[time-nuts] GPS Spoofing

[Grant Saviers]

Interesting comments about navigation of ships and GPS spoofing 
potential.  As a recreational offshore captain/navigator, over the years 
I've used a few generations each of RDF, Loran, GPS, and radar.  My most 
recent extended nautical travels have been on 100 passenger 
"exploration" ships, 300' +/- with open bridge policies, where I spent a 
fair amount of time with the crews in different oceans of the world.

Disasters are most often caused by stupid, drunk, or asleep 
captains/crews, not by the electronics.  Further, the SOLAS regs for all 
commercial ships require a lot of redundancy, a minimum two each radars, 
AIS, and GPS and more are usually on the bridge.  There may be a 
"molasses tanker" out there with only a compass, but it is illegal and 
isn't going into any managed port in the world.

In any close to obstruction/land situation, radar, visual fixes, and 
soundings are the primary redundant navigation tools.  GPS is mostly 
ignored as a position plot tool, except for anchor watch, once 
anchored.    Electronic charts, often integrated with the radar are one 
plot tool and paper charts with detailed course plots, soundings, and 
visual fixes are another.

Offshore navigation is a different ball game, mostly radar and visual 
watches and hourly GPS position plots by the navigator in the log and on 
the paper voyage chart.  AIS is required on all ships and that makes the 
radar returns much more useful.  Current radars, with integrated AIS, 
all the plotting tools for traffic, closest approach plots, etc. get 
constant attention.  As mentioned, underway, a GPS often updates the 
autopilot course setting, usually track mode to avoid current drift.  A 
magnetic/gyro/inertial compass system corrects the autopilot for real 
time rudder control.

Navigation tool redundancy and crew alertness keeps ships from going 
where the ducks are walking.

Charts are frequently suspect.  On one cruise, we were in an estuary in 
remote NW Australia, nothing but blue water on the chart, no soundings, 
no depth contours.  I asked the captain about this, a bit surprising to 
me to be there in a large ship.  He had been there before, saved the 
prior successful GPS plot, and that in combination with go slow, forward 
looking sonar, soundings, and a mud bottom worked for him.  Charts with 
notations "soundings from 1894" means be extra careful.  My handheld GPS 
didn't agree with charted positions of several reefs and islands by 
tenths of NM or more even after careful cross checking of chart and GPS 
reference datums.  My old lorans sometimes showed me a few hundred yards 
"on the beach" when comfortably at anchor.  In coastal waters we learned 
that Loran C positions were fairly repeatable and somewhat inaccurate. 
Offshore, I'd check now and then any electronic position with celestial, 
the ultimate backup/crosscheck and just for practice. (pretty hard to 
spoof, also).

One navigator showed me a factor that might have contributed to the 
Concordia sinking in addition to stupid.  The web update to the required 
electronic chart (all charts paper and electronic are updated monthly on 
commercial vessels) moved the obstruction marker buoy symbol and depth 
number images on top of each other, so one had to look very closely to 
see the depth number, which obviously was not enough for the Concordia.  
So, stupid crew plus a tiny slip of a cartographers mouse sank it.

I would also note that many recreational boaters, trust GPS to take them 
in pea soup fog to the dock waypoint or through the tight pass.  YMMV.

Grant KZ1W