EVS-Islands

Haifa Harbor - DigitalGlobe Image from Google Earth (1:4,600)

Let's begin this comparison of shoreline accuracy by superimposing four shorelines onto a 1-meter resolution image of Haifa Harbor. Why Haifa? Just because I happened to be making a land-sea mask in this part of the world. My initial study focused on using NGA's Prototype Global Shorelines (PGS) created from Landsat ETM+ imagery. As you will read and see this global shoreline is one of the most detailed available. It does have a few negative considerations, like "stuttering" shorelines, non-mapping of water-inundated land (swamps, marshes, etc.) and hand-tracing of shorelines is more accurate than the software generated shorelines...but, that said, NGA PGS shoreline is a great first-step toward a global shoreline for all of us.

Haifa Harbor - 1:2,000,000 (Red) and 1:250,000 (Blue) (1:4,600)

The 1:2,000,000 (Red) shoreline is, by virtue of it's name, designed to be used at 1:2,000,000 or smaller resolutions. It works well for coarse shoreline delineation to get "an idea" of where the shore ends and the sea begins. The 1:250,000 goes by the acronym W(orld) V(ector) S(horelines). It was originally extracted from Tactical Pilotage Charts. It gives one a useful shoreline for a wide variety of projects. However, when one pushes it beyond 1:250,000 it quickly proves inaccurate in tracing the actual shoreline.

Haifa Harbor - NGA PGS (Yellow) and EEVS Precision (Lite Blue) (1:4,600)

Now for the good stuff! The NGA PGS precision shorelines do a fairly good job of tracking the actual shoreline. They are deemed appropriate at scales between 1:125,000 to 1:250,000. Some areas could be pushed to 1:62,500, but be careful. I am careful and I push them below 1:62,500, for digitizing purposes, in order to confirm my Landsat derived hand-made vectors are of greater precision than the PGS. And mine are! For the purposes of this comparison the PGS shorelines would give a person an accurate shoreline impression as long as it was not superimposed over the actual shoreline. The lite blue line is what I call E(xtremely) E(nhanced) V(ector) S(horelines). They are derived from DigitalGlobe's 1-meter resolution imagery found in Google Earth. Oh, that all the world were imaged at 1-meter cloud-free resolution. I feel that is a long way coming and I don't think the global shoreline vector files derived from them will be free. I suspect they will be quite costly.

Back to the land-sea mask, or what started this current study. The most accurate vector based global land-sea mask is mine. I call it GSHHS Derived Global Land-Sea Mask and you can download a copy here. Be forewarned, the file is 83.9MBs zipped and 340MBs unzipped.

The most accurate future global shoreline map should be the NGA PGS precision product, but it needs work to finish it (correct stutters, map non-mapped areas, etc.). Once it is finished, it will be THE GLOBAL SHORELINE. Until then, WVS rules.

Enjoy!

Labels: Land-Sea Mask, NGA, PGS, Tactical Pilotage Charts, WVS

Posted by Mr Minton at 11/25/2007 07:43:00 AM | Permalink | Comments | links to this post

Just Thinking About It

Back in 1964, I was checking out reference books at a library. I happened upon a Geographical Dictionary. Leafing through that dictionary, I came upon an entry describing a small obscure island, Eiao. The entry stated that this island was 8 miles by 4 miles with elevations of 3,000 feet in the island's central plateau. At one time the island was inhabited. During the later part of the 19th century, the French used the island as a prison. Now the island was overrun by feral animals which were decimating what little native vegetation remained.

From that single incident came my fascination with islands. Like many people with a keen interest in islands, my reasons are many and most of these reasons having nothing to do with geography (stress, explorer, master-of-my-domain, etc.). What I discovered early on was the absence of readily-available maps. There were a few WW-II maps in out-of-print books, but no available collection of island maps.

Scripps Institute of Oceanography and Maps

That led to Scripps Institute of Oceanography and their excellent map collection. SIO does ocean-oriented research throughout the world. Consequently, they have a paper map collection of the world oceans and islands that rivals the best in the world. During the early 1980s, I spent many weekend hours pouring over this collection of maps and charts. The map librarian and I became acquainted and he allowed me access to their entire map and chart collection. It was both a joy and a curse, a sensory overload. I had access to every island paper map in the SIO collection and all I could think was, "So many islands, so little time."

