Friday, April 17, 2009

How To Make A Really Big Map

New Britain Island - Landsat ETM+ S-56-00_2000 Image Resized (1-375,000) Northwestern End

New Britain Island - Landsat ETM+ S-56-00_2000 Image (1-375,000) Northwestern End


New Britain Island - EVS Precision Map (1-75,000) Northwestern End

New Britain Island - EVS Precision Map (1:75,000) Northwestern End (Select image to see full-size)

I'm such a pushover when it comes to reader-requested mapping projects. Some of the projects are relatively simple (a single island with a couple of layers of information), others more complex (a single island with many layers of information) and finally the very complex or lengthy mapping project consisting of many layers of information or covering a large area with a variety of detail. My last mapping project, the northwestern end of New Britain Island, covered a large area and the reader was keenly interested in the offshore reefs.

Back in January 2009 the reader's initial query was deceptively simple, "Have you done any reefs in New Britain, Papua New Guinea?" During the summer of 2008, I had mapped all of the islands located between and including Papua New Guinea to the Cook Islands which consisted of over 10,000 island polygons using Landsat ETM+ as my base imagery. One of those islands was New Britain. I figured the project would be a snap. The reader was only interested in the northwestern end of New Britain. I would ensure my original island polygon was sufficiently accurate and then I would digitize and classify all reefs into a single vector layer called Reef Shallow. No problem!

Along with trying to be an attentive husband, loving father, generous grandfather, concerned and effective teacher and all-around good guy, my life got very busy. So busy that the New Britain Reef project got pushed behind my busy life. I informed the reader that I would have to table his project for the time being and since the project was "gratis", he waited. And so the New Britain Reef project resided in my computer waiting for my return. All of the shoreline was finished along with a portion of the reefs. My initial work had taken about 20-hours to complete. The week before last, the patient reader asked if I would have an opportunity to complete the project. Frankly, I had forgotten about it. Feeling guilty for not completing it way back when and knowing that I had a week's vacation coming, I promised to finish the project. And I did! It took about 10 more hours, but I finished it!

I studied my completed work and realized that this mapped area was really large. Since this reader did not require my vector files, I would have normally sent the reader a screenshot (raster image) the size of an 8.5 X 11 page of their project and that would be the end of it. However, this reader would be unable to appreciate the detailed reef information at that scale (1:350,000). I decided to construct a large format map at a scale of 1:75,000 giving the reader sufficient detail for his ultimate purpose, finding new fishing locations.


New Britain Island - EVS Precision Map (1-75,000) Northwestern End with Grids and Labels

New Britain Island - EVS Precision Map (1-75,000) Northwestern End with Map Tile Labels


W3

1 of 15 EVS Precision Map Tiles (W3)

I have constructed numerous image mosaics, but never a map mosaic. Using images from Global Mapper and mosaicking them into a single image using Image Forge software and finally constructing lat/lon grid lines in MS Draw, I created "a really big map" (thus the title of this post). All of you mapping professionals, calm down. I know mapping software exists that will make "a really big map" in a single step. Typically, it costs money and that's something I don't have much of. Anyway, back to my mosaicked map. "The really big map" consists of 15 tiles , each tile similar to the one displayed in the above image. My final map is 3050 X 4220 pixels in size or about 65 x 89 cms (25.5" x 35").

My final thought, I've got to start charging for my work! I spent 30+ hours on this project. At a modest fee of $20 per hour, I should have made $600 for this project. Instead, New Britain Island's northwestern end is mapped in great detail and I'm none the richer for it.

Enjoy!

Wednesday, April 15, 2009

Musandam Peninsula OM - A Revisit of a Previous Post


View EVS Islands Projects in a larger map


Oman - Landsat ETM+ N-40-25_2000 (1-175,000)

Oman - Landsat ETM+ N-40-25_2000 (1:175,000)


Oman - EVS Precision Map (1-175,000)

Musandam Peninsula - EVS Precision Map (1-175,000)


Musandam Peninsula - EVS Precision Map with 50-Meter Contour Intervals (1-175,000)

Musandam Peninsula - EVS Precision Map with 50-Meter Contour Intervals (1:175,000)

The above EVS Precision map project of the Musandam Peninsula was completed around the first week in September 2008. It was my last mapping "hurrah" before the start of this school year. I remember loving the look of this unusual piece of land calling it a "peninsula of peninulas.


Musandam Peninsula Oman from EO (musandam_AST_2004067_lrg) Resized

Musandam Peninsula, Oman from Earth Observatory (EO)

It seems that the folks at EO dusted off a 2004 ASTER image of the Musandam Peninsula and posted it to their site today. The peninsula is still lovely to look at!

Enjoy!

