Thoughts - Green Dots on a Deep Black Sea

Cook Islands - Landsat ETM+ Coverage

I have been very busy digitizing islands. In fact, I have just completed digitizing over 3,000 large and small islands making up the western portion of the Solomon Islands. I'm currently digitizing island shorelines in Papua New Guinea. All of this work is being done at EVS precision - using Landsat ETM+ base imagery and digitizing shorelines at a scale between 1:25,000 up to 1:50,000. Some of you have reminded me that because Landsat ETM+ is typically 14.5-meter resolution that digitizing at the scale I work at is overkill. After spending close to 3-years working with 14.5-meter resolution imagery, the maps I am able to produce work quite well. I wouldn't want to sail into a lagoon using only ETM+ imagery, but I could do excellent advance planning and rough navigation using maps derived from this imagery. I know my maps are often far superior in shoreline validity when compared to some of the very old nautical charts still in use today.

Anyway, this is supposed to be about green dots on a deep black sea. In between work on the Solomons and Papua New Guinea, I completed EVS precision mapping of the Cook Islands. The first image in this post struck me. It took seven Landsat ETM+ mosaics to cover the Cook Islands. If you study the above image at full size, almost all of these very small islands are impossible to see. Yet, they enable countries, like New Zealand, to claims (EEZ) enormous chunks of liquid real estate. These very small islands play a significant role in the distribution of world resources. Interesting.

Cook Islands - Marplot Map with 25-km Ocean Masks

The Marplot map of the Cook Islands shows their distribution more clearly. They are scattered about within a peculiarly shaped border, that is if one finds rectangles peculiar. Actually, their EEZs would spill outside of their international border. I'm not sure if the border serves to limit their 200-NM EEZs. If that is the case, too bad for New Zealand and the Cook islanders.

Penrhyn Atoll CW - Landsat S-04-05_2000 Image (1:125,000)

Just to confirm that there are islands located in this broad expanse of black ocean, here is one of them, Penrhyn Atoll. It has a classic atoll shape with a lagoon full of near surface coral heads. I have not researched Penrhyn as I was in a digitizing mode when I worked on it, not in an investigative one.

Penrhyn Atoll CW - EVS Precison Island Polygon Map (1:125,000)

This is my EVS precision map of Penrhyn Atoll. It is a solid shoreline depiction. My customary mapping effort of an atoll would include a number of additional layers of information - reefs, beaches, vegetation. As my client requires only shorelines, I digitizied and mapped the island's above-water shoreline and portion of it's awash reef. Again, it is a solid shoreline mapping effort and one that I will revisit to complete the other layers of information I would typically map.

I am still here, just not able to post as often as I would like. When I finish my paying mapping project (Whoopee!), I will share many, many, many more islands. Until then...

Enjoy!

Banaba Island - Freedom For The Banabans

View Larger Map

Banaba Island

Set These People Free! They got shoved off of their island back around the turn of last century, because they wanted the British, New Zealand and Australian interests to stop extracting phosphate, derived from bird guano, from their island. They got the old colonial wham-doozie! A contract that they didn't understand, a deal that gave away their island resource for next to nothing, a forced evacuation and after their island had been brutally plundered and not repaired, they were allowed to return. Just recently the courts awarded them $10 million for the past injustices - a mere pitance for all that they have been subjected to. There are a couple of web sites that tell their story: Abara Banaba, Wikipedia article and streaming video of a fly over of the island by Greenpeace.

The map was digitized within Google Earth using DigitalGlobe imagery and ultimately constructed in Marplot. I spent the last six to eight hours working on this relatively small island, but I learned alot while working on it. It is my first island I attempted to digitize all visible buildings. I think I did an excellent job.

What got me started on this particular island was the British Hydrographic Office Catalog of Nautical Charts. They publish a 1:50,000 scale chart (HO 979) of the island proper. This seems to me to be too large scaled to be of real use. In addition, they have a chart showing harbor details at 1:12,000 scale. I personally like my 1:3,125 view of the harbor. Also their chart was originally constructed in 1884 and updated in 1964. My data is from DigitalGlobe imagery gathered in the last couple of years. However, their charts are designed to be used by boaters, mine are not. I make maps, very nice maps, but just maps.