My first idea was to amass my own collection of paper maps. This was the pre-digital and pre-affordable computer age. I would need to copy thousands of maps. No way! At 15 cents a copy, it just wasn't practical or affordable. I thought I could trace the shorelines of islands. Too many islands, too complicated, too time-consuming. Again, No way! So I stopped. That's right, stopped any island map aquisitions activities and went about my non-island map life.

Affordable Computers...

Which takes us to the late 1990s, the time of affordable computers, mega-storage, www and information everywhere. This was a time of digitizing paper map collections and one of the first was the Perry-Castenada Map Collection at the University of Texas. I'm not sure of the precise date that their collection went on-line, but many paper maps were freely available as raster images. Anyone could download them and amass their own digital map collection. During this time, I downloaded 100s of island maps. It was great fun, but I knew there was more. Raster maps were okay, but what about vector maps. They were considerably smaller in size and offered greater flexibility. In 2001, I went the way of the vectors.

The Way of The Vector

Not having money to spend on software, I became the master of the freeware and shareware world. I finally settled on Marplot for my mapping software and World Vector Shorelines from the NGDC Coastline Extractor. I remember downloading 100s of vector tiles onto 1.4MB floppies, taking them to my computer where I uploaded them into Marplot. It took about a month and I managed to load all of the world shorelines at 1:250,000 scale, 20,000+ placenames of major cities, 86,000+ placenames of islands and various other opportunistic layers of information.

Free Landsat ETM+. Global mapper and Google Earth

It was about three years ago that I discovered free Landsat ETM+. Turns out Bill Clinton, at the urging of Al Gore and others, made Landsat ETM+ georectified photo mosaics available for free. I spent that summer downloading all of the Landsat ETM+ images that covered oceanic islands of the world. It was at this time that I actually spent hard currency on my mapping passion and purchased Global Mapper. This software can load Landsat ETM+ in MrSid format effortlessly and has an above average digitizing toolkit. The final piece to my mapping project puzzle was Google Earth with DigitalGlobe imagery. As more and more hi-res imagery comes online, the quality of my work will be greatly enhanced.

What Now? Just Do It!

Since 2004, I have been making island maps. Since 2005, I have been posting my island maps on to my blogsite. Since 2007, I have been producing high-quality vector maps of islands. To date I have produced close to 200 maps covering easily 1,000+ individual islands. People ask why I give these maps away. They suggest that I need to charge for my work. I say to them, "We'll see." What I mean to say is, "No, I will not charge for my raster images of maps." If you want them, you can download them and use them. Just give credit where credit is due.

Today and In To The Future

I would love to guide an international team of neo-cartographers to complete a world shoreline map, including islands, using Landsat ETM+ imagery. Yes, I know all about the NGA Prototype Global Shoreline. Trust me, my world shoreline will be more accurate. I would love to guide that same team of international neo-cartographers in fashioning the next-generation Digtal Chart of the World. I would love to make all of this information available for free. In the spirit of Google, free.

Enjoy!

Labels: DCW, DigitalGlobe, Eiao, Google Earth, Landsat ETM+, Marplot, neo-cartographers, NGA, PGS

Posted by Mr Minton at 8/25/2007 08:12:00 AM | Permalink | Comments | links to this post

This is a preliminary map of the Orkney Islands. So far I've used MS Live Local imagery, Landsat ETM+ imagery, NGA Prototype Global Shoreline and World Vector Shoreline to complete this first look at the Orkneys. Landsat ETM+ is largely cloud-obscured and only workable for the southern portion. MS Live Local uses some proprietary imagery that acts alot like Landsat, but I'm not for sure. NGA PGS makes up the northern end of Scotland and WVS was used as a check against other vectors. The project started with a viewers request for bathymetric data. He got the data, but the shorelines were WVS (1:250,000). My mapping is better than WVS, but not EVS quality. It is good, but not great. More will be coming on the Orkney's.

Enjoy!

