2010-12-13 Labs 8 and 9 Post 2

For the most part, the export from ArcGIS to GIF files went smoothly.  I was able to export the maps with the legends in place and the legends did not wobble.  The only problem I encountered was with the Dot Density Map.  When I animated it, the map itself moves in relation to the title, legend, and neat-line.  I do not know why this happens.  I tried to solve this problem by exporting the maps a second time.  This failed.  I tried a third time.  This too failed.  Finally, I decided to leave it like that.  The movement of the map itself is fairly gradual throughout the animation and does not really detract from the information being presented.  It just looks bad.  I will try to export the map a fourth time to see if that rectifies the problem.  The problem may just have to do with the computer I am working on.  There were several other times when the images on the screen moved without any user input.  Possibly, restarting the computer just prior to exporting the maps would be advantageous.  The only problem I had when going from Arc to Earth was that I did not include all of the files associated with my GIS file.  I just had to go back to my computer, load the other files onto my storage device, and then go back to the instructor’s computer.  The program worked very smoothly.  Although, when I tried to zip my files, the instructor’s computer would not allow me to open the zipped files without using WinZip.  I got around this by not zipping the files.

I used two other methods for displaying my some of my data.  The animations of the Choropleth, Gains/Losses, and Graduated Symbols maps change images fairly quickly.  This is useful for determining patterns.  However, these maps are intended for an audience of students.  Students often find note-taking useful.. Therefore, I made slower animations of these maps and put the choropleth maps on the website as individual images.    The slow-moving animations have about a ten second delay between maps.  This would allow students to jot down some notes about the information.  Changes to the speed of the animation were made by shift clicking all of the layers in the GIF and changing the timing by hundredths of a second.  Likewise,  the individual images of the maps allow students to examine the maps, and hence the symbolization of the data, without time constraints; this, in turn, allows the students to take better notes on their material.  The individual images were inserted in the html text so that they opened in a new tab.  This was achieved by using the code: target=”blank.”

ArcGIS is fairly expensive and not everyone has it.  Google Earth is a free download, at least the basic version is free, and anyone who has a computer or a local library can use the program.  By converting ArcGIS files into KMZ files, cartographers make maps that are relatively limited in their audience into maps that are useable by everyone.  Transferring the files from ArcGIS to Google Earth was certainly worth the time because it opens the information up to many more people.  Similarly, transferring the ArcGIS files to a GIF format, cropping them in photoshop, and animating them makes the maps available to a broader audience.  Any web browser is capable of displaying GIFs.  This format appeals to an even broader audience than Google Earth because people do not even need to download anything.  They can just view the images.

The data we originally collected from the Census Bureau was also in a format that everyone could use on their computer.  Arguably, people could get all the information they needed to understand population change in Maine, New Hampshire, and Vermont from the Census Bureau’s website.  However, turning the numerical data conveyed by the Excel spreadsheets to a graphical representation also opens the information up to a broader audience.  Transforming numbers in a spreadsheet to a map, and then transferring that map into a program that everyone can use was well worth the time.  This process also takes data that are not readily relatable to the phenomenon they are representing and turns them into data that are easily related to the data they represent.  These skills would be very useful in a job setting where I might have to collect data from a local source and relate that data to a national or worldwide source.  For instance, if I were doing a research project for botany and trying to map the occurrences of a particular plant, such as Filago pyramidata, to other occurrences in the world, and show that information on a research company website, collecting data from the web, integrating it with the data I collected, and then showing that data on a common medium–such as GIFs or Google Earth–would be very useful.

Yes, animating maps of this type definitely helps convey the information to the intended audience.  The phenomenon that is being portrayed in the maps is the number of people in a specified area in each decade for the last 109 years (by decade).  Obviously, we cannot put people themselves on the map or use the states themselves for the background.  We use symbols or colors to represent the people in the area.  We use a shape image to represent the earth/area.  Since we can use time to portray itself, that seems like the best option.  Obviously, showing a real-time version of the data would not be an effective way of showing the change in population in an area over time.  However, scaling the time and having several different speeds is very useful.

