Tag Archives: GIS

Extensions need to be authorized

Recently, I wanted to use the Spatial Analyst extension on ArcGIS 10 and got this cryptic message.


I knew for certain that the extension was licensed. It showed as installed and licensed elsewhere in the program. After getting help from the campus IT techs we found that the extension had to be turned on by going to Customize –> Extensions and filling in the screen shown below. Then everything will work properly.


This is similar to the situation in Excel where the statistical toolpack has to be enabled in the program properties screen separately in order to work. I just wish ArcGIS had made a more explanative error message! 

Develop an Elevation Profile

In surveying we have been making elevation profiles from differential leveling data for as long as I have been teaching the course. Recently, I came across a help page from the University of Georgia that uses ArcGIS to associate data points with elevation data from a DEM, which is then graphed in Excel, with an XY scatter chart. The data can come from GIS points or a line digitized on the map. The edit – divide tool sets the points along a line at any spacing desired. Spatial analyst will extract the elevation from the DEM at each point.

Here is a section of the pipeline that goes through the Waynesboro Watershed.
And here is the resulting profile.
This method can help reduce the error form field measurements. And it can be used for much longer spans. 

Access or Excel

Yesterday in FORT 120 we started looking at Microsoft Access, a relational database management system. It has similarities with Excel, the spreadsheet solution. This raises the question, why learn both programs and when should one be used over another.

At first glance there are similarities. Both store data in tables of rows and columns. Both perform calculations and have form and report-writing utilities. However, there is a big difference.

In an interesting article on the Microsoft web site Emma Nelson pinpoints the big difference. Excel is best when all the data fits into a single table (known as a flat file). When the data should be divided into multiple tables that relate to each other through common fields, Access can best handle the job. In a relational database the data is divided into multiple tables to make storage more efficient and search quicker. Access has all the tools needed to manage a relational database.

Borrowing from the Nelson article, use Access when you:

  • Require a relational database (multiple tables) to store your data.
  • May need to add more tables in the future to an originally flat or non-relational data set.
  • Have a very large amount of data (thousands of entries).
  • Have data that is mostly of the long text string type (not numbers or defined as numbers).
  • Rely on multiple external databases to derive and analyze the data you need.
  • Need to maintain constant connectivity to a large external database such as one built with Microsoft SQL Server.
  • Want to run complex queries.
  • Have many people working in the database and want robust options to expose that data for updating.

And use Excel when you:

  • Require a flat or non-relational view of your data (you do not need a relational database with multiple tables). This is especially true if that data is mostly numeric–for example, if you want to maintain a financial budget for a given year.
  • Want to run primarily calculations and statistical comparisons on your data — for example, if you want to show a cost/benefit analysis in your company’s budget.
  • Know your dataset is manageable in size (no more than 15,000 rows).

In Geographic Information Systems (GIS) we combine geographic data with attribute data. This requires multiple complex data tables that use a relational database tools. In fact the geodatabases used in ArcGIS desktop use Access databases by default. When you make a map with data in ArcGIS you are, in effect, using a relational database.

Use GIS to study cover type changes

The Penn Pilot web site provides historical aerial photos for most of Pennsylvania. You can either view the photos on the site or download them to use with other applications. Recently, in GIS class we used ArcGIS 9.2 to position the old aerial photos from 1937 over a current map of the southern portion of the Waynesboro Watershed using a process called georeferencing. Much of the area that is now the reservoir and its watershed was actively farmed. Here is an image of the map layout showing the farm fields outlined in green. (Click on the image to see a full screen version.)


When we remove the overlaid aerial photo we can see what the land looks like today with just the bouhdaries of the old fields. It’s evident that when forestry students in the 1950’s started working in the watershed they put the pine plantations on the old fields. This makes a lot of sense considering that it is much harder to put a plantation in existing forest.
historic_watershed2.jpgFor instructions of how to perform this operation click here. To use Penn Pilot click here.

Lessons Learned

The more I use ArcGIS the more I realize that it is a very complicated piece of software. You have to be careful to understand each function to deal with problems as they arise.

Here are a few tips I have learned the hard way:

  • Always make sure that the projection of the data frame is defined. This sets the stage for all new data that is added.
  • When exporting data from one data layer to a new data layer choose to put the new data in the data frame projection. This really helps when using data from the PASDA GIS server. It will also help when doing calculations.
  • Always try to put data in a geodatabase rather than leave it in a shape file. The geodatabases recalculate the shape area values by default.
  • When you want to determine the areas of polygons in a data table there is a calculate geometry option that lets you put the areas in any units, not just the data frame units.
  • Take advantage of the export map option to use a map document as an image file, like this map below.

  • Project_Boundary.jpg