Labels: EVS, Global Mapper, island, Landsat, Marplot, MS Live Local, NGA, Orkneys, PGS, UK, WVS

Posted by Mr Minton at 6/08/2007 12:38:00 AM | Permalink | Comments | links to this post

Malta - DigitalGlobe Image from Google Earth (1-50,000)

The first part of this post is called "Mapping Malta". Not necessarily a big island, but the shoreline has been quite a challenge to digitize. I am about 3/4's finished with the first pass using DigitalGlobe's imagery from Google Earth as my base and the digitizing capabilities of GE to map the shoreline. In the process, I have mapped moles, groins, breakwaters, quays, piers, slipways, ramps and a few dry docks. Thanks to DigitalGlobe's fantastic imagery, my shoreline mapping challenge is greatly magnified. But that's okay. Once this shoreline is finished, it will be a WOW!

Malta - DigitalGlobe Image from Google Earth (1-6,250)

Malta - DigitalGlobe Image from Google Earth (1-3,125)

Malta - DigitalGlobe Image from Google Earth Detail Shoreline (1-2,500)

The three detailed shoreline images above are extracted from the southern side of Malta. The coast appears to be made up of a base rock layer that sea and air manages to split apart and gradually weather these rocks into pebbles after many, many years. The rocks on the last images are very large. The rectangular-shaped boulder is over 17-meters by 17-meters! If that rock were in the sea, alone, it would be an island. Based on my mapping criteria, it is part of the shoreline. How should I classify this portion of the Malta shoreline? Rocky? That doesn't do it justice. Big Rocky? Better. Lots of Really Big Rocks Scattered Within 100-Meters of the Unfragmented Shoreline? Now, I like that name. I will work on the nomenclature to transform "wordy" into a succinct EVS label.

Malta - Landsat Image N-33-35_2000 Detail Shoreline (1:2,500)

I couldn't leave this post without giving you Landsat ETM+ 2000's comparable image of the same shoreline area. Digital Globe's HiRes imagery is amazing, from a mapper's perspective. The most accurate global shoreline freely available, NGA's Prototype Global Shoreline (PGS), approximately tracks the Landsat ETM+ shorelines. It is an impressive effort that will be useful in a wide variety of projects. But when you compare it to Digital Globe's derived shorelines - Forget about it! As you would expect.

Enjoy!

Labels: DigitalGlobe, EVS, Global Shorelines, Landsat, Malta, NGA, PGS

Posted by Mr Minton at 5/23/2007 11:15:00 AM | Permalink | Comments | links to this post

Dar es Salaam - VMAP Layers (1:2,000,000)

I spent yesterday working with VMAP data. VMAP is the new Digital Chart of the World (DCW). It contains layers of information in vector format. The entire world is contained in 4 files V0eur, V0noa, V0sas and V0soa. Global Mapper brings in all or individual layers of information. I decided that I wanted my World Shoreline map to contain some of these layers. I reformatted and loaded: Global Bathymetry, Elevation lines (contours), airfields (almost 10,000) and metro area polygons that delineate populated areas.

Dar es Salaam - VMAP Layers (1-1,000,000)

This is the classic scale for the DCW, now called VMAP. This scale is able to accomodate a number of layers of information: World Vector Shoreline, airfields, metro areas, roads (green), railroads (red), name labels and water/land contours.

Dar es Salaam - VMAP Layers with Prototype Global Shoreline (1:125,000)

This is Dar es Salaam at 1-inch : 2-miles. More name labels show up at this level of zoom. The PGS consists of great detail and gaps due to reputed cloud cover. It makes me want to digitize a shoreline to EVS precision.

The additional layers of data are informative. They bring my World Shoreline map up to about 5-gigabytes in size with over 190-layers of unique information.

If you would like to know how I worked the VMAP data, drop me a line and I'll explain.

Now I remember what I wanted to tell you. This whole project began with a look at Europa Technologies' $5,595 one-year license Global Discovery, their premier digital world reference map. This mapping package consists of a number of DCW layers - free information that Europa Technologies repackage and sell. More power to them! I won't buy their mapping package, but if there are consumers that require it, sell it to them. Or those consumers can do like I did and make their own World Map. It takes time and very little money (Global Mapper @ $199). And it's easy!

Enjoy!

Labels: Europa Technologies, EVS, Global Discovery., PGS, VMAP, WVS

Posted by Mr Minton at 4/01/2007 04:07:00 PM | Permalink | Comments | links to this post

Comparison of Four Different World Vector Shorelines

If you look at most commercial digital atlases, their data is based on Operational Navigation Chart (ONC) data, which was originally compiled and maintained so that pilots could use these maps as reference tools when viewing features from 20,000 feet. They worked well for their intended purpose and were reasonably maintained. They were largely fashioned from pre-satellite data, much of it derived from aerial photography dating from WW-II through pre-Vietnam. Any cartographer worth their mapping "salt" will tell you that much of the data is, at best, only moderately accurate. ONC shorelines are close to actual shorelines, but not very close. The ONC shorelines were never intended to mirror the actual shoreline. I mean the scale is 16-miles to the inch!