The faster the animation is played, the easier it is to see patterns.  In Maine, New Hampshire, and Vermont, there is a fairly noticeable trend in population growth in the area along the coast, closest to Boston; there is also significant growth in the areas of Vermont closest to New York.  This trend is fairly visible on all of the maps.  However, the animation of the multi-class choropleth maps is probably the least helpful.  This may be because these maps have too many classes for an animation.  The gains/losses maps seem to be much easier to pickup patterns from when they are animated.  The dot density and graduated symbol maps show trends very well when animated.

Yes, not only have I considered including the static maps, but I had decided to do so as an extra way of displaying my maps.  As I mentioned above, the intended audience is upper-level high school students.  Students often find note-taking ver helpful for learning their subject.  By being able to view the static images on the webpages, students can take time making notes on the data.  Kinesthetic learning may be better for some students than viewing animated, visual representations of the data.

When I was creating my GIF files, I found that you could make them play continuously (loop) or play through once and stop.  For the purposes of this project, playing continuously is a little bit more advantageous because seeing the animation multiple times often makes patterns more evident.  However, playing through once could be advantageous for other uses of GIFs.  For instance, if you were trying to make a very long GIF of several thousand photos, from an expedition or research project, running through a lot of images at a fast rate could be very impressive.  The looping feature can be changed when the first layer (prior to the first image) is selected.  Going through the images only once might be more impressive than seeing them over and over again.  Also, varying the speed for the animations can be useful.  Faster speeds are better for showing patterns; slower speeds are better for seeing the information displayed on each individual map.  This feature can be adjusted for individual layers in the GIF, or all layers can be selected (shift click) and the speeds can be set uniformly.

They should be pretty useful.  The only drawback to the KMZ files is that a person needs to have Google Earth on his/her computer in order to view them.  In theory, the Google Earth Plugin would allow people to view the files with a web browser, but getting the files to a point where the plugin will work seems to require scripting the entire file in html text and was very confusing.  On the other hand, Google Earth is really the ultimate locator map.  Everyone would recognize the image of the globe.  Google Earth starts at the global level and then zooms into the KMZ file.  The KMZ animations look much better than the GIF files.  Part of this may be attributed to cropping; part of it may be attributed to the background and part may be attributed to the familiarity most people have with Google Earth.  The KMZ file look very good and the timing between the images is very even.

I chose to use two different methods for displaying the maps.  The first was a slower GIF file and the second was to display each of the maps as individual images for specific map types.  As I mentioned before, the slower GIFs allow users to see the information on each map better, but may make it harder for users to see patterns.  The KMZ file looks more professional than any of the GIFs.  Once again, both the slower GIFs and the static images will allow the intended audience to take notes on the information they are viewing.

 

2010-12-8 Labs 8 and 9

Rather than cramming too many things into a small area or covering up part of the map (and hence covering up the data–which are the real point the map), I decided to include a link to a website that works as a locator map for almost any location in the world.  This avoids the problems associated with placing a locator map on the main maps. The states themselves will be close to the center of the map; since Maine, New Hampshire, and Vermont form an L-shape that has been rotated 90 degrees counter- clockwise, the states themselves are asymmetrical. By moving the states slightly to the left of center, the overall image of the map user’s focus will be drawn to the center of the total area of the three stated. Rather than putting the projection type on the map itself, I decided to put the projection name in the descriptive text on the main project page.  All of the maps share the same projection, so it would be redundant to have the name of the projection on every map.  Furthermore, having the projection name on the maps in the animation causes even more wobble.  Since I am apparently bad at cropping the maps to begin with (or my computer is just particularly grumpy), the fewer elements I place on the map, the better.  A legend was supposed to be placed in the bottom right hand corner just above a text box showing the projection type.   Since no projection type was placed in the text box, the text box was eliminated from the design.  Instead, I put a small scale bar in the bottom right hand corner and then placed the legend on top of that.  These elements are slightly lower on the grid than I had originally intended.  This difference can be seen when comparing figure 1 and figure 2.  Some of the legends, especially the graduated symbol legend, went into the space I had intended to use for the informational text.  Any additional commentary on the maps was placed on the webpage rather than on the maps.   This web design still keeps the maps and the data being displayed on the maps in the natural line of sight, allowing the users to focus on the map itself. The balance will be very close to centered (latitudinally). The overall final map design can be seen in figure 2.

