Lab 2: Raster Modeling in ArcGIS

Introduction

 This lab teaches you how to do raster spatial analysis and introduces map algebra.  Put your lab data on your workspace on the network drive (e.g., Z:). If unsure, ask the TA how to map your network drive or visit this webpage. Because you will use Arc/Info tools to process the coverage data, make sure there is no "space" in the file path (e.g., "My Documents" is not allowed). Answer the questions at the end of the instructions of each exercise and produce the requested outputs.  Return them to the TA by the due date.  Labs should be typed, include your name, be well organized, and be stapled together.

Instructions

Skim Chapters 4, 5, and 6 of the Using ArcGIS Spatial Analyst pdf (Using_Spatial_Analyst.pdf) on the I drive  (The pdf file can be found in the I:\Students\Data\GIS\ArcGIS Documentation\ArcGIS9.1_documentation\ESRI_Library\ArcGIS_Extensions folder.). Read Chapter 7, do the following tasks, and produce maps as specified in each task. Make sure you save your results from each question.

Log onto the Oregon Geospatial Enterprise Office web site and click on "Spatial Data Library Alphalist". There are six columns in the table. The E00 file is an ArcInfo interchange file (also called an ArcInfo Export file).  ShapeFile is an ArcView shapefile.  GRID is an ArcInfo raster data file.  Metadata is descriptive information about the data.

Set up for Spatial Analyst Exercises:

1) Download data with the specified format, and unzip them:

* Click on the Shapefile for County Boundaries- Oregon (2007 BLM 1:24,000).

* Save it to disk in a location that you have access to. 

* Navigate to the file using Window's Explorer. 

* Right-click on the zip file and point to 7-zip to extract (unzip) its contents. 

* Repeat this process to download and unzip the following E00 files:

o Cities (USGS, 1:2,000,000),

o Highways (USGS, 1:2,000,000),

o State Boundary (USGS, 1:2,000,000), and

o Vegetation/Species (Oregon GAP vegetation, 1:250,000).

o Population (USGS). Use this link to download the population data.

2) Convert the E00 files into ArcInfo coverages.  

* To avoid confusion, it's best to save these coverages in a new folder, separate from the files that you unzipped. Make sure there is no space in the file path name.

* Open ArcToolbox, and search for Import from E00 and then choose the first E00 file and an output name and location for the coverage.  

* Import each of the other files into ArcInfo coverages in the same way.

3) Add data to ArcMap

* Start ArcMap and open a new empty view. 

* Click the -ADD Data- button and add all of the ArcInfo coverages and shapefiles to your map by double-clicking on them individually and choosing a feature class. 

* For each coverage, you will specify the feature class that you want to add to the map (coverages and geodatabases may have multiple feature classes while shape files only have one type of feature per data set).  The feature classes used by this exercise are:

o Highways - arc,

o State Boundary - add as a polygon shapefile,

o Counties - polygon,

o Population - point,

o Vegetation/Species - polygon,

o Cities - point. 

* Once all of the data layers are added to the map, right-click on the data frame properties at the top of the table of contents (named 'layers').  

* Click on the general tab and choose feet for the map and display units. If these fields are grayed out, then first set the linear unit to foot under the coordinate system tab and modify button - hit "Apply", then set the units under the general tab.

4) Projection information - Can check this in the Coordinate System tab under Layer Properties:

The projection information was imported with the coverage. If any data you brought in doesn't have its projection information defined, then you would have had to put this information in manually (using the Define Projection tool). The projection parameters are as follows:

Projection: Lambert Conformal Conic
Datum NAD83
False_Easting: 1312336.0
False_Northing: 0.0
Central_Meridian: -120.5
Standard_Parallel_1: 43.0
Standard_Parallel_2: 45.5
Scale Factor: 1.0
Latitude_Of_Origin: 41.75

5) Now set the Spatial Analysis options as follows - Under the Geoprocessing menu, click on Environment settings and change the following:

* Processing Extent: use the Oregon State Outline file

* Under Raster Analysis, Set Cell Size to "As Specified Below" from the dropdown menu, enter 5000 in the text box below

These two must be done before you begin your raster processing or you will get incorrect results!

You will now do several spatial analyses using the Spatial Analyst Toolbox:

* To make Spatial Analyst raster layers permanent, be sure to check the output location and name for the result raster in each tool.