So from this data source a product called Digital Chart of the World (DCW) was produced. Every atlas company producing small scale mapping uses some or all of the DCW data. Since my area of focus is global shorelines, I can state with certainty that if DCW shorelines are used in a map product, those shorelines are only moderately accurate. But for many map users, moderately accurate works. My Michelin maps got me from city to city and allowed me to navigate with a solid degree of certainty.

World Vector Shorelines, at a scale of 1:250,000 (4-miles to the inch), were fashioned from Tactical Pilotage Charts and other similar sources. Their shorelines are more accurate than DCW/ONC shorelines. However, they also more often miss the shoreline than track it.

Landsat ETM+ 2000 Coverage

Now we get into the realm of space-based mapping. With a virtually cloud-free image of earth, Landsat gives us a georectified imagery base from which extremely precise shorelines can be extracted. And it was done! The Prototype Global Shoreline, using a carefully researched, crafted and tested digitizing algorithm created a global shoreline that can be used at between 1:125,000 to 1:62,500 scales. If DCW/ONC shorelines are moderately accurate and World Vector Shorelines are extremely accurate, Prototype Global Shorelines track the shoreline with a high degree of reliability. That is until one zooms in and studies the PGS vectors as they relate to the Landsat image from which they were derived.

EVS and PGS Vectors

Upon closer inspection, they do not track the shoreline very often. They are typically anywhere from 15-meters to 200-meters off-center. Their vectors have a habit of "stuttering" unneccessarily as can be seen in the above image. They also do not account for large sections of shoreline that is water innundated, per NIMA's request.

EVS Would Digitize the Water Inundated Areas

EVS vs PGS Precision

Finally I get to my product, Enhanced Vector Shorelines (EVS). All of my shorelines are done by hand. The vector plotting errors inherent with this methodology are minimal compared to PGS errors. EVS provides the most accuracte shorelines extracted from Landsat ETM+ 2000 mosaics. Unfortunately, that only includes the coast of Morocco, Taiwain, Madagascar, Baja California, Caspian Sea and many islands scattered throughout the world.

Now to repeat the question I posed at the beginning of this post, "Who cares?" Should the world shorelines be mapped at EVS precision? Is the PGS, which is pretty good, sufficient for most projects? Is pretty good, good enough?

I think that a world shoreline at EVS precision would be extremely useful. I believe, that properly constructed, it would become the base mapping data for military, government and private sector projects for years to come.

I know I have a large number of GIS savy readers. What do you think? Should the world be mapped at EVS precision? And if so, why? Feedback would be appreciated.

Enjoy!

Labels: DCW, digitize, EVS, Global Shorelines, Landsat, ONC, PGS, SRTM, WVS

Posted by Mr Minton at 3/11/2007 04:50:00 PM | Permalink | Comments | links to this post

Pepel Island - PGS Shorelines 1:30,000)

The shorelines in the above image are PGS precision shorelines. They are typically excellent in tracing Landsat imagery shorelines. That is until you have shorelines that contain wetlands. According to individuals at MDA that worked on the project, wetlands were to be excluded as shoreline. I can appreciate the logic behind the decision to exclude these water-logged areas. However, there is a problem.

Pepel Island - Landsat Image N-28-05_2000 (1-30,000)

This is the Landsat imagery that Prototype Global Shoreline (PGS) was derived from.  You can see that PGS tracks only the non-wetland shoreline.  Enhanced Vector Shoreline (EVS), my product, not only tracks the island shorelines, but includes the wetlands.

Pepel Island - Marplot Map (1-30,000)

This is my solution for mapping Pepel Island.  The entire island is mapped to include both land and wetlands.  The two warfs at the sourthern end of the island was derived from Google Earth Imagery.  If I'm a dude in a boat heading into Pepel Island port, I definitely would want to see the wetlands.

Pepel Island - Landsat Image and EVS Mapping

This side-by-side shows the Landsat image used to create both PGS (red lines) and EVS (the completed map).  EVS most nearly depicts the base imagery.

