Remote Sensing

I used Wayne County Landsat images. After spending some time experimenting with displaying different band combinations of Red, Green, and Blue values, including popular Landsat band combinations, I moved on to successfully display three bands simultaneusly. Below are my subset images of Wayne County. First my Unsupervised Classification:

After three tries I decided to give up on gettting all, or any, of my Supervised Clasification fields above 50%. I think the biggest problem with it is the roads in orange. Clearly there aren't huge chunks of roads like this. Surprisingly that field is the greatest percent reference accuracy. I think those orange road stripes across the water are cloud cover. But I think the urban area may be correct, as that is Detroit in the top right, and then further south Toledo. The vegetation could be correct to, but you would think there would be a lot more of it. 

Attribute and Spatial Queries

1. US cities that have more than 500,000 population are generally located in the east.

2. Separate State layer.

3. Cities in state. 

4. Cities in state with over 25,000 population. 

5. Roads that intersect my state 

6. Cities that lie within 50 miles of Lake Superior.  

7i. The following darker areas are more than 950 ft in elevation.

7ii. The following darker areas are between 925 and 1050 ft. 

7iii. These are areas that have less than 1000 ft in elevation and are also less than 50 degrees in slope. I assume the slope degree is set in the “z factor” box of the slope dialogue, i.e. putting 50 in there indicates all areas 50 degrees and under.

7iv. These are areas that have more than 1000 feet in elevation and less than 30 degrees slope. 

My Experience Loading Vector Data into QGIS and Creating Thematic Maps

The maps below illustrate the percentage of a country's population that live in urban areas as related to the rest of the world. Each of the four maps is based on the same urban population data set, and all divide the data into five class, as seen in each Legend. The difference between the four maps is how those five classes are divided. Below I display each map and describe and explain the patterns observed based on  the four methods of classification. In order to highlight the patterns, I select four diverse countries based on the data, the US, Australia, Vietnam, and Bolivia.

Sorting data by Quantile is to take the list of countries, ranked from greatest urban population percentage to least, and divide them into five equal groups. Here we see that Vietnam is in the lowest group, Bolivia in the middle group, and the US and Australia in the highest group.

The Natural Breaks method ranks the countries in order of urban population intensity, then divides that list into five groups that share the same general attributes. Here we see that among our four countries, Vietnam has now moved up into the second group, and there are no countries in the bottom group, which denotes no urbanization. Bolivia has now moved from the middle group into the second group. This shows that even though there are roughly half of the countries above and below Bolivia in urban population percentage, in reality, Bolivia belongs in the second most urbanized group in the world. The US and Australia are in the top quintile of natural break rank, not surprisingly.

Now we have the Standard Deviation map. Standard Deviation seeks to group the countries by the margin with which they differ from the norm, so the closer the number is to zero, the closer that country's percentage of urban population is to the global average; the greater the number, the greater the deviation from the average in the direction of greater urban population; the less the number, the greater the deviation in the direction of less urban population than the norm. Vietnam has now moved up again, from the fourth group to the third (or middle) group, indicating that it is only slightly below the global norm, even though it was in the lowest quintile. Bolivia remains in the second group, indicating that it is slightly above the global norm even though it did rank in the middle quintile. Australia is still in the top rank, postively deviating from the norm by the maximum among the groups. But the US has now dropped a rank to second, indicating that its percentage of positive deviation is less than the maximum group. It is not one of the very top urbanized countries. Australia ranks 11th, and the US ranks 23rd.

Here we have the Equal Intervals map. Equal Intervals classification is done by taking the range of data values, from highest to lowest and dividing that equally by, in this case, 5 classes. The US is back up into the top rank with Australia. This map really shows that despite their differences, most countries rank in the top half of urban population, even Vietnam and Bolivia. 

All of these maps demonstrate the advanced state of urbanization around the world. Vietnam, the least urbanized country highlighted, exemplifies this trend. Even though it is in the bottom fifth of the countries in urbanization, and is still naturally grouped with those in the bottom half, it still has a relatively comparable urbanization, varying only slightly less than the norm.

My Experience Collecting Data with GPS

          I used an iPhone as my GPS receiver with the Easy GPS app, and a trusy assistant, my son Logan, age 9. We embarked on our bikes, and Logan was instrumental in finding hydrants, reading and transcribing the coordinates, and creating very useful descriptions of the hydrants. Thanks to Christine's heads-up in her blog post, before hitting the road I changed the settings of Easy GPS to display coordinates in Digital Degrees.

Screenshot of the locations of 10 Ann Arbor fire hydrants on Google Earth. 
Click to magnify, or open KML file in Google Earth.

          Upon uploading the coordinates into Google Earth, I noticed two things. One, as we took our coordinates at all of the locations, Easy GPS said the reading could be plus or minus 4 to 7 meters. The 4-meter differential was at the location farthest from buildings. This translated into Google Earth showing some coordinates matching exactly the fire hydrant location, but others across the street, or in the middle of the street.
          And two, there is an acute triangle formed from the fire hydrant locations which denotes our path. I'll leave it to others to explore Street View around the tip of the triangle to discover what our sweet destination was!

Discovering Google Earth

I learned a lot in this about Google Earth in this Unit. I discovered the Look function on Google Earth, which I thought was a really interesting way to look around. It gives you a sense of the shape and proximity of neighboring counties, cities, and states from a static perspective.

I also found the Ruler function interesting and did not know that existed. I also did not know that you could explore Mars, the Moon, the Sky, or use a flight simulator (really fun).

Turning on and off the 3D layer gives you a good indication of the difference between the satellite imagery and what can be done on top of that by applying 3D information gathered from aerial photography and remote sensing. The 3D layer is the necessary cherry on top for satellite imagery because it allows you to explore the topography of the natural and urban landscapes.

The Tour Guide function is also very interesting and fun to use. These are where you will find the highest resolution and most detailed application of the 3D layer.

And this is all just the tip of the iceberg...