The changes to the map layout include:

1. Eliminating info text, projection type, and locator map (figure 1) from maps
2. Adding scale bar to maps (figure 2)
3. Putting a map locator link on the html page
4. Putting additional commentary on the webpages, rather than the maps themselves.

Overall, these changes bring the map viewer’s attention to the map and data displayed on the map.  The information that has been eliminated from the maps is still present on my webpages; it  just is not creating clutter on the maps themselves or causing additional wobble in the GIFs.

Visual Elements on Web Pages:

Webpages Maps

background Data being displayed in maps

Informational text Map background

hyperlinked text State images

Images/photos County images

Dividing lines between body and links State and county borders

Title Legend

Less important informational text Title

Maps Informational text

While “a foolish consistency is the hobgoblin of little minds, adored by statesmen and philosophers and divines,” as Emerson said, an intentional and wise consistency can open new doors for the mind.  By seeing a set of  webpages consistent with each other, people should be able to intake the information conveyed on these webpages more easily.  The visual hierarchy of the webpages and the maps should be consistent.  The webpages use a dark background (black) with more important information in brighter colors (white) and external links (less important) in darker colors (red).  Prior to this evaluation, the hyperlinked text to the annoying Google Earth Plugin (that never fully worked) was as bright as the other hyperlinked text on the page.  Since the Google Earth Plugin was useless for this project, I removed this and put a link in that leads to the download for Google Earth.  Since Google Earth is necessary to use the KMZ files, it makes sense to have it on the same level in the visual hierarchy as the files themselves.  Brighter colors stand out on dark background and bring page-viewers’ focus to what is important.  The maps should do the same.  In the above list of visual elements, items of less importance are shown in darker colors while items of more importance are shown in brighter colors.  The same is true of the maps and webpages.

Color differences for the population change will be the best means of communicating this change.  The change could be represented with different patterns–this would be ideal for a black and white map.  However, since computer monitors are capable of displaying a decent array of colors, a color gradation would be best.  Furthermore, using a spectrum of different shades of the same color would emphasize an increase or a decrease in county population.  By using variations of the same color rather than different colors to represent population change, map users can intuitively determine increases and decreases rather than having to look at each individual color in the legend.  As an overall hierarchy, the webpage background will be the darkest.  Text on the webpage will be bright.  The map background will be white, and hence brighter than the webpage background.  The states on the map then the webpage background.  The states on the map background will be darker than the map background–and hence stand out from it.  Finally, text on the map will be darker than the background.

The visual hierarchies for the webpages are slightly different than the ones for just the maps.  When people look at a map, they can see the entire map at once.  The webpages are longer than what a typical monitor shows.  Users must hence scroll down the page.  On the maps, having an asymmetrical setup is advantageous because the map reader can see the entire map and their eyes are allowed to naturally flow across said map.  On the webpages, however, having everything left justified is not necessarily advantageous.  Since web-users will have to scroll down the page anyways, centering the pages gives them a more professional look.  Prior to this evaluation, my webpages were left justified.  I have now centered the text on the pages.

My main project page has fun facts about the three states, an Ohio map, and links to images of the some of the maps I created in ArcGIS.  At the moment, the links to the images are at the bottom of the page; the Ohio map is at the top, and the fun facts are in the middle.  I had placed the Ohio map on the page in order to have a representation of what the Maine/New Hampshire/Vermont maps would look like.  Since the maps are now on the page, the Ohio map can be deleted.  Also, the fun facts should be below the Percent Change and Proportional Symbol maps.  These changes have now been made.

It was fairly easy to export the maps from ArcGIS to either GIFs or JPEGs.  After the maps were in GIF or JPEG format, however, they needed to be cropped using Photoshop.  This was not fun.  When the cropping is off by as little as one or two pixels, the maps wobble when they are animated (with a GIF animator).  These cropping differences can be corrected, to a certain degree, in Ulead GIF animator.  When you click on the map, you can use the arrows to move the maps around in relation to the others.