* For all the maps you will submit and be graded on the following...

o Make sure that the SA# is part of their title

o Identify the tool used for each map in a text box or with the title/subtitle

o All maps must be in the shape of Oregon. This may be done many ways. Here is one - Clip to Shape

* Right-click on the dataframe and select Properties

* Select the Data Frame tab

* In the bottom section, under 'Clip Options' select Clip to Shape from the dropdown menu

* Click 'Specify Shape'

* Select 'Outline of Features'

* From the dropdown menu choose the Oregon state boundary layer

* Click OK, the click OK again

o Change the symbology of the state boundary to be hollow with a 2 point outline.

o Change the symbology of the counties to be hollow with a .4 point outline.

o Make sure the correct datasets are displayed with your raster results to make them understandable - not all maps need all the data!!

SA #1: Calculating Distance from Highways using the Euclidean Distance Tool

* Under the Spatial Analyst toolbox choose Euclidean Distance.

* Use the highway coverage as the Input source.

* Make a map of the result, add your name using a text box, and print it.

SA #2: Allocating Space using the Allocation Tool

* Under the Spatial Analyst toolbox choose Distance, Euclidean Allocation.

* Use Cities as the Input source

Make a map of the result and print it with your name.

* Using a text box describe the map in your own words directly on the map.

SA #3: Calculating Population Density using Density Functions

* Under the Spatial Analyst toolbox choose Kernel Density

* Use the population layer as the input data

* Set Population Field to POP85

* Set Search Radius to 100,000
Make a map of the result and print it with your name.  

* Include a legend that shows the numerical values for the population density.

* Using a text box describe what the units mean, directly on the map.

SA #4: Highway Accessibility by County Using Zonal Statistics

* Under the Spatial Analyst toolbox choose Zonal, then Zonal Statistics,

* Set Input source to the county shapefile,

* Set the Zone Field to alt name

* Set Value Raster to the result of SA#1 (Euclidean Distance From Highways)
Make a choropleth map with a classified legend using the zone mean statistic

* Use a text box describe what the units mean, directly on the map.

SA #5: Rasterize a Vector File and Use it To Calculate Vegetation Diversity Using a Neighborhood Statistic

* Convert the vegetation coverage to raster.

o Under the Conversion toolbox choose To Raster, Feature to Raster,

o Set Input Features to Vegetation,

o Set Field to com_code

o Set Output cell size to 5000

* Calculate the vegetation diversity in a neighborhood surrounding each cell.

o Under the Spatial Analyst toolbox choose Neighborhood, Focal Statistics

o Set Input raster to the vegetation raster you just made

o Make sure the Neighborhood is a rectangle (3x3) and the units are Cell

o Set Statistic Type to variety,

* Make a map with a legend of the result.

* Using a text box, describe what the units mean and why they have the range they do

SA #6: Reclassify the Raster Vegetation Variety Neighborhood Surface into Two Classes.

* Under the Spatial Analyst toolbox choose Reclass, Reclassify

* Set Input raster to the variety surface created in SA#5,

* Set Reclass Field to value,

* Click Classify

* Set Method to equal interval,

* Set the number of Classes to 2, click OK

Note the Old Values and New Values now listed.

* Make a map of the result.  

SA #7: Using the Raster Calculator to Change Cell Values (to m from feet), the Reclassification Tool to Reduce the Number of Classes, and Vectorization

* Under the Spatial Analyst toolbox choose Map Algebra, Raster Calculator.

* Double click on the distance to highways raster created in SA#1, click on " / ", type 3.28, and click OK.  

* Reclassify the result into 3 natural breaks classes.  

* Convert the raster to a vector by choosing the Conversion Toolbox, From Raster, Raster to Polygon (it doesn't matter which field you use for this operation).  

* Make a map of the result that symbolizes based on GRIDCODE and include a legend

SA #8: Implementing Map Algebra Commands Within the Raster Calculator
First review Appendix A of the tutorial book.  The Spatial Analyst Raster Calculator allows you to use all ArcInfo MapAlgebra GRID commands and functions.  Learning map algebra is like learning any language.  Once you know the syntax then you need to increase your vocabulary.  In this exercise you will implement one map algebra function using the Raster Calculator.  

* First rasterize the highway vector coverage (using the Highway ID field) that you downloaded and call it HW.  

* Then choose from the Start Menu: All Programs -> GIS -> ArcInfo Workstation -> ArcDoc

* Select Contents -> Command Reference for ArcInfo Extensions -> GRID -> Alphabetical List of Grid Functions. This is a comprehensive help system for ArcInfo Map Algebra. Read the help topic for Thin.  

* Open raster calculator and type: Thin ("HW") and then click OK.  

If it works then you have implemented your first MapAlgebra command that is not available through buttons or toolbox commands within Spatial Analyst.  Make a map of the result and include a description of the result in a text box as well as your name.