Mexican Charts - Arrrggghhh!

[phiggins]

Beware of some of the Mexican charts....

Three years ago I took my boat as far as Mexico. While in Isla Mujeres we heard of a Swiss boat that had run aground south of Puerto Morelos about 35nm south of Cancun. A few of us decided to sail down and see if we could help him get off the reef.

Unfortunately he was very aground with his double keeled steel boat and all our efforts failed to get him off. The local tugboat guys said they would tow it off for $20,000 which he didn't have. After two weeks the pumps were failing and the boat was being smashed to pieces by the surf so he abandoned the boat and flew home with his dog.

Before he left he told us to take anything we wanted before the vultures got it. One of the things I took was his "Cruising Guide to Belize and Mexico's Caribbean Coast" second edition.

As I plan to sail back there next fall I ordered all the charts from Mexico to Panama from Bellingham. As I haven't been able to find any BSB charts for Mexico I decided to scan all these paper charts so I could use them in SeaClear. After doing so I decided to check them using Google Earth.

Good grief the Puerto Morelos chart 28201 has its longitude off by +2 minutes, nearly 2 nm! The chart was from a Mexican survey in 1905 but even still it hasn't changed that much.

I've often wondered why our Swiss friend ran aground but if I look at the cruising guide from his boat it uses the same chart for information and has the same error.

Thank Goodness for Google Earth. I'll check the rest of the charts.

[Charlie]

I have heard comments about Mexican and other charts before and was wondering if someone would clairify this point. Is it true: that when using a paper chart based on an older datum If you use proper costal piloting techniques such as range and bearing you will place you boat more accurately on the coast then you would by using the GPS lat and long. To take it a step further, assuming an accurate piloting technique, the charts are accurate enough to navigate by. It seems to me that the solution to the problem is to use older techniques with older charts.

[Ziggy]

You can't just take the Lat/Lon from your GPS, which will probably be referenced to the WGS-84, and plot it on a random chart. Instead, you need to determine the chart datum and set your GPS to it. Many Mexican charts use the NAD27-Mexico datum. I don't know whether that's the problem with using the Mexican charts, but it may be.

[phiggins]

Yes I guess using older techniques would probably work. The Cruising guide I mentioned does do that but includes the latitude and longitude lines which could throw off an novice and could use them as fact.

The chart does not specify a datum but does say "Adjustments for plotting positions obtained from satellite navigation systems based on WGS-84 cannot be determined for this chart."

I guess I'm surprised that using WGS-84 with any published chart would cause such as disprepency. Guess I'll be a lot more careful now that I know how much they can be off.

[orion1]

I seem to remember reading somewhere recently that Nigel Calder once said; "If one cruises to the west coast of Mexico by GPS alone they will be on a reef within a week".

Has anyone else read this statement, or similar words from Mr. Calder?

[Viking Sailor]

Originally Posted by phiggins

Beware of some of the Mexican charts....

As I plan to sail back there next fall I ordered all the charts from Mexico to Panama from Bellingham. As I haven't been able to find any BSB charts for Mexico I decided to scan all these paper charts so I could use them in SeaClear. After doing so I decided to check them using Google Earth.

Good grief the Puerto Morelos chart 28201 has its longitude off by +2 minutes, nearly 2 nm! The chart was from a Mexican survey in 1905 but even still it hasn't changed that much.

Paul

Since you have already scan in the chart, can't you use Google Earth to re-calibrate the chart and thus reduce its error?

[Lodesman]

Originally Posted by phiggins

Good grief the Puerto Morelos chart 28201 has its longitude off by +2 minutes, nearly 2 nm! The chart was from a Mexican survey in 1905 but even still it hasn't changed that much.

What datum does the chart use? Or is there a correction to WGS84 (NAD83)?

[Jmolan]

This is a good read. The Pardey's pilot into the anchorage then get out the GPS to check their position, which puts them on land. If the charts were made 100 years ago, does it not make sense they are not as accurate as the GPS today?