The html script for creating a Google Earth Webpage (viewed with the Google Earth Plugin) would not allow the program to function using Google’s sample scripting because the page needed a new API code.  After I created a new API code, the page would still not function.  After wasting time creating several different API codes, I realized that Google’s coding for “create an API code here” was still in the page.  After I removed this, it worked perfectly.  There does not seem to be an easy way to open a KMZ or KML file with the Google Earth plugin.

 

In some ways, the Dot Density Maps imitate the actual phenomenon the best.  The number of dots increases in areas where population increases and the area in the map stays the same.  As more people come into the area of Maine, New Hampshire, and Vermont, the number of people increases, but the area remains the same.  Thus, the increase in the density of dots mirrors the increase in the density of people.  However, the dot density maps are not perfect.  The dots are randomly placed throughout the counties.  The people are likely clustered in cities.  Hence, areas in the counties that are farther away from the cities likely have fewer people than the dots indicate and areas near the cities likely have more.  This is a limit of the dot density maps.  The fact that the map is showing density seems, to me, to mirror the actual phenomenon better than the other maps.  At the same time, the choropleth maps convey the information that is actually captured by the data best.  Since the data we collected are only at a county level, we do not actually know that the population is clustered in cities.  By showing the data at a county level, the choropleth maps portray the data better than the phenomenon.

Since the data sets for the Choropleth Maps are diverging (losses and gains), a double ended spectrum makes sense.  The spectrum that I had originally chosen was from red to green.  This looked very good to me.  There were clear distinctions between all of the  different categories with this color scheme.  The color values were different enough that the categories could be distinguished.  The color hues were aesthetically pleasing.  The color intensity was also good.  However, I am not colorblind.  By using the colorbrewer website, I was able to devise a color scheme that was equally well distinguishable and was colorblind-friendly.

The terrain color in all of my maps (except the choropleth maps) is either some variant of green or gray.  Maine, New Hampshire, and Vermont are all fairly vegetated.  As the lecture pointed out, the color green is commonly associated with vegetation, forests, and nature.  These stereotypes are fitting with the New Hampshire, Vermont, and Maine countryside.  The lecture on color and cartography also said that shades of gray are associated with the following adjectives: quiet, reserved, sophisticated, controlled.  Certainly this is true.  However, gray is also the color of a lot of the rock in these states.  There are some variations, of course.  Much of  the rock in Acadia National Park has a reddish tint to it.  There are many different shades of granite in Vermont–one of the major sources for granite countertops and headstones in the U.S.  A lot of the rocks are either gray or some variation of gray.  Hence this seemed like a good choice for the color of the “land” in some maps.

After looking at the colors on the dot density maps, it seems that the shade of yellow used for the dots is slightly greenish.  Since the background is green, this makes the yellow dots blend in with the background slightly.  This is an example of simultaneous contrast.  By making the yellow slightly less blue, it will stand out better from the background.  Since these maps wobble quite a bit in the GIF, I am re-exporting them with the updated color.  Hopefully, this will get rid of the wobble and make the symbols stand out from the background more.

 

2010-11-1 through 2010-11-15 Lab 7 Log

The classification part of this lab went well.  After opening the DBF spreadsheet containing the Maine, New Hampshire, and Vermont data, it is possible to copy and paste the percent change data into a new spreadsheet in Open Office.  By doing this, the data (and hence the histograms) can be saved without distorting the data for ArcGIS for future reference.  After the histogram was constructed in Open Office, finding natural breaks in the data was easier when the data were sorted by either ascending or descending values.  However, there were only slight changes in between any of the points on the histogram–the curve was fairly even except at extreme ends of the histogram.  This may be, in part, due to the fact that there were over 500 data points plotted on the graph (all counties in Maine, New Hampshire, and Vermont  and eleven sets of data for each of those counties).  After the graph was enlarged, nine “natural” breaks were found in the data; hence, there are ten categories for the choropleth map.

