Skip to main content

Up until now in this column we have only considered the potential loss of GPS satellites during a voyage. However, our GPS satellites are in no way uniquely vulnerable, nor are we uniquely dependent on them for navigation. We also use satellites for communications and weather prediction; hence, we need to be self-sufficient and able to carry on without these capabilities as well. So far we have discussed the large-scale loss of satellites in fairly abstract terms, but for this article we are going to create a very specific hypothetical scenario.


The following scenario is based as closely as possible on the 1859 Carrington Event, a powerful solar storm. We will focus only on the immediate effects such an event would have on a small cruising boat at sea, including how it would affect the satellite and radio resources commonly used by cruisers. For those unfamiliar with the history of this event, here is a summary: In early September 1859, a solar coronal mass ejection hit the earth, causing spectacular auroras worldwide and also causing telegraph systems (the only large electrical systems that existed at the time) to fail catastrophically. The Carrington Event remains the largest solar storm to affect the earth since humans have used electricity in a widespread fashion. A storm of similar magnitude also occurred in 2012, but it narrowly missed Earth. NASA predicts that the probability of Earth being hit square on by a coronal mass ejection of this size over the next decade is about 12%.

The image compares the size of Earth to that of a Solar coronal mass ejection.

The image compares the size of Earth to that of a Solar coronal mass ejection.

For our scenario, we’re going to suppose we are a cruising trawler making a rhumb-line passage of 240° True from San Francisco to Kahului Harbor on Maui, a trip of more than 2,000 miles. For argument’s sake we’ll assume that the main engine is a simple diesel that operates in the absence of electricity; otherwise we’d be without propulsion and adrift.

Let’s say that the first night after we left San Francisco we observed a very weak aurora. While it was visible only because the night was moonless and we were offshore and away from the city lights, it was still a bit surprising to see an aurora this far south. Then, during our daily noon sights, we noticed that there was an unusually large number of sunspots. This, it turns out, was just the opening act of the sun’s performance.

Now our single-sideband (SSB) radio appears to be limited to line of sight, which is very consistent with the auroras. Of all of our equipment, our SSB radio is by far the most vulnerable to space weather, and we would expect skywave radio signals to be disrupted or eliminated by a solar storm of this magnitude. Our satellite resources, such as GPS and Iridium phone, seem relatively unaffected at this point. We might actually be experiencing some small degradation in GPS accuracy, but in open ocean we have nothing to compare that with, other than dead reckoning and celestial. So as far as we can observe our GPS is fine.

Now, on the fourth day underway, we’re at the halfway point, somewhere around 30° North and 140° West, literally a thousand miles from anywhere, averaging 10 knots in following seas.

During our morning Iridium satellite download we get our usual weatherfax maps and a couple of routine emails. There is a tropical storm, Brigid, west of Acapulco and heading westward generally toward Hawaii. It is still more than 600 miles away and should pass well south of us.

But we have one additional message from our weather service provider, marked “Urgent Space Weather Warning.” Is this an old message that came late, telling us about the auroras from the other day? No, this is new.

A sample solar storm alert from NOAA.

A sample solar storm alert from NOAA.


06:00 UTC is 20:00 local time in Hawaii, about an hour after sunset. Hawaii will be on the night side of the planet for the first 11 hours or so of the storm, and due to a latitude of only 20° North, the distance from either of the magnetic poles is close to ideal to minimize the effects of this solar event. For our boat and our destination port, things are much better than they could be for an event like this, and much better than they will be in other places.

Before sunrise we take a celestial fix on Jupiter and the stars Betelgeuse and Schedar, compare it to our GPS position, and lay out our dead reckoning track from these. We will continue plotting our GPS position hourly for as long as we can.

The sun rises on a beautiful day in the Pacific tradewinds, the weather is clear with following seas, and other than the sparsity of radio traffic on our SSB, everything is pretty routine. We still have about 14 hours before the coronal mass ejection (CME) arrives. Much like a hurricane, a CME allows us ample time to prepare to the best of our ability. Unlike a hurricane, however, avoiding the solar storm entirely is not a possibility.

We spend the day in preparation and take inventory, looking at both the best-case and the worst-case scenarios. Our first actionable priority will be stowing as much of our small electronics as we can in anything that might pass for a Faraday cage. Anything from a covered wire-mesh garbage can to a sealable metal coffee can or the oven will do the trick. We disconnect as much of our electrical equipment as possible, pump our holding tank, top off our potable tanks with the watermaker, and move diesel to our day-tanks.