Monday, April 13, 2009

The Great Blue Hole Located on Lighthouse Reef, Belize


View EVS Islands Projects in a larger map

Great Blue Hole, Belize Resized (greatbluehole_ali_2009083_lrg)

Lighthouse Reef and The Great Blue Hole from Advanced Land Imager (ALI) on NASA’s Earth Observing-1 Satellite

On April 3rd the above image was posted to Earth Observatory's Image of the Day. I read the article with mild interest. I was aware of the these features, cenotes, being found throughout this part of the world. After downloading the 4 MB image of the entire reef, I figured that it would be an excellent island mapping project. I checked out the Landsat ETM+ image and it is cloud free and quite detailed. So, I began to map it.


Great Blue Hole - Comparison Landsat to ASTER Images

Lighthouse Reef and the Great Blue Hole - Comparison Landsat to ASTER Images

The small cays that are located on the reef at first glance looked to be fairly simple. They were not! Wanting to do justice to the details shown on the base image, my digitizing was ultimately complex. I worked at scales between 1:25,000 to 1:12,500 for most of my digitizing.


Lighthouse Reef and The Great Blue Hole, Belize - Landsat ETM+ N-16-15_2000 Image (1-220,000)

Lighthouse Reef and The Great Blue Hole, Belize - Landsat ETM+ N-16-15_2000 Image (1-220,000)


Lighthouse Reef and The Great Blue Hole, Belize - EVS Precision Map Preliminary (1-220,000)

Lighthouse Reef and The Great Blue Hole, Belize - EVS Precision Map Preliminary (1-220,000)

10-days and about 20-hours later, I am done. All of the layer colors used are my usuals, except for the ocean layer. I went with a darker blue. In my opinion, the deep blue ocean color accents the many shades of blue used in detailing the reef system.


Great Blue Hole (National Geographic) belize-blue-hole-reef-731526-sw

Great Blue Hole from National Geographic

If you are curious about this feature follow the links found throughout this post. Wikipedia also has an informative article about the Great Blue Hole you might find interesting.

From a mapping perspective, I am pleased with my finished product. However, to truely appreciate the amount of detail included on the map one needs to study the vector map. It is really cool! Oh, well...I'm glad it's done. Now on to the north coast of New Britian and more reefs.

Enjoy!

Tuesday, March 24, 2009

Manihi Atoll FP - First Landfall in the Tuamotus


Manihi Atoll Locator Map


Manihi Atoll - Landsat ETM+ Image S-06-10_2000 (1-110,000)

Manihi Atoll - Landsat ETM+ Image S-06-10_2000 (1:110,000)


Manihi Atoll - EVS Precision Map (1-110,000)

Manihi Atoll - EVS Precision Map (1-110,000)

Manihi Atoll is one of French Polynesia's premier destinations for the tourist wishing to view pearl farming first hand. According to the Tahiti Tourism North America, Manihi Atoll has over 60 working pearl farms. While one is looking for those perfect pearls, the Manihi Pearl Beach Lodge beckons. This atoll is one of the first landfalls for the visiting yatchsmen traveling from the Marquesas to Tahiti.


Manihi map-big (500)

Manihi Atoll - Pacific-Image Map

In doing research for Manihi Atoll I came across the Pacific-Image map. As to names of the various motus, it is complete. As to cartographic quality, it looks like a map, it seems like a map, and it is a map, one designed for the tourist. If you are into high quality tourist maps of Pacific Islands, specifically the islands of French Polynesia, for only $237.60 you can purchase a CD loaded with 72 of these very attractive maps.


Detail Manihi Map Image-Pacific

Manihi Atoll - Detail from Image Pacific Map


Manihi Atoll - EVS Precision Map - Detail Tairapa Pass and Paeua (1-30,000)

Manihi Atoll - EVS Precision Map - Detail Tairapa Pass and Paeua (1:30,000)

Although the Tahiti Pacific-Image maps are excellent for their purpose, I like mine better. Compared to the Tahiti Pacific-Image map, my EVS precision map of Manihi Atoll takes into account a wider variety of layers of information. In addition, my maps are vectors, whereas Tahiti Pacific-Image's maps are raster images.

As to the production of this map, it was not particularly difficult to digitize and ultimately create the island map. However, my very busy life caught up with me and what should have been a 10-hour project, took two weeks to complete. I have other island projects in the works, but they will be completed and posted at this much slower pace. Until then...

Enjoy!

Sunday, March 8, 2009

Nisos Aegina GR - An Athen's Island Getaway.


Nisos Aegina - Locator Map


Nisos Aegina - Landsat N-35-34_2000 (1-85,000)

Nisos Aegina - Landsat N-35-34_2000 (1:85,000)


Nisos Aegina - EVS Precision Map (1-85,000) Modified

Nisos Aegina - EVS Precision Map (1-85,000)

Nisos Aegina is located only 17.5 kilometers from Athens. The island serves as both a "bedroom" community for Athenians and as a summer getaway for city dwellers. Many of the homes located on this island are second homes of Athenians. Wikipedia has an in-depth article about the island.