ArcGIS allows users to manually set category breaks.  This is fairly easy to do–they can be set by either moving lines on a histogram or by typing the values into a box on the right hand side.  I found that typing the values was faster; when I used the mouse to move the dividing lines on the histogram, I found I ended up with values like -6.99974 rather than -7.  ArcGIS does not allow users to create categories that contain no data.  This is a slight problem when you are trying to make the same categories for more than a century of population data.  Not all of the data contain values at either end of the spectrum.  This should not be a problem as long as the scale for all of the maps is clear.

About 80% of the changes in population were increases and 20% were decreases.  This diverging data set requires the use of a double ended color spectrum.  Originally, I had constructed a diverging color spectrum using ArcGIS’s preset color options.  This color scheme was soon replaced with a teal to brown scale using RGB color codes from Colorbrewer.org.  Oddly, the one color code for this spectrum produced a very different color in ArcGIS than what it showed on the color brewer website–one of the beige colors appeared to be red.  However, when I used the CMYK codes for this color, it looked right (the CMYK code is 12, 20, 45, 0).

The histogram for the total population data was similar to the one for the percent change data in that it had a fairly even increase when arranged in ascending/descending order except at either end of the histogram where there were very distinct changes.  After I zoomed in on the histogram, there were six points that looked like good breaks in the data, so there are a total of seven categories for this data: 1) 2500-9999, 2) 10000-24999, 3) 25000-49999, 4) 50000-79999, 5) 80000-149999, 6) 150000-249999, and 7) all values greater than or equal to 250000.  The colors for these maps, the graduated symbol maps, took some time to work out.  The backgrounds were and are a pale greenish color.  When the counties had white lines and the blue circles had black lines, the counties were more noticeable than the graduated symbols.  When the boundaries were made gray and the circle boundaries were made white, the important data stood out.

As in the graduated symbol maps, I used a greenish background for the dot density maps.  Since land is either green, brown, or gray in that area (with vegetation, dirt, or rock), one for these colors (or variations on these colors), this color scheme makes sense.  Gray borders on the counties are visible, but do not distract from the information that is conveyed by the map.  Bright yellow dots stand out from the olive-green background and were hence used for the symbols in this map.

When I looked through the symbol choices in ArcGIS for the proportional symbol maps, I found an almost unlimited number of choices.  Hence, cars seemed like a viable choice for conveying the presence of people.  If cars are present, people are present.  ArcGIS even offers Aston Martins as a choice for symbols.  Who wouldn’t want an Aston Martin?  This choice for symbolizing population was too tempting.  The population of Maine, Vermont, and New Hampshire are, therefore, represented by James Bond cars of varying sizes.  I used a light gray background with thin, white borders on the counties.  This background makes the dark-gray Aston Martins stand out.

2010-10-3 Through 2010-10-11 Lab 5 Log

Lab 5 went very smoothly.  There were no problems linking the .dbf of Maine, New Hampshire, and Vermont population data to the ArcGIS file for the states.  The central meridian for Maine is at -70 degrees, 22 minutes, and 30 seconds (according to the Maine Revised Statutes page).  Below are the definitions.

.DBF/ Dbase file: A .dbf or a Dbase file is a data base file.  Most of the data contained in dbf files are similar to data that would be contained in an .xls (Excel) file.  Dbase files are a fairly simple form of tabular data.  Microsoft Excel is no longer capable of exporting .dbf files, but Open Office is capable of exporting this file format.  Dbase file format works very well with ArcGIS.

Query: To query is to ask questions or interrogate.  “Query” and “inquiry” share the same etymology.  In ArcGIS, as in all computer software, a query is a search tool that allows you to, functionally, “question” the database in order to extract information from it.  A specific type of query in ArcGIS is the ability to Select by Attributes. In an attribute query, ArcGIS users can select a feature such as state name and question the database by asking it to pull forward all values equal to “Maine.”  Users can also include multiple states by saying “Maine, Vermont, OR New Hampshire.”  Similarly, if an ArcGIS user were trying to find population data for cities of a certain size, he/she could select a cities layer for the map, select population, and then use greater-than/less-than symbols to identify the particular size of city.

