Wednesday, July 29, 2009

Remote Sensing Module 5:




I sometimes have a hard time differentiating one land type from another. In this case grass/agriculture land covers gave me a lot of trouble. In the Unsupervised map I took that cluster of fairly regular shapes in the north west portion to be an area of High Urban Density. I didn't even include a land cover category for agriculture because I didn't think any sizable farm land existed on the image. Part of my issue was not paying strict attention to the area I was viewing and making assumptions about the image based on past experience. I thought the main body of water running through the center of the map was Escambia Bay, hence I assumed that area in the North West portion to be Pensacola. Nope. That central water body in the image is Perdido Bay making that North West portion of the image a sparsely developed section of Alabama. Upon further review, I took those North West sections to be tracts of farm land. I added the Agriculture land cover category for the Supervised image. I still ended up with a confusing mish mash of land cover types in portions of both maps and agreement between land type categories between each map is sketchy.

Wednesday, July 15, 2009

Remote Sensing Module 4:



Usefulness/Pitfalls of Image Rectification: Once an image has been rectified with a reference, locating points of interest become much easier. In the Pensacola image (above at left) it could be difficult to identify specific buildings, a school for instance. But, after rectifying, the coordinates of a particular school could be taken from the reference map (above at right) and the school could then be pinpointed on the image. A particular issue I had with rectification arose after I placed the first check point. The next one would vanish. I assumed my control points were bad, so I started over with a new set. I ran into the same issue: vanishing check points. I tried them in several places and noticed a distinct pause before rejecting the potential check point. I tried several more until the next one found a permenant home. It required quite a bit of trial and error with check point placement until I had the requisite five sited. My final check point error was 2.27.

Thursday, July 9, 2009

Remote Sensing Module 3:




Roads: The roads are bright because the asphalt paved roads absorb more heat and remain warmer overnight. In addition, the streets have already begun to absorb heat from the early morning (6:45 am) sun.

Natural and man-made vegetation: Overnight cooling of vegetation combined with its higher thermal inertia results in the vegetation being cooler, i.e., darker, in the early morning as it requires a longer period of sun exposure to heat.

Sidewalks and patios: The lower thermal inertia of the concrete allows it to more rapidly heat in the morning light. If the sidewalks and patios are made of certain materials, overnight heat retention may play a role in its brighter appearence, much like the asphalt roads.

Storage sheds in back yards: The sheds appear uniformly dark (cooler), much like the houses. The material from which the sheds were built, wood for instance, would have a higher thermal inertia requiring longer sun exposure for warming. Possibly the materials stored in the sheds may contribute to their cooler thermal appearence. A quantity of stored water or building material within a shed could impact the image. The constant shade provided by the shed would lower the temperature variations the material within experienced thereby increasing solar heating times.

Automobiles: The cars that have been sitting all night appear cooler as the early morning sun has not begun heating them yet. Though, cars that are idling or have recently been operated show hot spots from the heat of the internal combustion engines.

Bright spots on many of the roof tops: Ground Temperature is in the lower fifties (Fahrenheit), so the hot spots are either furnace vents from heating units or chimnies above active fireplaces.

Thursday, July 2, 2009

Remote Sensing Module 2:














The bright blue areas in the multi-spectral image are very hard to recognize in the panchromatic image. When viewed as a color infrared composite the areas stand out in bright red suggesting it must be some type of vegetation. Given this and the fact that the blue colored areas are located in the water, I would guess that they represent a algae bloom, a very dense concentration of seaweed, or some other type of sea plant.

Friday, June 26, 2009

Remote Sensing Module 1:


What problems might you infer or identify in using this type of photograph?


Plants and foliage showing in red tones in not at all intuitive and could cause misinterpretation of the photo. Visually, the dark contrast between the bright beach and the ocean gives the impression of rapidly deepening water, which isn't at all accurate to the area. Seeing a familiar area from the overhead vantage offered by an aerial photo may cause problems as the bird's eye view of a location looks quite different from our regular line of sight experience of it. Also, at a certain height, details become too small leaving a lot of different areas looking much like any other in the absence of large features like mountains, lakes, rivers, and seas. Without a stereoscopic image the photo lacks the dimension of depth, making it impossible to get a sense of the relative difference in elevation between different locations.

Tuesday, April 28, 2009

Module 11: Google Earth















Average wind speed at Station MKGM4 at Muskegon, MI (to the northeast of site) is 11 knots (12.7 mph)with gusts of 15 knots (17.3 mph). It's beyond the BERR recommended 200 to 300 meter distance fromdwellings to minimize noise pollution. It is also beyond the BERR recommended 400 to 800 meter distance to avoid adverse reactions to Shadow Flicker. Major shipping routes out of Chicago and Milwaukee are to the southwestand west, respectively. The site is a good compromise with regards to major bird migration corridors,and should be minimally impactful to most species. Its proximity to land should help minimize buildingand transportation costs.

Saturday, April 25, 2009

Before and After













I drew my isolines as closed loops that went beyond the boundaries of the state. That's what you see in the first picture. I used the pencil tool, and once they were complete I clicked on the path and set the stroke and fill as normal. I used ColorBrewer to come up with the grayscale. After drawing all my isolines, I created shapes with the pen tool to conceal the portions of the isolines that went beyond the edge of the Georgia map. You can see the concealing shapes in the bottom photo. I changed their stroke to black so they would be visible. You will have to change the fill of the Georgia map from that beigey/orangey color to No Fill and make sure it is in a layer above the isoline shapes.