Let's talk about the map. I have spent in excess of 6-hours on this map. The challenge was in determining the layers and finding a good color scheme that is pleasant to look at, yet imparts accurate and meaningful information. I used my entire suite of mapping software to complete this project - 1) Global Mapper to create vector layers and identify lat/lon grids. 2) MS Paint to draw the lat/lon grid lines and finally 3) ImageForge to complete the labels and their placement on the map. I realize most decent mapping packages come with these features all included. I am a creature of habit and like my end products. So if it isn't broken, why fix it!

Enjoy!

Thursday, March 5, 2009

Thoughts - Lukuga River - A Water Road into Africa's Heart of Darkness

Lukuga River - EVS Precision Map (1-1,122,800)

Lukuga River - EVS Precision Map (1:1,122,800)


Lukuga River Composite - Each Image at scale 1:50,000

I remember spending long hours digitizing the many features on this river map. It is 290 kilometers in length and contains 989 river islands. When finished, I eagerly posted a few images of this grand river. I titled the post something about "the heart of darkness". The original Lukuga River post gets lots of views. I think it has to do with "the heart of darkness" label and less about the Lukuga River. Well, it hit me this evening. The reason the Lukuga River map doesn't get the respect it deserves is you the viewers have not seen the entire course of the river. So enjoy the slideshow entitled "Lukuga River - A Water Road into Africa's Heart of Darkness".

Wednesday, March 4, 2009

Thoughts - Bigger is Good, Smaller is Better


View Larger Map

Nisos Aegina - Landsat Island Image Mosaic N-34-35_2000 (1-125,000) Modified

Nisos Aegina - Landsat Island Image Mosaic N-34-35_2000 (1-125,000)

I am currently reworking my map of Nisos Aegina. It is a bedroom island outside of Athens, Greece. A nice little getaway when life becomes too hectic in the big city. But this isn't the reason for this post.

You know that I construct island maps using Landsat ETM+ mosaics as my base imagery. I have a Landsat ETM+ digital library of all of the world's shorelines. I use Global Mapper software to digitize, creating vectors from which i construct my island maps. This software handles the Landsat ETM+ mosaics with ease. I have invested over $500 in Global Mapper and do not regret a cent of it. The software does exactly what I want it to do and Mike, the boss and software guru, keeps making improvements. But this isn't the reason for this post either.

This post poses the age-old question "Is Bigger Better?" as it applies to the world of imagery.

In the world of Landsat derived island specific orthorectified imagery, bigger is not necessarily better. This is the world of imagery anywhere from multi-GB files to a few hundred MB files. These very large image files test the limits of most software packages. Often their formats, (i.e., MrSID) require special handling to open and work with. Since I primarily work with island imagery, I figured that if I were to cut out the ocean and save the island and it's surrounding reefs at a scale allowing me to work at increasing zoom ranges, I could create the same high quality maps from much smaller sized imagery.

First I fashioned a series of Greek island mosaics (see above image). These were followed by a series of Tuamotu Atoll images. I call my creations Landsat Island Image Mosaics (LIIM). Here are a couple of examples showing the zoom capabilities using the Nisos Aegina LIIM as base imagery.


NIsos Aegina - Detail Western Shore LIIM N-34-35_2000 (1-25,000)

NIsos Aegina - Detail Western Shore LIIM N-34-35_2000 (1-25,000)


NIsos Aegina - Detail Aegina Harbor LIIM N-34-35_2000 (1-12,500)

Nisos Aegina - Detail Aegina Harbor LIIM N-34-35_2000 (1-12,500)

I typically digitize shorelines and certain land use features at scales between 1:25,000 to 1:12,500. I know that this pushes far beyond the intended limits of Landsat-7, 14.5 meter resolution imagery. I have digitized at these limits for over four years. The resulting maps have a level of quality and precision that exceeds most widely available island mapping. My island maps have been used by a number of universities to further research and as a platform for spatial analysis by insurance companies.

The size advantage with these considerably smaller island images is significant. My LIIMs range in size from 2 MBs up to 45 MBs for very large islands. All of my LIIMs are saved as georectified JPGs. Nisos Aegina LIIM is 3.47 MBs and is a 15.8 MB Tif file. The Landsat image containing this LIIM, on the other hand, is 106 MBs. The image degradation is slight and barely noticeable. So, I will keep using my LIIM of a specific island from which to create my island maps. If you would like to use one of my LIIMs, drop me a line. I'll fix you up.

Enjoy!