Fields (in a Table): Fields are the columnar data for a particular attribute in an attribute table.For instance, a field might contain all population data for 1940.

Records (in a Table): Records, in tables, are all of the information for a particular feature.  Records are the information found in a row.  For instance, records for an entry such as Aroostook County in Maine might include FIPS codes; population data for 1900, 1910. 1920, etc.; and total area.

Attributes: Attributes are information about geographic features that are not spatial data.  Attributes are typically found in a table linked to a spatial representation on a map.  Attributes of a state might include the state’s name, the state’s area, the state’s population, etc.  Attributes can be found in cells in an attribute table.

Join Function: The join function in ArcGIS allows users to link multiple data sets together in one table without the hassle of transferring all of the data.  It essentially makes a secondary table an appendix to the primary table, forming a Relational Database. A relational database is  a table, or similar database, that has two tables linked together by a common data-field.  For this lab, I was able to link two databases together using the FIPS codes in two tables as the common field.

Monitor Fire: Monitor fires are very dangerous!  They can occur when Arc GIS user are trying to import data or when a sparkler is placed inside a computer monitor.  Based on the amount of dark gray smoke coming out of a monitor when it burns, monitor fires can also be very bad for the environment.

Calculate/ Field Calculator: The field calculator does precisely what its name implies: it calculates the value of a particular field based on an equation entered into the calculator.  In this lab, we used the field calculator to calculate the percent change in population in each decade for our states’ counties.

2010-9-29 Through 2010-10-6 Lab 4 Log

One interesting feature of ArcGIS is that you can edit data.  After adding the editor toolbar, you can select a particular set of data (either attributes or vector data) and modify them.  For instance, this summer, I did a plant community study.  All the trees in a 20 x 80 meter plot were marked with a G.P.S.  After these data were downloaded into ArcGIS, some of the points were clearly outside of their actual location (under the forest canopy, the G.P.S. had an accuracy of approximately 5 meters).  By using the editing feature of ArcGIS, it would be possible to move the wayward points back into the area of the plot.

Another interesting feature of ArcGIS is that you can add text and pictures to a created map.  If you were trying to create a map of a particular geological feature, let’s say waterfall, in the state of Ohio, you could take photographs of all the major waterfalls in the state, locate the coordinates for waterfall in ArcGIS, and place a photo of that waterfall at the “actual” location on the map.  This could be useful for either geologists or tourist information centers.

On ESRI’s website, there is a page on banking.  This was very surprising.  Obviously, reality agents, and hence banks, need to know where properties are that are being sold or have a mortgage.  Initially, G.I.S. software seems like overkill for this.  However, the ability to have all the information about a particular location in an attribute table could be very convenient.  Furthermore, G.I.S. software could be very useful for the banking industry when corporate offices are attempting to discern the best location for future branches.  ESRI’s website claims that their software will “enhance your [organization’s] understanding of risk, customer interaction, and economic conditions using spatial models based on geographic and geodemographics.”  This sounds like a bit of a stretch, but people have apparently figured out someway to make this work.  If nothing else, it is a good marketing strategy.

GEOINT (Geospatial Intelligence) is ESRI’s mapping program for military intelligence work.  Most of the information on it seems slightly vague, but there are some interesting features listed on the website.  ESRI claims that map production, G.I.S., and imagery analysis are typically thought of as different “domains.”  GEOINT allows users to perform all of these tasks in one program.  Also, GEOINT allows users to access “multiterabyte” data sets through one interface that is accessible online.

ArcGIS is a neat program.  The user interface seems reminiscent of Windows 95 or Windows 97; however, everything is fairly logically laid out.  Most of the features can be used easily without having to spend long periods of time looking for how to use them.  While the user face seems a little outdated, the capabilities of ArcGIS are very impressive.  Maps conveying almost any type of information can be created in ArcGIS and huge quantities of data can be linked to these maps.  The fact that ArcGIS cannot be used on Apple computers (while using an Apple operating system) is slightly annoying.

The maps I created in ArcGIS are below.

