EVS-Islands

Pukapuka Atoll - Landsat Image (S-03-10_2000)

Pukapuka Atoll - Map

Pukapuka Atoll (10˚55'S., 165˚50'W.) lie about 42 miles NNW of Nassau Island. They consist of Pukapuka, the farthest N; Motu Ko, 3 miles S of Pukapuka; Motu Katava, 1.5 miles NW of Motu Ko; and several smaller islets and sand cays. A sand cay lies at the W end of a reef, 3 miles W of Motu Katava. The islands, islets, and sand cays lie on or within the barrier reef, which surrounds a shallow lagoon.

The barrier reef is unbroken and there is no access to the lagoon for other than small boats. On its W side the barrier reef is always awash; its S side is awash at LW, and its E side is partly awash and partly dry sand. The sea breaks over the sunken reef which extends W from Motu Katava.

Pukapuka is densely wooded with coconut and other trees; some reach a height of 24m. The island is bordered on its N and W sides by wide, flat reefs, and the surf beats heavily on the narrower reef on the weather side. Three villages are situated on the S side of the island near the lagoon; a large church and a school are situated in one of the villages. The Resident Agent lives on Pukapuka.

Motu Ko, at the SE end of barrier reef, is flat and sparsely wooded; the tops of the trees rise to a height of 30m. An aviation runway is situated on the S side of Motu Ko.

Motu Katava is densely wooded in its W part. The trees on the W side rise to a height of 38m, and the coconut palms on the E side rise to 24m.

The current sets strongly onto the E side of the Danger Islands.

The current across the reef extending W from the islands is dangerous. It runs S during the ebb and N during the flood, attaining a rate of 3 to 5 knots. This reef is said to break in deep water even on a calm day.(SD Pub-126)

I have spent the last 3 evenings working on Pukapuka Atoll. The project began as a test of the Protype Global Shoreline map file for Pukapuka Atoll. The PGS files, although a tremendous improvement over WVS, are too coarse for a well-done EVS island.

Pukapuka Atoll - ISS Image ISS002-E-9915 Realigned

The EVS map for Pukapuka Atoll is a very well done product. I love the work that I was able to do on the various reefs that are part of this atoll. I did use a shuttle image (see above) to resolve shoreline that was cloud-obscured. This is one of my best maps to date. Air Rarotonga has information on Pukapuka Atoll.

Enjoy!

Labels: atoll, Cook Islands, EVS, island, Landsat, map, Marplot, Pacific

Posted by Mr Minton at 6/17/2006 03:22:00 PM | 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

PGS (Red) vs EVS (Orange) - Landsat Image Detail (Enderbury Island Kr)

The above image is from a Landsat ETM+ 2000 image (S-02-00_2000) that was used in the development of the NGA Prototype Global Shoreline (PGS) (see previous post for link). Since I previously digitized this island, I used it to compare my vectors (EVS) to the newly released NGA vectors (PGS). The image retains the pixel shapes (little squares) that EarthSat used in creating the PGS. The PGS is the "red" line. My EVS is the "orange" line. This island is a very simple shoreline that was easy to digitize. It appears that the PGS vectors are nearly complete for the island. They did not create vectors for the southern end of the island that had minor cloud obscuration. I did create vectors to cover that area using other sources, WVS, internet maps of the island, etc.

PGS vs EVS - Detail (Enderbury Island KR) Anti-Alias Pixel Effect

The above image is identical to the previous image, but with the "anti-alias pixel" effect activated. One can easily see the smoothing of the pixels. This is how I prep images prior to digitizing. My motivation is that shorelines are not pixel shaped. The look more like the smoothed shorelines. So, I elect to smooth shorelines prior to digitizing them.

Either way, the PGS vectors are impressive. They track the EVS vectors closely. There are some discrepencies, but they are minor. Last night, I loaded all 28 PGS vector files into Marplot, my mapping software. It took about 3-hours to download, prep, import and define layer attributes. Now, I am able to work with these vectors prior to digitizing from Landsat ETM+ imagery. Although the PGS vectors are stunning when one compares them to WVS, they do not resolve shorelines as precisely as I am able. Consequently, EVS lives! I will benefit from the PGS vectors, but my mapping will continue.

