Today we are going to make some spatial data from scratch.

After today you should be able to:

Create a .dbf table

Convert it into a set of points (shapefile)

Create a CSV table

Address match a table

Edit and print metadata.

We all understand street addresses better than decimal degrees, so why can't we just get our computers to think that way? Well, to some degree (pun intended), we can. If you have appropriate software (like ArcGIS), and appropriate reference data (like TIGER street files from the U.S. Census Bureau) you assign coordinates to a record based upon its street address. This act of assigning coordinates to street addresses is commonly called address geocoding.

Address geocoding is especially popular in marketing, parcel delivery, and other geo-business applications because it allows for easier geographic analysis of mailing lists, customer lists, and competitors. It is also part of popular web-based applications like www.mapquest.com that ask for an address to help you make a map. Almost all of these begin with some version of the TIGER (Topologically Integrated Geographic Encoding and Referencing) street files, but have cleaned these up and added information. On a nationwide basis the clean up can be expensive. That's one reason why free services (like MapQuest) might be off a little on addresses like 1600 Pennsylvania Ave or 14 College Farm Rd, but expensive pay services can get it really close.

The good folks at NOAA have helped track plenty of hurricanes over the years. We want you to pick one from their site and enter its name and your name on our sign-up sheet. This way, everyone will have different hurricane and resulting map. To get to the top level of the hurricane archives, follow this link:

http://www.nhc.noaa.gov/pastall.shtml

Choose a year, then a hurricane. Scroll down to the table that includes the latitude and longitude information. Many, but not all, have Lat Long coordinates in a table. The more recent ones will have better data, but the older ones may have more exciting paths to follow. Don't forget to enter your name and the storm's name on the sign up sheet.

Simply highlight the entire table and copy it, then paste it into a blank Excel spreadsheet. (You might need to use Paste Special...and paste it as Unicode text)

Now that you've created your own hurricane shapefile, take a shot at entering some basic metadata for it. Fire up ArcCatalog, find your hurricane shapefile and view its metadata. Thankfully, ArcCatalog has gone through the trouble of automatically entering some of the metadata, such as the Spatial Domain and Attribute Information. Notice that by choosing a different style sheet in the Stylesheet drop down menu, you can change the way the metadata is organized and displayed but the metadata itself doesn't change.

Click on the Edit Metadata button next to the Stylesheet drop down menu. Notice there are two levels of tabs for navigating through the metadata. Explore the tabs to see how you'd go about accessing the different types of metadata for editing. Fill out the Description boxes and the Data Set Credit under the Identification->General Tab, the Identification->Time Period, Status and Keywords (don't worry about the thesaurus) information. Remember, you created the shapefile but not the data, so be sure to give credit where credit is due. The federal standards for metadata can be found at http://www.fgdc.gov/metadata/metadata.html. For a taste of well-done metadata, look at the examples found at http://njgeodata.state.nj.us/. To access the data and metadata at this site, you need to create a login, but it's worth the effort because they have a lot of data that you might find useful in a few weeks.

A different way to create data from scratch is to use a geocoding system based on street addresses.

Go to an online Yellow Pages type directory and find at least ten street addresses in Middlesex County (maybe something like, 10 tire stores, or ten public libraries, or ten bookstores, or ten churches, or ten wellness centers, etc.). Open an new Excel file and enter these addresses with the street address, town, zip code, and the name of the facility or store as your fields. The first row in the Excel file must be the names of your fields, not actual data. Use Address, City, Zip, and Name as your first row field names. Save this as an Excel file, THEN as a CSV or comma delimited file.

Add the table in ArcMap. Open it and see that it all made it alright, then close it. Right click on the data and choose to Geocode Addresses. Add a geocoding service from Y:\intgeo\rowan\avtrn\middlese that is called street.mxc

Make sure that you Zone is your zip code.

And set your geocoding options to include ALL 4 possible Output Fields. Hit OK.

If you have less than 75% matching, you should use the Interactive matching to see if you can improve your odds. We'll go over this one in class for sure.

You can add the streets layer from the middlse directory for background.

What is the coordinate system for these data points?

When you are done with your map you can export the layout as a JPG or some other useful graphic formats.

You should turn in 2 maps. One of your hurricane, one of your address matching. Make 'em nice, and save your data for future use. Also turn in your edited hurricane metadata.