This world map shows data for coal production by country along with major rivers and drainage areas.

This Ohio map is on an NAD 1983 State Plane Coordinate System for Ohio. The color gradation is representative of the base 10 log of population in each zip code.

Ohio Household Size

Ohio Rivers and Interstates

Map Document: a map document is the format type of anything created in ArcGIS.  It is abbreviated “.mxd” after the name of the file.  Map documents can store maps, symbols, hyperlinks, toolbars, and any other information created at the same time as the map itself.  ESRI developed mxd files.

Table of Contents: in ArcGIS terms, the table of contents is the area (usually on the left of the screen) that displays all of the layers of a map.  Here, users can check or un-check boxes; this controls the information displayed on the map.

Data Frame: the data frame of a map in ArcGIS defines nearly every feature of the map.  From the data frame, an ArcGIS user can change map projections, coordinate systems, and other information vital to the map.

Map Layer: a map layer, in the G.I.S world, visually represents the attributes of a particular set of data.  Map layers can only store one type of information.  For instance, if a person had mapped all the trees in a research plot as individual points and all of the herbaceous stratum of the plot as areas, said researcher would need two layers to visually represent this data on a map: one layer for points and one layer for areas.

Attribute Table: the attribute table in ArcGIS contains all of the textual and tabular data that specify geographic attributes.  Usually, the attribute table for an individual layer is opened.  If a county layer’s attribute table were opened, an ArcGIS user could view information about household size, ethnicity of the population, and other such information.  These attributes could then be visually represented on the map with a color gradation.

2010-9-20 through 2010-10-4 Lab 3 Log

The National Institute of Standards and technology (NIST) has developed a a system for naming/numbering counties and states within the United States; this system is known as the Federal Information Processing Standard, commonly referred to as “FIPS.”  The U.S. Census Bureau says that FIPS codes were developed so that all U.S. government agencies could have a standard way of referring to all geographical areas in the country.  FIPS codes are a set of numbers that specify a particular county in a particular state (in some cases, even more detailed regions are specified); initially, only odd numbers are used in FIPS codes.  By using only odd numbers initially, the Census Bureau has the capability to add additional FIPS codes using even numbers if new counties are created.  FIPS numbering order typically corresponds with the alphabetical order of counties within a state–this may change if additional counties are added.  Linked to every FIPS code is a data set containing various types of information.  For the purposes of this class, the most relevant information is population data.

Processing data is, as Dr. Krygier points out, an extremely joyful task.  The Census Bureau has data from 1900; however, these data are in a text file format.  In order to use the data in an ArcGIS system, they have to be in a DBF format.  This is not a difficult task.  It is just slightly tedious and time-consuming.  The text files should be opened with Excel and column width should be manually adjusted.  The FIPS codes, which are numbers, need to be read by computer programs as text.  In order to trick the computer into doing this, an apostrophe can be placed before each code.  Likewise, the years need to be made into text by putting a “Y” in front of each year.  The Census Bureau lists all counties as “Fill-in-the-blank County.”  Only one word (the county name) can be processed by ArcGIS.  To allow the G.I.S. program to work, “_County” must be deleted.  The Census Bureau has the data from 2000 and the 2009 estimate saved in an Excel format.  After this is copied and pasted into an Excel sheet with the 1900-1990 data, the commas must be removed.  After all the data from 1900-1990 (text files), 2000, and the 2009 estimate have been properly formatted into a single spreadsheet, they must be converted to DBF format.  Since Microsoft Excel has not been capable of doing this since at least the 2007 redesign, Open Office should be used to open the Excel file.  In Open Office, the opened Excel sheet can be saved in a DBF format.  All of this work is very straightforward and simple.  It is tedious only because there is a significant amount of copying, pasting, and reformatting.

2010-9-1 through 2010-9-22 Lab 2 Log

Creating web-pages in html was a new experience.  It is, actually, fairly fun and easy, albeit tedious.  There are several websites that have color codes for html writing.  Numerous other sites have information on commands that can be used for either changing font color of an entire page or for a single word.  By simply doing a Google search for “font color change html,” you can any code or command.  Using Notepad to write html scripts was simple and free.  I tried to use Text Edit on my Apple, but going between Text Edit on my own computer and Notepad on Ohio Wesleyan’s computers made a few changes to the font I had used.

