vertical shear plays a major role in hurricane development and intensification. This has been shown quite dramatically, for instance, in the recent Atlantic hurricane season. African waves have been plentiful, but large vertical shear over the Caribbean, apparently related to the presence of the strong El Ni o, has virtually eliminated tropical cyclogenesis to an unprecedented degree, as Chris Landsea recently noted.

However, this does not mean that vertical wind shear is the only important factor. When the intensification of an individual storm is being considered, vertical wind shear is intimately tied to the rest of the dynamics. In particular, during a trough interaction, the vertical wind shear cannot be separated from the trough interaction; rather, they evolve in a highly coupled manner. We have numerous examples of storms that intensified despite high vertical shear, all during interactions with upper tropospheric troughs. It is critical to understand the role of vertical shear in an integrated framework that includes the trough dynamics.

2.4 How does the tropical cyclone core respond during trough interactions?

Eye wall cycles are common in intense tropical cyclones. We have provided evidence that such eye wall cycles are excited during trough interactions. The nature of intensity changes in existing tropical cyclones might have to include the details of inner core dynamics. If so, it is no wonder that we have made little progress on prediction. In general, we need much more data from the inner core at all levels during such interactions.

2.5 Does the nature of tropical cyclone-trough interaction differ for existing storms from pre-hurricane disturbances?

An argument can be made in terms of PV dynamics that trough interactions should have similar impacts regardless of the stage of the tropical cyclone. But eye wall cycles do not occur in pre-genesis stages, and the tropical cyclone internal dynamics obviously differs greatly for a mature storm. This issue still must be resolved.

2.6 Do identifiable parameters exist for determining whether a tropical cyclone will intensify or weaken in response to an upper tropospheric trough?

This is the fundamental forecast problem. Forecasters are well aware that "bad troughs" exist that produce weakening of tropical cyclones, although only one such case has appeared in the literature to our knowledge. Many researchers have speculated on this issue, but no method has proven successful. It is likely that our understanding will have to improve before such an advance can be made.

2.7 Hurricane Opal

Hurricane Opal intensified in the Gulf of Mexico when an upstream trough was quite distant from the storm, then rapidly filled when the trough got close. Was this a favorable or an unfavorable trough interaction? Or both? By what mechanisms did the deepening and filling occur? And, of course, what about other factors such as sea surface temperature fluctuations? Opal seems to be an excellent case for testing any theories of trough interaction.

2.8 What sort of numerical model is needed to simulate tropical cyclone-trough interactions?

Real cases are greatly complicated by other factors such as SST fluctuations, both pre-existing (as in warm eddies in the Gulf of Mexico) and tropical-cyclone-induced. As a result, there is a tremendous need to simulate tropical cyclone-trough interactions with idealized numerical models in which such variables can be controlled. We believe this will prove to be a daunting task, because it is essential that details of the tropical cyclone core be resolved in order to simulate the response of the tropical cyclone to the trough. This almost certainly requires the resolution of a cloud ensemble model or something close. The simulation of tropical cyclones on such scales is still in its infancy; for instance, we have been hearing of considerable sensitivity to the details of the microphysics when less than 5 km resolution is used. Such difficulties were anticipated long ago by Ooyama (1982). It may be that successful simulation of hurricane-trough interactions will have to await the resolution of these issues.

Acknowledgements. This work was supported by Office for Naval Research Grant N0001494-I-0289 and National Science Foundation Grants 9529771 and 9612485.


Ooyama, K.V., 1982: Conceptual evolution of the theory and modelling of the tropical cyclone. J. Meteor. Soc. Japan, 60, 369-379.

Return to TC INT 98