Biggest Map, Smallest Map

The week before last I worked on two maps for two different non-profit organizations focused on wilderness protection: The Wilderness Society and the Washington Wilderness Coalition. TWS needed a thumbnail of a map for a newsletter; WWC needed a giant backdrop map for a display they are putting together with a graphic designer. As it turns out, the TWS map is the smallest map I have ever made (2.5" x 3") and the WWC map is the largest I have ever made (60" x 45").

Here is an image to give you a sense of the difference.

Biggest on left, smallest on right. Click on thumbnail to embiggen.

Excellent article about GIS in NY Times

The NY Times published this very intriguing piece about GIS earlier this month. It focuses primarily upon urban design and architectural applications, but it's nice to see the technology getting such prevalent exposure.

WA URISA Conference

Last week I attended the Washington URISA conference for the first time. Overall, I found the conference to be informative and useful and was impressed by the variety of sessions, particularly the volume and quality of open source presentations. I presented with David Howes and Gretchen Peterson on the Lone GIS Professional, a group that David started last year. It was an interesting session, in that about 1/2 the time was allotted to group discussion and David recorded audience input in real-time. The objective of the talk was to spread the word about the Lone GIS Pro group, get more people involved, and perhaps more importantly, find out what would be helpful to Lone GIS people.

It was surprising to me how many attendees of the conference are Lone GISers--either solo practitioners or the only GIS person at their business/agency/non-profit. The need for (and value of) a group like the Lone GIS Pro was reinforced in many of the sessions, based on their heavy emphasis on enterprise GIS systems with massive budgets, staff, and procedural requirements. It really is a differnet world when you're working on your own, and I'm grateful to have such an active, energetic and helpful group to turn to when I have a question or need some backup.

If you are a Lone GIS professional and are interested in joining the group, please email me--the group is invitation-only.

From Images to Data--Image Segmentation

Last year I searched for freely available image segmentation software as an alternative to eCognition. My original search was related to a project that I decided not to pursue; however, this past fall and winter I worked on a project that required image segmentation . So, after doing additional research, I was able to determine that Brazil's National Institute for Space Research (INPE) produced a freeware program called SPRING that does image segmentation. The USFWS worked with INPE to produce an English translation of the software, along with a manual available here. The software itself can be downloaded here.

After working with this software for the better part of a day, I was able to get fairly decent segments out of my imagery, but the classification was not very good. I also found the software difficult and counter-intuitive to use, and although it is ‘translated’ a lot of the menus don’t make much sense.

So I started searching again and found Berkeley Image Segmentation which is great. It is commercial software, but I was able to get the beta release for $900. The creator of the software, James Scarborough, was surprisingly accessible and extremely helpful anytime I had a question about what the software was doing. He has already integrated several of my suggestions into the most recent version, and I’m sure he’ll continue to solicit and incorporate user suggestions. The software is much, much easier to use than SPRING, and it has several classification options built in (K-means, K-nearest neighbor, and neural networks) and while they are better than SPRING they still weren’t quite good enough. James suggests in the ImageSeg manual using WEKA for segmentation classification. This is open-source data mining software, and I used the expectation maximization clustering algorithm with great success, but there are many others available.

I have sample output on the CORE GIS portfolio page, along with a brief description of one of the projects for which I’ve used it (third one from the top).

Park(ing) Day in Seattle

Park(ing)Day took place on Friday, September 19, and was a "one-day, global event centered in San Francisco where artists, activists, and citizens collaborate to temporarily transform metered parking spots into “PARK(ing)” spaces: temporary public parks." The Trust for Public Land hired CORE GIS to create a map of the park locations in Seattle, to hand out to curious passers-by and help people find the various parks.

This photo of one of the parks was taken by Jeanine Anderson.

Sustainable Ballard: Local Food Production and Population

CORE GIS is located right on the edge of the Ballard neighborhood in Seattle. The non-profit Sustainable Ballard has been active in numerous sustainability and environmental issues, and each year at this time they ramp up their "Eat Local" campaign by promoting the 100 mile diet.

This seems like a great idea, for a number of reasons--it supports local farmers and local economies, reduces fossil fuel consumption, builds community, and encourages all of us to think about where and how our food is grown.

I decided to take this idea to its logical extreme--what if everyone in Seattle decided to follow the 100 mile diet? Do we have enough agricultural land to support that many people?

I started with 2000 U.S. Census data, specifically the SF1 100% population count at the block group. Then I used 2001 NLCD land cover data and extracted the two agricultural classes--cultivated crops and hay/pasture. I arbitrarily chose a spot in the middle of Seattle, created 50 mile and 100 mile buffers, then counted up the number of people and the acres of agricultural land within the two distance bands.

As it turns out there is far less acreage in cultivated crops than in hay/pasture, and within 50 miles, each acre of cultivated crops would need to feed 172 people! If we assume that all of the hay/pasture can be converted into cultivated crops, the number of people supported by one acre drops to 22. Moving out to 100 miles improves the situation, with just under 31 people per acre of cultivated crops and just under 7 per acre for all agricultural land. (The cartographic and tabular results are shown in the thumbnail image to the right; if you'd like to download a screen-resolution PDF of the 30"x36" poster, just click here. Please contact me if you'd like a printed version).

However, according to one study, a meat-based diet requires 9 acres per person! A diet that is primarily plant-based (with some milk, cheese, and eggs) requires 3/4 of an acre. Clearly, we still have far too little agricultural land within our 100 mile radius to feed our entire population.

This suggests to me that our region has a lot of thinking to do about our food security and food sovereignty. What happens if there are shortages of fossil fuels, in particular diesel, for transporting food across the vast distances it currently travels? Are there areas within our region that have prime agricultural soils that are not being cultivated? What policies can/should be put in place to ensure that we have the capacity to feed a much larger proportion of our population from local farms? How should our food distribution network change to make it easier and more efficient for local farmers to get their produce to market?

Stemilt Partnership

For the past several months I've been working with the Trust for Public Land on the Stemilt-Squilchuck Community Vision. Our client is Chelan County, and we've been working closely with a large citizen group known as the Stemilt Partnership. This photo was taken just after our most recent Partnership meeting last Wednesday at the Malaga Fire Hall. The project started in response to a proposed land exchange in which the Washington Department of Natural Resources was planning to divest itself of four sections in the upper part of the basin. As it happens, these four sections form the central core of many of the land use and land management activities that affect the entire watershed. The Partnership was formed to help the County determine what should happen with the land in light of all of the stakeholder groups that are using these public lands.

This project has been particularly interesting for me to work on, because it contains all of the elements that I really love about my job--interesting data and modeling challenges, lots of cartography, the opportunity to meet with and learn from a wide range of citizens and agency people (farmers, real estate developers, wildlife biologists, policy wonks, etc), public speaking and lots of travel to the watershed. The timeline is fairly compact for a project of this size and complexity, but I think the fact that so many community members are heavily engaged will help us produce an excellent result.