Enjoy! 

Labels: EVS, island, Landsat, map, Marplot, PGS

Posted by Mr Minton at 11/12/2006 12:19:00 AM | Permalink | Comments | links to this post

Comparison of EVS, PGS, WVS and SRTM (1)Shorelines

A picture is worth a thousand words. The above image contains a small segment of Eiao Island FP. The scale is 1:12,000. What are visually displayed are the shorelines of various global shorelines (PGS, WVS and SRTM (1)) as they compare to my own EVS shoreline. They all fall short of depicting the actual shoreline and that is to be expected. A shoreline is a dynamic geologic entity. It is constantly in motion. There will never be a penultimate map (a static entity) that depicts a global (dynamic) shoreline. With that said, which of the four shorelines works best for your needs. All four shorelines have their uses.

The World Vector Shoreline, "old reliable", works for most mapping needs at a scale of 1:250000 or above. When you work above 1:2000000, the map becomes too busy. But between those two scales, WVS works fine.

Prototype Global Shoreline, when completed, will be "THE" map file for work from 1:62,500 up to 1:1,000,000 scale. It will provide details that will make WVS look like a clunky jalopy compared to a sleek race car. PGS will impact every country in the world. It will require a country-by-country acceptance. It will cause nations to redefine maritime limits and that is not an easy notion for nations to agree upon. Nations will gain or lose valuable real estate because maritime base lines and economic zones will be shifted because PGS shorelines will shift. If the shift results in a lose of real estate, my suspicion is the losing nation will not buy the new boundary. What an exciting proposition for the global community. Much of the above comes from a source within NGA that has the enviable task of completing this monumental map file.

SRTM (1)(SWBD) shoreline was added to show how far this vector mapping program has advanced. It is a solid vector map file that improves mightily upon WVS, but it does not accomplish the precision that is found in PGS. However, again from my source in NGA, there is a more precise global data source within SRTM. It is SRTM Level-2. It is available for the US today, but NGA has access to global coverage at the level-2 precision. I am purposely vague on SRTM Level-2 as I have not done much work with it, but I am certain it will perhaps rival PGS when released to the eager public.

Finally, my own Enhanced Vector Shorelines effort and how it compares with the other three shorelines. I like to think that EVS is the most accurate of the four, but that isn't quite so. The three are global in scope and mine covers islands and some shorelines. The places I have mapped at EVS resolution are more exact in depicting the actual shoreline than any of the other three shoreline sources. My ultimate goals with EVS is to see all shorelines mapped at that level of precision, to develop a consistent decision-making process as to map or not to map certain features and to bump this effort up into a legitimate concerted, well-funded effort that will result in a timely completion of the project within 2 to 3 years.

EVS, in my opinion, is the ultimate map file that can be extracted from Landsat-7 ETM+ 2000 imagery. When 1-meter resolution imagery is released and the coverage is global, the new and improved ultimate map could be created, EEVS (Extremely Enhanced Vector Shorelines). Until then, I will keep myself busy working on my labors of love – creating shoreline maps of islands.

Enjoy!

Labels: EVS, Landsat, map, PGS, SRTM, WVS

Posted by Mr Minton at 7/05/2006 11:35:00 AM | Permalink | Comments | links to this post

PGS vs EVS - Detail (Motu Ko, Pukapuka Atoll CW)

So, I decided to test the PGS vectors (red lines) against Pukapuka Atoll, Cook Islands digitized at EVS quality. There is a cloud issue on one of the motus. Fortunately, the clouds were not a problem with Motu Ko. I tried to digitize by moving vectors to trace the visible shoreline. I moved more vectors then I didn't. At least 70%+ of the vectors had to be moved to track the visible shoreline. The moves were slight, less than a few millimeters which equals 10's of meters on the ground. Some of the vectors traced reefs that were classified as shorelines (northern part of Motu Ko).

The PGS vectors are close, but no cigar. For EVS quality the PGS vectors require a great deal of adjusting. To produce EVS islands, I need to digitize off of the Landsat image. So, I will keep doing what I do best, making EVS shorelines.

Enjoy!

Labels: Cook Islands, EVS, island, Landsat, map, Marplot, PGS

Posted by Mr Minton at 6/14/2006 11:18:00 PM | Permalink | Comments | links to this post