Water, water everywhere!

Last semester I took a Water Resource Economics class and did a final project that looked at the role of water meters in reducing water use. My inspiration for this project was hearing that the Republic of Ireland is just now starting to install water meters for the purpose of charging people based on their volumetric consumption. Now, there are many wealthy countries that only charge flat fees for water, that is, not based on volume used, but Ireland is one of the few places that did not charge for water, period. Water in Ireland has been free, a service provided to its citizens, but relatedly, Ireland also has one of the highest water use rates in Europe.

So I decided to look at what happens when consumers pay for the amount of water they actually use. I found 24 case studies wherein water use before and after water meter installation was reported, from countries on 5 continents, and I'd like to share some of the maps that I made to correspond with that data. (If you're interested in particular cases, I'd be glad to share my paper and/or sources.)

First are the maps showing daily per capita use and daily household use, depending on what the paper provided measurements for.

And the reductions in water use after meters were installed:

 

Interestingly, the greatest reductions occurred in Yerevan, Armenia, though there are many reasons behind the different consumption and reduction rates. Other noteworthy sites included Hamburg, Germany; several sites in Colombia; British Columbia; California; and Isle of Wight, UK.

Where the Donors At?

Though I haven't posted anything for a while, it doesn't mean I've been completely absent on the map-front. Au contraire, I've given myself a few projects to do at work showing where the donors, donations, and lecture attendees come from, and today I'll share a map of donors for [place of work omitted].

I chose to show this map by zip code for two reasons. First, I have zip codes for all the donors. Second, there are zip code map shapefiles easily obtainable online. Future variants of this map will be more neighborhood specific, though I'll probably just translate all the addresses into GPS positions and map those, rather than find a way of using street maps or something crazy like that.

You would think, for those of you familiar with Chattanooga, that you should be able to identify wealthier areas of town based on the number of donors for that zip code. This map is sort of surprising in that regard, as the heaviest donor base is actually in Marion County, which would not have been my first guess.

More of the Same

I have version 2 of the Hamilton County, Tennessee, watershed map ready to go today. Prepared again using QuantumGIS, with Elevation data downloaded from the National Map and transformed into a lovely elevation/terrain model.

I had to download the data because, even though there is a WMS service for USGS data that lets you pull it directly into Quantum or your GIS software of choice, it limits the size of the map you can "print." What that meant for me was that when I tried to create an image of the map, it was considered too large, and the images served up by the WMS (Web Map Service.. I have to remind myself of what these acronyms stand for so that someday I'll actually remember them) didn't show up in my map image.

I like this map okay but I like it less when I go back and look at the original HCWQ map that motivated this project in the first place. Oh well, it's a learning project... and I can't do it exactly the same way as Adam did because that would be unoriginal!

A Quantum Leap

I just had QuantumGIS installed on my work computer, which is exciting for multiple reasons. First off, I can give myself projects to work on when there's a lull. Second, I'm all for supporting open-source software use when possible. Third, it explicitly gives the finger to ESRI, who charges an arm and a leg for ArcGIS licenses as well as fees for admission to their user conference (even for non-profits).

The first time I downloaded Quantum was before I had ever used Arc, and I was naturally bewildered about how to do anything with it. I had downloaded some gigantic files and an outdated tutorial that required the use of GRASS alongside Quantum and was doing my best to come to some sort of understanding, but really to no avail. After working with Arc and coming back to Quantum, I must say I'm a fan of the interface now. It seems far less cluttered but still presumably with all the functionality (however that works). I haven't done any analysis with it yet, just trying to get my feet wet by making some maps.

Here's my first offering exclusively using Quantum: a labeled map of the watersheds (HUC 12 level) in Hamilton County, Tennessee. There's all kinds of data I'd like to put on the map, so it's hard to limit it for clarity's sake, but I'm still looking at making some future changes, like adding a shaded relief basemap from USGS and coloring the individual watersheds different colors instead of having chunky borders around each one.

Land Use, ArcGIS, and You!

.. Well, maybe not You, but at least Me.

There's some pretty cool data hosted at the McGill Geology department, and I took a peek at the first set on Global Cropland and Pasture data from 1700 to 2007 (whoa!).

It's in NetCDF format (yeah, I hadn't heard of it either), which is pretty cool because it holds a bunch of time slices of data, so in just one file you have ALL those years from 1700 through modern-day. Plus it's a relatively small file and really easy to flip through those years in Arc. I'm still trying to figure out how to get ArcMap to play nicely with it in raster format, as the Raster Calculator seems displeased trying to operate on the data.

So here's what I came up with: a combined map of land use for crops and pastures in the year 2007. Pastures are blue and crops are green, and overlapping areas are various shades of turquoise, going all the way to dark dark green-blue. (Very pale greens and very pale blues means there's not that much of either, but there's even less of whichever color isn't showing.) Pretty intuitive with the different areas (northern parts of Africa/Canada/Russia aren't used for much of either, midwestern US is for both, and so forth).

In other news, some of the land use analysis and mapping I did for the Land of Sky Regional Council in North Carolina is going to be discussed next Tuesday in the US Department of Transportation Eco-Logical Webinar on Green Infrastructure and Transportation Planning! Klasse!

Fun with GIS

I'm trying to beef up my GIS portfolio in hopes of finding an awesome full-time position somewhere (anyone?). And naturally I'm doing this by having a little fun, that is, by doing some small and easy yet interesting projects. I found a really nice collection of links to data repositories (located here), with collections ranging from precipitation and temperature maps to land cover and health. It's just a matter of figuring out what to do with it all.

So I started out with an observational exercise: looking at the difference in precipitation levels across the US over the past half century. For being so simple, it's actually pretty telling stuff, and it at least seems to match with observations people have made in the Southeaster-ly parts of the country, namely that many bodies of water were once regularly higher than they are today after our near-annual droughts.

Caveats: the precipitation data from the PRISM project is part of a research project that models precipitation for entire areas based on a small number of actual measurements; while that idea makes me put on my skeptical face, PRISM is apparently considered some of the best data out there. However, the data is not awesomely labeled; I believe the numbers in all the data files were in millimeters but wouldn't stake my life on it based on the lack of metadata. Furthermore (and here's the iffiest part).. the data files are averaged over a period of years. There's a file of averages from 2005-2009, 2000-2004, 1995-1999, 1990-1994, and.. 1961-1990. Yeah.

So the difference displayed by this map is between average precipitation levels during 2005-2009 and average precipitation levels during 1961-1990. No fancy intermediate averages or anything like that, just comparing the average levels during those two sets of time. Open it up and have a look!

I'll likely be posting these sorts of things on a regular basis, so check them out or contribute your own ideas, data, and maps!