If any of my readers know the parties involved in developing the NGA PGS vectors, I would welcome the opportunity to use my EVS efforts to QC their efforts. Drop me a line with a contact or have them get in touch with me. If any of you would like pointers on how to process the NGA PCS files (28 total) drop me a line. I'll be glad to share what I have learned in working with them

Posted by Mr Minton at 6/14/2006 11:48:00 AM | Permalink | Comments | links to this post

PGS Overlay on Suwarrow Atoll

One of the readers of EVS-Islands suggested that I look into an improved WVS (World Vector Shorelines) file. I am aware of something called WVS Plus, and as yet have not bothered to download it. This was supposed to be much better. Rumor is, this is a world shoreline vector file(s) extracted from Landsat-7 imagery. I thought, "Hell, that is what I am trying to do? This file is going to make my efforts of no consequence.

Well, the file exists! And it is available at the Prototype Global Shoreline Data (PGS) page. The files are easily downloaded. They are in .shp file format. They are zipped and ready to load into your mapping software. Global Mapper loads them and they can remain zipped. They are very large files unzipped (average between 10 to 100-mb per file with a total of 28 files).

After a preliminary look, I can confidently state that PGS is a major improvement over WVS. At 1:250,000, PGS shows a different shoreline than WVS shorelines. At larger scales (1:25,000) they show a shoreline that is very close to Landsat-7 imagery, whereas WVS is clunky and coarse.

To compensate for cloud cover, the method EarthSat employed was to leave the area blank (no vectors). I am sure NGA and EarthSat already have a solution in the works to approximate cloud covered shorelines.

Now my question, "Is EVS necessary? Is this shoreline what EVS strives to be? No! I still have a mission. Their shorelines are just that - shorelines. They used a land masking method that appears to not include human eyes. There are irregularities, not as irregular as WVS, but misalignments. My EVS efforts of islands and shorelines correct most of these irregularities.

The final process EarthSat and NGA should consider is "human eyes" to confirm what software solutions have suggested. If that is to be done, great! If not, then my quest remains - Enhanced Vector Shorelines for the World!

Posted by Mr Minton at 6/13/2006 08:29:00 AM | Permalink | Comments | links to this post

Caldey Island - Landsat Image (N-30-50_2000)

Caldey Island - Map

Caldey Island lies off of the southwestern coast of Wales. Small in size, but rich in history. The island is about 3-miles from the city of Tenby. It has a small population besides the 30 or so monks that live on the island. I came across a few interesting links that will give you a great deal of information. The first site is the official Caldey Island website. It gives you a very complete picture of this interesting island. The next link is about Welsh holidays and Caldey Island has it's own web page. Their are other links, but I'm sure the two I've given you will be sufficient to get you started.

Caldey Island (51°38'N., 4°41'W.), 56m high, lies 1 mile SE of Giltar Point. It is mostly bounded by cliffs of moderate height, the highest being on the S and NE sides. A prominent monastery, with a round tower, stands near the center of the island. A main light is shown from a conspicuous tower, 16m high, standing near the SE end of the island.

Offing Patches, consisting of several small areas of foul ground, lies within 1 mile of the S side of the island and has a least depth of 6.8m. Drift Rock, with a least depth of 9.1m, lies about 1.3 miles SE of the light and the tidal current, setting W, occasionally causes a considerable sea over this rock.

A ledge of rocky shoals, with a least depth of 3.7m, fronts the E side of the island. Spaniel Shoal, the E and outer of these dangers, lies about 0.5 mile E of the E end of the island and is marked by a buoy.

Eel Spit, a rocky ridge, extends about 0.3 mile N from the N extremity of the island and has a least depth of 1.8m. Highcliff Bank, with a least depth of 1.9m, extends 0.8 mile N from the N side of the island and is marked by a buoy.

St. Margaret’s Island, bounded by cliffs, lies 0.2 mile W of the NW end of Caldey Island to which it is connected by a drying reef.

(SD Pub-142)

Posted by Mr Minton at 6/11/2006 01:34:00 PM | Permalink | Comments | links to this post