It becomes an exercise in setting up Layers appropriately. It's a bit slow (and tedious). I'm sure there's a more efficient method to do it, but I have not puzzled it out. Good luck!

Sunday, April 12, 2009

Module 9: Flow Maps




















Maximum stroke size was 30 points. I grouped Antartica into the Unknown category and added the question mark circle to indicate unknown countries of origin. I also did not include Alaska or Hawaii. Since I'm only showing immigration to the country itself and not to individual states it seemed unnecessary to add anything beyond the contiguous U.S.

Friday, April 10, 2009

Module 8: Contiguous Cartograms




















I'm not at all sure about the legend. I wanted to convey that the countries with the greatest increase in size indicated a greater GDP. I used a divergent color scheme to further highlight which countries had higher GDP's. The size increase from the cartogram application made it easy to see which countries had higher GDP's, if I was familiar with the geography. But, a country like South Africa, whose actual shape I'm not terribly familiar with was difficult to determine at a glance.

Failed to include the number of iterations I used. It was 8.

Module 8: Non-Contiguous Cartograms





















The biggest problem I had with this map was the projection. If I applied the Robinson World projection, or any other world projections, prior to running the cartogram application, the cartogram output would be at such a large scale it was imperceptible on the small scale world map. I applied the GCS 84, ran the cartogram, then added the Robinson projection. The GCS 84 gives Alaska a stretched out look. The Robinson projection looks more like I expect the world to look. I was guessing on this legend as well. I used the 5 color divergent scheme here, too. So, I added 5 circles. They are only representative of circles that would fall in each division. I prefer the contiguous map. It makes more sense to me visually. In particular, the non-contiguous map is very confusing to look at without the further differentiation a color scheme provides. I made the circles slightly translucent to reveal the base map beneath. It's a bit gaudy looking.


Wednesday, April 1, 2009

Module 7: Dot Distribution Map

























I ended up using a dot value of 3, 1 dot is 3 houses per square mile. I used a dot size of 0.6 points, but later added a stroke of 0.2 points. I choose not to display county names, rivers, lakes, or wetlands so the dots would stand out. I found the map very busy looking with all layers displayed. Plus, I had so much fun making 2289 blue dots I just had to show them off :)

Friday, March 20, 2009

Choropleth Map Part 2






















After coming up with an average population growth for each division, I followed the equal interval formula in the text and came up with a class interval of 4.74%, a high value of 29.6%, and a low value of 5.9%. Equal Interval was probably not the best classification method for this data. The 4th class break, 20.13 - 24.86, is not used. The growth in the mountain division is skewed by Nevada's 66% growth, which is more than twice that of any other mountain division member and, eight times a much as the division slowest growing state, Wyoming at 8.86%. The fact that the entire Northeast and Midwest fell into the first break created a uniformity that's visually unappealing. The grayscale was set manually by percentage in Illustrator. I scaled the divisions down to 75% and moved each around. Seperating each division looked so good on a classmate's blog I had to steal it.


Thursday, March 19, 2009

Choropleth Map Part 1





















I chose the Albers Equal Area Conic Projection for Hawaii and Alaska because... well, because ArcMap had a specific Albers projection for each state. I assumed they're optimal for each one. It would be odd if the Alaska Albers Equal Area Conic Projection was best suited to, say, Belgium and not the state it was developed for. I chose the North America Albers Equal Area Conic Projection for the lower 48 just to keep the Albers theme going. Actually, the North America Albers projection looks the way I expect the contiguous United States to look. I choose the color scheme to enhance the contrast between states that experienced more or less growth. The red/pink color offered enough visual differentiation in the middle classifications for me to see clearly.

Monday, January 19, 2009

State Preferences












I live in South Carolina. I have family both here and in North Carolina. I'd like to stay relatively close. Most of my green states were based strictly on their cold climates.

Good map / Bad map

1.) Bad map of The Six Duchies from The Farseer Trilogy by Robin Hobb (click map to expand)


2.) Good map of Westeros from A Song of Ice and Fire by George R. R. Martin (click map to expand)



I chose the map of Westeros (#2) from Martin's A Song of Ice and Fire series as an example of a good map and the map of The Six Duchies (#1) from Hobb's Farseer Trilogy as an example of a bad map. Clicking either map will open a seperate window with a better view. Map #1 shows very little detail. Map #2 does a better job of conveying the location of forests, hills, and mountains. The rivers in map #2 and the coastal regions appear in more detail. Map #2 also includes major north/south and east/west roads, while map #1 does not show any travel routes. The second map also includes the names and locations of all cities, towns and landmarks. While reading the books from the Song of Ice and Fire series, I often referenced the map to see where cities lay in relation to other locations visited in the story. Aside from the location of a handful of major towns, the map from The Farseer Trilogy simply didn't offer enough detail to be useful. Perhaps not coincidentally, I enjoyed the Martin series much more than the Hobb series.

Map #1 from: http://crooty.deviantart.com/art/Robin-Hobb-Map-10431866

Map #2 from: http://www.coldones.eu/images/maps/westeros_complete.jpg