MDI Lab Procedures

This ArcView lab is intended to acquaint you with the scientific method and hypothesis testing, learning several important GIS procedures, managing the data of a project, and how to deal with the stress when procedures don't go as planned.

Background Information

MT. Desert Island contains numerous granite mountains, the tallest being just over 1500 feet tall with a base at sealevel. Forest cover is a mixture of broadleaf deciduous and needleleaf evergreen. Typically, broadleaf deciduous trees are found in warmer mid-latitude climates such as the Eastern Deciduous Forest of North America that spans the Humid Subtropical and Humid Continental climate types. Needleleaf evergreen trees tend to dominate harsher climates such as Subarctic found with the Boreal Forest or Taiga.

The microclimate can vary significantly from south-facing slopes to north-facing slopes. South-facing slopes tend to be warmer, have higher rates of evapotranspiration and thus are usually drier. North-facing slopes tend to be cooler and wetter.


The forest cover on north- and south-facing slopes will differ significantly.

General Procedure

You will have two data sets to work with. One is a grid (raster) data set (mdi_elev25) of elevations of the island with a 25 foot contour interval. The second is a vector data set (shapefile) that contains land cover (primarily vegetation, Veg79) classes for the island. The original data can be found in /scratch/gt202/mdigis. You will make a copy of the mdi_elev25 grid and veg79 data sets to analyze. Using the elevation data you'll derive a data set of the aspect (north-facing, east-facing, etc) of the slopes and using Map Query, separate out first, the south-facing slopes and then the north-facing slopes. The veg79 data is in vector format while the elevation data is in raster (grid) format. The data sets must be in the same format to be compared. You will convert the raster slope data to vector (shapefile) format and determine what vegetation types are found on the south-facing slopes and north-facing slopes.

To compare the two derived data sets to see if they are, indeed, different, you will perform a statistics test call Chi Square using MS Excel.


Specific ArcView 3.1 Procedures

Everyone should complete this lab using ArcView 3.1 on the PC's.

Everyone needs to make a directory (called your username) in /scratch/gt202/mdistudentlab/ to contain your lab data. Make a subdirectory called mdi1 and in mdi1 make two subdirectories called  veg79 and another called tmp. If you want to begin the lab again make another directory called /scratch/gt202/mdistudentlab/username/mdi2 with the appropriate subdirectories. I want you all to learn how to keep all of your data in one directory (with multiple subdirectories) separate from all other data.

Start ArcView 3.1 on the PC. Open a blank project. Make sure the following extensions are set;
    - geoprocessing
    - spatial analyst

Note: Words on boldface with arrows between indicate the ArcView menu items to navigate to.

Check this by making the Project window active and going to File--->Extensions and scrolling down the list.

Open a new view and make it active.

Next, set the working directory to your project directory (File---> Set Working Directory) on the Sun fileserver (S:\gt202\mdistudentlab\username\mdi1).

If  I was logged onto a Sun, the path to my project directory would be; /scratch/gt202/mdistudentlab/tbw/mdi1

If I am working on a PC the path to this same directory would be; S:\gt202\mdistudentlab\tbw\mdi1
Why is the path different on the Suns and PCs? The PC's mount the remote Sun file systems as disk drives. The /scratch disk on the Suns should be  mounted on the S: disk drive on all of the PC's. If it isn't, open Windows Explorer and click on the Map Network Drive button and map everest:/scratch to S:. See me for help on this if you have a problem.

To copy the mdi_elev25 grid to your project directory do the following;

View--->Add Theme (**** set  Grid Data Source ****) /scratch/mdigis
Theme--->Convert to Grid to save in your local directory. Use same theme name with a .grd extension. Delete original grid theme. Save project.

To copy the veg79 vector/shapefile to your project directory do the following;

View--->Add Theme (set  Feature Data Source) /scratch/mdigis/covs/veg/veg79. Theme--->Convert to Shapefile to save in your local veg79 directory. Use same theme name with a .shp extension. Delete original theme. Save project.

The veg79 data set has 25 classes. The default legend is not too readable. Read in a better legend by doing the following;

Make veg79 active, bring up legend editor. Load .avl file in /scratch/gt202/mdigis/legend_veg.avl. When load legend dialog box appears take default values. Save legend in your directory with same name.

Using the elevation data, derive the aspect (whether a slope is facing N, NE, E, SE, S, SW, W, or NW),

Make mdi_elev25 active and Surface–Derive Aspect.

The aspect grid you produce includes slopes of all 8 compass points. We are interested in just the north-facing and south-facing slopes. You need to use the Map Query function to extract, first, the south-facing and then north-facing slopes.

Analysis–Map Query for south-facing slope and then north-facing slope. The resulting table will include values that meet the query (values = 1) and values that do not meet the query (value = 0). You want to create a grid of only the true (1) values and convert to grid and then convert the grid (raster data) to a shapefile (vector data).

Open table of map query 1, make active and Table--->Query for true values. Make view active and Theme--->Convert to Grid. Call the north grid northaspgrd and south grid southaspgrd. Delete Map Query Theme. Theme--->Covert to Shapefile, convert northaspgrd and southaspgrd to shapefiles. Use names northasp.shp and southasp.shp.

Next you will take the southasp.shp vector theme and intersect it with the veg79 vector theme. Then the northasp.shp vector theme and intersect it with the veg79 vector theme. View—GeoProcessing Wizard, Intersect two themes, northasp.shp and veg79.shp and then southasp.shp and veg79.shp. Make sure the output file is set to your project directory. The default name for an intersection is intsct1.shp. I suggest you change the name to something like int_northasp_veg.shp. Be descriptive....... when you're working with a lot of themes it keeps you from gettng confused.

Open table of each intersection keep it active and export in dBASE format ( File--->Export ) to open in MS Excel. Use a file name such as table_int_southasp_veg.dbf.

The statistical analysis will be done using either MS Excel or SPSS. I will give a lecture soon on the procedure.



Assume that you will not do any GIS for one year and then will have to familiarize yourself with all aspects of this lab. Prepare a writeup of all procedures and what they mean. Prepare map(s) of your work with appropriate explanations and post to your web site. This final posting should be professional in its execution.