This is one of those rare times that you would want to use a computer printer if you have one. Up until now, we haven’t been especially concerned about the aforementioned Tropical Storm Brigid, which was more than 600 miles away from us last time we checked. But the possibility of being adrift out here in its path gives us a renewed awareness of its significance. With our Iridium phone we download and print the latest storm tracks and surface analysis maps, as well as plot the relevant information from these on our Pilot Chart. Assuming we still have propulsion after the CME, we’ll only have about four days to Maui, so we also print the 24-hour, 48-hour, and 96-hour weatherfax surface forecast maps, as this may be the best external weather information we’ll have available.

A map showing the route of a hypothetical Pacific cyclone that is heading west towards the Hawaiian islands.

A map showing the route of a hypothetical Pacific cyclone that is heading west towards the Hawaiian islands.

Verifying our aneroid barometer with the surface analysis charts and our electronic barometer is also critical, as it may be the only direct weather information we have other than our eyes for the rest of the trip. Similarly, verifying our magnetic steering compass is a necessity as well.

Any other navigation and weather resource we will need for the remainder of our ocean passage and our approach to Kahului, such as excerpts from the U.S. Coast Pilots and tide tables, should also be printed in case our hard drive becomes inoperable or inaccessible. Enlarging and printing the portion of the Pilot Chart that includes our route may prove beneficial. Important files should be transferred onto a flash drive and secured in one of the makeshift Faraday cages.

Plotting our track on the Pilot Chart, it is apparent that our rhumb-line route crosses the great-circle track from Los Angeles to Honolulu that is preferred by container ships and other commercial traffic. Thus, we will need to make sure to restore or replace our running lights after the CME arrives.

We must also decide whether to keep our engines running as the CME passes through or to shut them down and try to restart them afterward. If our engines are dependent on electronic systems in order to operate, shutting down is probably best, and we need to be prepared for them to never start again. At the other extreme, a very simple diesel engine with a pneumatic or other mechanical-only start may not be affected by the CME at all. As with automobiles, for any kind of electromagnetic pulse, older engines are better than new and diesel is better than gasoline. For cruising boats, sails are better than any of them, if you happen to have them or are able to jury-rig one.

In our case we’ll simply assume that our engine is an older diesel. Given the fact that we’ll be on the night side of the earth when the CME hits, we’ll take a chance and keep our engine running and engaged but disconnect any electrical loads from it. We’ll power down our radar and disconnect the coaxial, but realistically we will have to write it off as a loss and hope for the best.

Finally, we make a nice meal from our perishables because anything we can’t keep cold will have to be thrown away. Nothing much to do now but relax, continue motoring into the sunset, and wait.

Luck favors the well prepared

In the evening twilight we break out the sextant and shoot the stars Alkaid, Antares, and Deneb, and draw the fix on our plotting sheet. We take a small chance and remove a cheap handheld GPS receiver from one of the Faraday cages to get a last comparison with our celestial. The receiver boots up fine but has no satellite signal. Whether this means that the satellites have already been damaged or that they’ve simply been taken offline to protect their electronics, we have no way of knowing. Regardless, we put the GPS receiver back in the Faraday cage.

As the sky darkens we see a faint green glow to the north; if we weren’t looking for it we likely wouldn’t notice it. It remains there, like a faintly luminous cloud, for the better part of an hour. Then suddenly the sky erupts in greens and reds, sparkling emerald curtains dance in the night above us. Static electricity builds quickly on metal surfaces, especially the rails and lifelines, giving occasional mild shocks. To our relief, the old engine rumbles on undisturbed.

The aurora continues throughout the night, sometimes strong and dramatic, sometimes retreating to a green glowing cloud in the northern sky, but never fading completely. The brightest stars are visible through it, and the horizon is sufficiently illuminated to allow celestial navigation sights through the night, as desired. Our wristwatch seems unaffected by the solar storm. Our small battery-powered backup running lights are similarly unaffected, so we mount and illuminate those for safety. Our first night, presumably the worst of the solar storm, is otherwise uneventful.

We have no updates on Tropical Storm Brigid. However, as we monitor and log our barometer hourly, we see no precipitous drops in barometric pressure. Watching the skies for telltale signs of an approaching low has similarly yielded nothing; the ubiquitous popcorn cumulus of the tradewinds march steadily westward, with no cirrus clouds anywhere in sight. Either Brigid has passed far to the south of us or more likely it simply dissipated as eastern Pacific tropical cyclones often do.

Our current ETA into Kahului has us arriving a little after midnight seven days later. Given the likelihood of extensive power outages on the islands, including lighted aids to navigation, we decide to slow ourselves a bit both to conserve fuel and to ensure that we make landfall on Mt. Haleakala soon after sunrise so that we can approach Kahului Harbor in daylight.

Fortunately, Maui’s electrical grid is small and much of it is wind-powered, so critical loads will likely be restored there much more quickly than on the mainland. Even more fortunately, electricity is not required to make a Mai Tai.

Good watch!