Creating the map mash-up was relatively easy.  There was, however, one glitch in the process.  When I tried to title the one point, Daniel Webster’s Farm,” nothing showed up on the map.  You apparently cannot use apostrophes in normal text when writing html text.   The apostrophe was interpreted by Google maps as the end of the text.  To avoid this problem, the location was titled, “Daniel Webster Farm.”

Creating html pages is interesting even if it is time-consuming.  For future reference, the most important trick to keep in mind is to close each command.  When trying to change the font on a line of text, write the command and the close-command first–then write text in between them.  This ensures that only the right text will display the desired characteristics.

2010-9-1 Lab 1 Post

By doing a Yahoo Directory search for “U.S. Census Bureau,” the website can be accessed from any of the hits on the first page of results.  All of these results are for the Census Bureau’s website.  A search on Google displays mostly Census Bureau webpages, but also includes Wikipedia articles and the like.  Finding the more historical population data (1900-1990) was slightly more difficult than finding the most recent data.  Eventually, the quickest and easiest way to find the historical data on the Census Bureau’s website was to search for “Maine population 1900” in the search box on the main page of the Census Bureau’s website.  From the search results, the one that titled “County Population Census Counts 1900-90” leads directly to a list of the states.  Since this study is looking at Maine, New Hampshire, and Vermont, the names of each of those states were selected individually.  The historical data for Maine, New Hampshire, and Vermont were respectively titled, Maine Population of Counties by Decennial Census: 1900 to 1990, New Hampshire Population of Counties by Decennial Census: 1900 to 1990, and Vermont Population of Counties by Decennial Census: 1900 to 1990.  More current census data from the 2000 census, titles County GCT-PH1. Population, Housing Units, Area, and Density: 2000, on Maine, New Hampshire, and Vermont are also available on the website.

The fastest way to access the 2009 estimates is to select “Detailed Tables,” select by geographic type and state, and then to add the selected areas to the selection box and click “show results.”  There appear to be no titles for these data.  They are only listed as 2009 Population Estimates.  The 2009 estimates for Maine, New Hampshire, and Vermont can easily be found on the Census Bureau’s website.

The Census Bureau’s website clearly states that the government cannot legally copyright any of its information; therefore, it may be used by any and all individuals.

The School of Business at the University of Southern Maine has a webpage on population change by county for Maine and several other New England states.  This should be useful for interpreting population changes.

The Vermont Department of Health has some interesting data on population change in the state.  The site breaks down the change to a county level and provides some information about how the projections are made from the census data.  However, it does not interpret the reasons for change very much.

Smart Growth Vermont deals extensively with the reasons for population change within the state.  This site discusses everything from landscapes to urban sprawl.

The New Hampshire government has a website that talks about reasons for the increase in population.  Since 2000, the state has increased in population by 52,000 residents.  This will be useful for learning about why the population has increased.

The U.S. Department of the Interior has a website called, “nationalatlas.gov.”  This website discusses the increase in U.S. population at both nation and state levels.  It may be less helpful for interpreting population change at a state level than some of the other websites, but it does talk about why some states have grown faster than others.

The commercial website visitmaine.com provides a convenient list of historical places within the state.  Perhaps more importantly, it provides links to the websites of the historical places.

Maine’s website provides useful information about both the state’s history and its geology–which often plays an important role in settling areas.  This site, not surprisingly, contains information on the government of the state from 1820 until today.  Its information on Native American populations is fascinating.

The Vermont Historical Society’s website has information on each county’s history.  It seems that this information would be very useful for looking at the population of each county.

Two of the major manufacturing companies in Vermont closed during the 1950s and 1970s.  These closings caused a large protion of New Hampshire residents to commute to Boston.  City-data.com’s website discusses the history of Nashua, home to one of these closed manufacturing companies.

The Concord Historical Society provides a good history of the state’s capital.  This may be important for understanding the change in populations in Merrimack County.