Mac use needs help!

[GRFox]

Ok I have a Garmin Dakota 20. I have downloaded and installed both NY and OH topo maps that I downloaded here at GPS file depot. I don't remember doing anything after down loading them but opening Garmin Basecamp and the maps were in there.

I am trying to download the CT topo maps and it doesn't show up in Bascamp and when I try to open the downloaded file it is just crazy symbolic text in a text document. The file says something about being Windows Bases. Any advice? I am suppose to be leaving on a last minute trip in a few hours!

Any help is greatly appreciated. Thanks.

[Boyd]

It helps if you tell us exactly which map you're talking about. Is this it?

http://www.gpsfiledepot.com/maps/view/638/

If so, there is no Mac version of this map available. Also note this is a new map created by someone in Russia. It may have some issues.

See the connecticut maps listed here. You can see which ones are Mac compatible. http://www.gpsfiledepot.com/maps/state/ct

[GRFox]

Yes that was the map. Sorry for not being more clear. There are only two maps listed with TOPO and that is the primary thing I need, neither of them are Mac compatible. I am guessing those are really my only options?

[Boyd]

There's a program called JaVaWa Mapconverter than is supposed to convert Windows maps into the .gmp format that is used by Macs. I have not tried it and don't know what issues it might have. I have looked at the Mapsbank NJ map myself, and it's a little strange due to the way the author used satellite imagery to produce groundcover.

Of course, there's also Garmin's 24k topo that you can download, but that costs money. https://buy.garmin.com/shop/shop.do?cID=255&pID=68271

BTW, a word of caution. Many of us have learned the hard way that it's a bad idea to install anything on your primary GPS a few hours before leaving on a trip. 

[maps4gps]

Also note this is a new map created by someone in Russia.

I wonder why they do not create maps of their own country before their stated purpose of making maps for the USA, Canada and Australia.

... it's a little strange due to the way the author used satellite imagery to produce groundcover.

I wonder how qualified/experienced the map author is to make those interpretations.  How does it compare to what you did in NJ?

Since the screen shot is not of a GPSr display, how does it look on a GPSr?

[Boyd]

There has been both public and private discussion of this at GPSReview and elsewhere.

Using Landsat imagery for landcover isn't a bad idea, but I think it's best to work with the National Landcover Dataset (NLCD) which has already classified and color-coded the data. The main issue is that the resolution is relatively low. You can still produce pretty good looking forest shading if you apply some smoothing to the image before vectorizing though.

I developed a system to do this last year and have meant to post here, but just haven't had the time. In New Jersey however, this is the wrong approach. The state already has really excellent high resolution (1:12000) landcover data available as shapefiles.

Here's a screenshot of my map and the Mapsbank map of the same area. I have not tried installing it on the GPS.

[Seldom]

I wonder why they do not create maps of their own country before their stated purpose of making maps for the USA, Canada and Australia.

I know that the USA and Canada provide free map data available on the web.  Does Russia?  Do EU members and the UK?

[Boyd]

FWIW, here's an example at 1:24000 from the Scranton, PA area with NLCD raster data converted to shapefiles using my technique.

[Seldom]

Snazzy, Boyd.  I look forward to seeing how you did it.

[maps4gps]

Using Landsat imagery for landcover isn't a bad idea, but I think it's best to work with the National Landcover Dataset (NLCD) which has already classified and color-coded the data. The main issue is that the resolution is relatively low. You can still produce pretty good looking forest shading if you apply some smoothing to the image before vectorizing though.

NLCD has a spatial resolution of 30m.
Landsat bands 1-5 & 7 also have a resolution of 30m.  Band 6 (thermal IR - good for vegetation health) is 60m and the panchromatic (black & white) band is 15m.
Problem with the smoothing in GlobalMapper is that the adjacent landcover type will be smoothed slightly differently resulting in gaps and overlaps.

BTW - Landsat 8, to be lauched in Dec., 2012, will still have a resolution of 30m with 15m for the panchromatic.

[Boyd]

Problem with the smoothing in GlobalMapper is that the adjacent landcover type will be smoothed slightly differently resulting in gaps and overlaps.

This is quite true, and you can see some examples in the screenshot above. But mapmaking requires the author to make some compromises and trade-offs. I don't think it's much of a problem at the zoom levels I personally might use. And the benefit of having forest cover on the map outweighs the registration errors that might be seen.

Also, I just delete all the polygons that represent agriculture or other areas without forest cover or development, so they just appear as the white background of the map. That eliminates any registration problems where a field is adjacent to the woods. I am only using two colors - green for any kind of trees and pink for developed areas. Anything else is just discarded from the raster image.

If I don't apply any smoothing, the stair-stepping of the vectorized pixels just looks wrong to me.

[maps4gps]

I just delete all the polygons that represent agriculture or other areas ... That eliminates any registration problems where a field is adjacent to the woods.

In the high-country of Colorado knowing if the vegetation is evergreen, dicedious, or scrub can be valuable for navigation and view locations.  For hiking, etc. agriculture is not very important as most/all is private land anyway, although hunters may find it helpful.

If I don't apply any smoothing, the stair-stepping of the vectorized pixels just looks wrong to me.

I see some/much of that in the Scranton, PA area you posted above.

[Boyd]

Well my maps are free and I make them the way I like. My idea was to mimick the style of USGS maps which use green to indicate forests and white to indicate fields. I posted a more ambitious landcover/land use map here: http://www.gpsfiledepot.com/maps/view/441/

By all means, you should use the techniques and styles that reflect your own priorities. But we've gotten way off the topic of this thread...