I couldn't leave well enough alone. I added a fourth image, a Marplot map at 1:20,000. The colors have been changed, a text box with appropriate credits is added, a nifty North Arrow and contours at 5-m intervals. I like the new look. One can soften colors too much. The colors used on this map are a bit bolder to draw attention to the features they depict.

Enjoy!

Thoughts - Easter Island Shoreline 12,615 Vectors Later

Easter Island - EEVS Map (1:100,000)

This is a continuation of my previous post. EEVS quality mapping requires lots of vectors. As the title states, at this preliminary stage, 12,615 vectors were digitized by me to create this shoreline. I worked in Google Earth using the path tool. The first pass was relatively fast taking about 6-hours. The final pass, the fine tune, took another 3-hours. I will have to work on this shoreline in Global Mapper to finish it. Two parts of the island, the north and western edges will be digitized using Landsat as they are not included in the Google Earth map. I have loads of work to do on this map digitizing layers of information.

My point is that DG imagery is great, but to execute a quality shoreline, one must do lots of digitizing.

Enjoy!

Thoughts - How I Do What I Do When I Do What I Do

I receive numerous requests for special project maps. Most of them I do, a very few I don't. Lately, representatives from non-profit groups have solicited my map-making talents to construct maps and secure imagery for their special projects. Their budgets are $0.00 for mapping. If the group's work furthers the quality of life of both human and non-human (flora and fauna) inhabitants of these fragile places, I try to help. They begin by asking if I use ArcGIS. I tell them I too operate on a $0.00 budget and it is too expensive. They then ask "how I do what I do when I do what I do" in making my maps. So I explain my map making processes.

When I first began making maps, I used Landsat ETM+ mosaics exclusively. They enabled me to create detailed shorelines, where none had existed before. I made lots of maps using this database of imagery. Then came Google Earth and then came DigitalGlobe hi-res imagery. Many of my current projects use this imagery for a base. It is how I use it that this post deals with.

(The following is the essence of an e-mail I sent to one of my non-profit contacts. It explains "how I do what I do when I do what I do" in making maps using DigitalGlobe hi-res imagery in Google Earth.)

Iwo To - DigitalGlobe Image From Google Earth Detail Suribachiyama (1-12,500)

Capturing and georeferencing DG imagery can be done. Depending on the level of detail required, the number of images captured and georeferenced can be extensive. For example, a relatively small island I recently digitized required over 30 DG images taken from GE. One must be patient and practiced at creating mosaics from these separate images. I ensure all images are at the same scale on GE, systematically select and save them to my desktop,

Creating a Photo Mosaic From Shuttle Images using ImageForge

and then create a mosaic of images in a draw program (ImageForge).

DG Image Georeferencing in Global Mapper

I take these mosaics and georeference them using my preliminary mapping program (Global Mapper), where I do the bulk of my digitizing.

Digitizing from Landsat ETM+ Image using Global Mapper

As I complete digitizing areas of interest from these base images, I import the vector files into my primary mapping program (Marplot),

Making Maps using Marplot

where I create my maps.

******************************

Another process, much simpler, but still challenging,

Digitizing using Google Earth tools

is to digitize island shorelines and information layers using GE's path and polygon capabilities. Both of these digitizing tools are extremely powerful and flexible for the type of mapping I require. After digitizing a number of layers of information, I process the KML file into a simple ASCII vector file and import them into my preliminary and primary mapping programs where I create maps from these GE-derived vectors.

******************************

Both methods have taken me over two years to master, but the end-product is cool!

Banaba Island - Marplot Map with Contours (1-20,000)

I am an accomplished map-maker when it comes to using DG/GE imagery. I currently work with two university-affiliated non-profit organizations supplying maps and images for their island projects. My maps have been viewed by over 25,000 visitors this year alone. Most are arm-chair travelers. However, a substantial group represents a "Who's Who" of the GIS community and the business community using GIS in their work.

My map-making processes can be taught to the right students. Given time (months), genuine enthusiasm (you have to love the map-making process and the final product), an aptitude for monotonous work (digitizing and tweaking the project) and the ability to constantly improve your map-making skills, I could teach someone, with these attributes, how to make maps of islands. Or, I could make your island maps myself. Whatever.

(And that is what I said about me and my maps. Interesting, huh?)

Enjoy!