SO432, Geographical Information Systems

Exam 1, Fall 2012



10@ 4 points



Short answers

 6 @ 10 points









Read the directions carefully.  You have a selection of questions for the definitions and the short answers.


Quality of your answers is important.   For full credit you should use correct terminology, and show that you understand the concepts involved.  Demonstration of understanding, and placing you answer in the context of GIS, is much more important than finding a random sentence from the book describing or defining the term, which will be an average answer.


All work on this exam is individual.  You may not any materials (books, notes, computers, calculators), and you may not use IM, texting, talking, or any other means to communicate with other individuals.


Red comments are not complete answers, but points for the review after the exam which is a continuation of the learning process.  They are important but not sufficient for full credit.


Definitions:  define 10 of the 11 terms with a concise sentence that clearly shows your understanding of the term.  Each is worth 4 points.


1.  Buffering: while we did this for a point, it can also be done for area or lines.  Examples would be critical environmental zones along a river or the Chesapeake shoreline.
2.  Database filtering: write a query, and get only those records which match the desired characteristics.  Query is boolean.
3.  DEM: gridded data

4.  Gazetteer: vector data

5.  Integer versus string

6.  Large scale map

7.  Meades ranch: half of class skipped ("Dorothy, we're not in Kansas anymore" but use WGS84)

8.  NGA

9.  Reproject data: e.g. what we did with the NLCD data, taking it from conic coordinates to geographic

10.  UTM zone:  UTM is not Mercator, but a transverse Mercator (cylinder turned on its side) and it has 60 zones, each 6º wide.

11.  WGS84


Insure that you left one blank.  There is no extra credit for answering more than 10.

The next five pages contain 5 questions, each of which is worth 12 points.  You can answer on the page or on the back of one of the pages. You must answer all 5.  For each answer you should clearly show that you understand the principles and use appropriate terminology.



I downloaded some NLCD data, and got all these files. 

·        What is NLCD, and what does it provide for me?

·        Is this vector or raster data, and which file(s) contain the actual data?  How many files must I send someone so they can display this data?

·        Do you have any idea what the files with “meta” in the name mean?




I downloaded some TIGER data for the blocks in my state, and got all these files. 

·        Can you tell from this information if this is your state?  How?

·        Which of these file(s) are required to display the data?

·        What is the name of this format, and is it a raster or a vector data set?

·        Do you have any idea what the file with the XML extension contains?




One of your classmates has two databases, one with the tracts for the state, and other with earnings for the tracts in one of the counties.  The join is set up above, but is not working.


What is your classmate trying to do with the join, and why do you think it’s not working?  Is there any hope other than redownloading the data?


Two databases, one with the geometry (shapefile), and the other with just data, but they are joined by fields with the ID of the tracts.

Length of the entries in the two fields must match, but the field names do not.  In this case the GEOID field has the state and county FIPS codes included, but the TRACTCE does not.  You could edit out the 24003 from the GEOID, or your get another field in the shapefile.



These are two versions of an NLCD map, with a legend showing the percentage of the map in each classification category.  The map on the top has been clipped to show just the city of Annapolis.

·        Why would some one perform the clipping operation, and is it a good idea?

·        One of the maps has a UTM grid, and the other a lat/long graticule.  Which is which, and how could you tell?

·        What projection does the data have, and how can you know? 


Gridded datasets have a projection, which gives the coordinates for the grid nodes at which the values are created.  You should be looking at the shapes of the "boxes" in the grid/graticule, and the orientation of the lines with respect to the edges of the map.  Unless you were to explicitly specify a different projection, the GIS program will use the projection of the data in the first grid you load for the map.



The map of Annapolis above has a series of red and black lines one it.


Are they:

·        Two different lat/long graticules?

·        Two different UTM grids?

·        One graticule, and one UTM grid?

Defend how you knew which case was correct, and why understanding that is important for GIS.


From the orientation of the lines, what can you infer about the projection of the map?


Datum shift on a conical map projection.


These four maps show two areas (they show the same area on each row, with different overlays).

·        Why do the maps on the left have three scalebars, valid for the top, middle, and bottom?  Which needs the three more, and why would you say this?

·        What is the purpose of the circles on the two maps on the right?

·        Why does the lat/long graticule look so different on the two maps?

·        Could the two regions use the same projection?  Why or why not.

·        Are these good projections?  Why or why not.




Mercator projection, which is conformal.  Scale distortion worse at high latitudes, and changes rapidly there as well. In high latitudes, a 1ºx1º block will be extremely tall and narrow, so it is hard to show very many parallels without having too many meridians.