Sailing with Lin & Larry Pardey



When I was a greenhorn deckhand, the Capt. came up from a nap and growls.... " show me where are we?".... I took a sharp pencil and made a dot on the paper (all we had) chart. So he sez, you sure bout' that? So I grab the radar and get bearings, grab the compass and get bearings....move my pencil point over a few clicks...ah yea, this is where we are, I say.
He grabs a dull pencil, makes a circle about a mile around and sez "we are in here somewhere", Never think you know exactly where you are., or you may end up on the rocks, makes you too confident.....
I never forgot that. I pilot a 125' steel vessel with 16' draft, and there are places we go where we rub bottom at times. We have to know exactly where we are. But this is done on a previous tracks. GPS is perfectly accurate on a repeatable course. With GPS early on many people went on the reefs out in the pacific. Instead of standing off 5 miles to allow for error, they plowed right up on the reefs because it was suppose to be water...

[Talbot]

Datum errors are the prime reason for inconsistencies between charts. This is particularly true if the original chart was created using a land datum.
it is only in the very recent past that land and maritime charts have begun using the same datum (WGS 84) and this is solely due to GPS. Prior to that, datum corrections were a fact of every day life.

Some of the pacific islands for example are perfectly aligned with each other as they used the same datum, but all were several miles out in accordance with GPS. Any chart that has a date of 1905 would immediately be suspect.

[Lodesman]

1905 was the date of the survey - not necessarily the date of publication. I've used charts that had modern datums, where the survey was done by Cook. One should never rely solely on GPS when navigating inshore.

[phiggins]

There is no datum specified for the chart so I tried to recalibrate it using Google Earth but since there was no landmark that I could recognize from 1905 chart I was wasn't very successful.

I then decided to make a new chart from Google Earth and information from the cruising guide by saving the image as a JPG, converting it to a BMP and then calibrating it with MAPCAL.

It actually produces a fairly good chart that is usualable with a GPS. I've attached the JPG file of it. I can easily see the deep water and reefs on it. As many have said you shouldn't depend solely on GPS close to shore but it does provide an extra measure of confidence when you don't have a good chart.

[Cheechako]

The actual paper chart was off (in the Sea of Cortez area) substantially when I was there in the 80's/early 90's. It was common knowledge with those who had been down there a while. Probably due to the age of the survey that established the chart.

[Pete the Cat]

Charts of the West Coast of Mexico are off. For instance, Isla Isabella is about .6 mile off its charted position. In going from SF to Florida, I found discrepancies (much fewer where the CIA operated in Nicaragua and Costa Rica!) of .3 to .7 mile, but not all that often. This was not due to datums being wrong (I corrected these often), it was the fact that GPS is a lot lot more accurate than the clocks and sextants the guys were using in the 1800's when the basic charts were drawn. Don't know about anyone else, but I could never get that close with a sextant so I have respect for them.

[phiggins]

The navigation program Fugawi has a pretty neat feature that allows it to interface with Google Earth and synchronize your chart with the GE picture. It even allows you to overlay the chart on the GE picture and adjust the transparency so you give see if the chart is accurate. You can then save the combined picture and create a new chart. Pretty cool.

Only thing I wish it had is some way to adjust the chart overlay horizontally and vertically so you could recalibrate the chart if it doesn't match the actual GE photo. Would have made it easy to fix these bad charts.

[phiggins]

I checked the rest of the charts in the Bellingham set (Yucatan Channel to Panama) that had no datum specified or a local datum and had no correction since the original survey, six total. The surveys were done either by the US or British navy back in the 1800 or early 1900's. The worse one was only off by less than 5 seconds , around 500 feet. At least they would put you in the ballpark using a GPS.

Only that one Puerto Morelos chart had the big discepencies of 2 nm. I manged to correct the chart by subtracting .036956 degrees from the longitude and adding .004949 degrees to the latitude for all the calibration points.

I think its criminal to keep publishing that chart. I guess it keeps the local tug boat companies in business.