Thoughts on the imminent loss of SSMIS data for tropical cyclone forecasting

The loss of these data will have significant impacts on the accuracy and timeliness of NHC’s tropical cyclone analyses, forecasts, and warnings.

Passive microwave sensors like SSMIS function much like a radar from orbit, cutting through high clouds and revealing the detailed cyclone structures beneath, which are often obscured in conventional visible and infrared imagery. For example, the first indication of a developing eyewall (an important trigger for rapid intensification) usually comes from a microwave pass. In weaker and formative systems, these data can help determine when a disturbance has enough organization to be classified as a depression, and can improve forecaster estimates of storm location by 60 miles or more.

Aircraft reconnaissance is the best way to assess storm location, intensity, and size, but In the Atlantic basin roughly ⅔ of all NHC advisories and forecasts have to be made without the benefit of such data, and nearly all eastern Pacific forecasts lack aircraft data. So satellite sensors like the SSMIS are an essential part of the forecaster’s toolbox - and are sometimes even better than aircraft at describing a tropical cyclone’s overall organizational structure.

When the SSMIS data stop flowing to NHC, the number of microwave passes available to NHC will be cut by roughly one half, and the ones that remain will be less detailed. Sampling gaps of 12 hours or more will become commonplace. While forecasters have access to a wide array of data, there is currently no practical substitute for the SSMIS. I would expect the following impacts:

1. Poorer estimates of the initial position of weaker systems will cascade into poorer track forecasts for these systems.

2. For tropical storms not under surveillance by aircraft, not being able to observe an incipient eyewall could delay real-time upgrades to hurricane strength by 6 hours or more.

3. The lack of microwave passes overnight will inevitably produce cases of hurried and abrupt changes to NHC track and intensity forecasts shortly after first-light visible imagery arrives (the “sunrise surprise”). In general, I expect the number of Special Advisories will go up once forecasters start losing the advance notice of structural changes that the SSMIS provides.

4. Forecasts of rapid intensification, which had been improving in recent years, could sometimes now be delayed by 6-12 hours because forecasters will not see the improved structural organization of a storm until it subsequently becomes apparent in conventional imagery, or until the onset of intensification has already begun and been observed by an aircraft. This could translate into the delayed issuance or expansions of some coastal (and particularly inland) warnings, or force undesirably abrupt changes in forecast intensities from advisory to advisory.

5. The risk of poorer forecasts from the loss of the SSMIS data will be particularly increased for the Caribbean nations, Central America, and the Pacific coast of Mexico, where recon flights are less frequent. And at greatest risk will be the marine community (see El Faro, Fantome, etc.).

(In addition to the direct impacts on forecasters, the satellite data also get ingested into numerical models. However, I don’t have a good feel for what the model impacts might be.)

This reduction in the availability of microwave data comes at the same time as reductions in weather balloon launches, staffing reductions at WFOs, the loss of modeling talent at EMC and the NOAA labs, and the threat of budget cuts that would gut NOAA Research. While there is resiliency in the hurricane forecasting enterprise, where strength in one area can cover for a weakness elsewhere, that resiliency is being stripped away, piece by piece. I fear that our hurricane products and services are suffering a death by a thousand cuts.