Exploring Spatial Variability of Environmental Data with Digital Elevation Models

In this lab, we will explore the spatial variability of environmental data with digital elevation models (DEMs) as examples, examine the effects of a sampling strategy and spatial resolution on the data. You will also practice several spatial analysis tools that are routinely used in an environmental GIS project. 

Getting Data

Download the data from D2L and extract it to your desired directory (e.g., your USB drive). Once you unzip the data, you should have the following data layers: 

1.tlhned10: 3DEP Seamless DEM 1/3 Arc Second grid for the city of Tallahassee, Florida;

2.tlhned30: 3DEP Seamless DEM 1 Arc Second grid for the city of Tallahassee, Florida; 

3.tlhsrtm30: SRTM 1 arc sec Shuttle Radar Topography Mission for the city of Tallahassee, Florida; 

4.tlhsrtm90: SRTM 3 arc sec Shuttle Radar Topography Mission for the city of 

Running ArcGIS

1.Open ArcMap.

2.In this lab exercise, you will need to use a set of tools from ArcToolBox. Therefore, you need to turn on ArcToolbox (Geoprocessing > ArcToolbox) after you run ArcMap.

3.You also need to turn on the Spatial Analyst (Customize > Extensions) extension.

Step 1: Compare the datasets visually

A.Display the four data layers with ArcMap. Spend a few minutes to examine the four different DEM datasets (their projection, unit, pixel size, symbolization, or whichever appropriate...). 

DEM data are very commonly used for environmental GIS analysis. There are multiple sources DEM data available for mapping, modeling and analysis. It is therefore critical to understanding their differences in order to choose the appropriate datasets for your GIS project.

In this lab, we will explore and compare two major types of DEM data: the 3D Elevation Program (3DEP) and the Shuttle Radar Topography Mission (SRTM). Note that the two types of DEM data are generated from different sources through different sampling strategies. 

Step 2: Resample the datasets 

Use Resample (ArcToolBox > Data Management Tools > Raster > Raster Processing > Resample) tool to aggregate EACH of the datasets to 200m cell size. Make sure that each output should be given an appropriate name, and the output cell size should be 200 (X) by 200 (Y) (using the default method NEAREST for resampling; do you know the differences among the three resample methods? If you are not sure, please check ArcGIS HELP)

Step 3: Derive contour maps from the original data

Use Contour utility (ArcToolbox > Spatial Analyst Tools > Surface > Contour) to create a contour map from EACH original data set. Make sure that each output should be given an appropriate name, and use the same contour interval of 10 (m). Leave the other options as default (take a look at the Help area so that you understand what each option defines).

Step 4: Derive contour maps from the aggregated data

Use Contour again to create a contour map from EACH resampled data set (produced from Step 2). Make sure that each output should be given an appropriate name, and the contour interval should be at 10 (m). 

Questions:

1.With your own words, describe some major visual differences between these datasets (Hint: try to compare a few pairs of images such as tlhned10 versus tlhned30, tlhsrtm30 versus tlhsrtm90, and tlhned30 versus tlhsrtm30). Do these differences result from the differences in sampling strategies or resolution or both? Why?

2.Comparing these outputs (visually and statistically –go to layer properties/source, you will see a statistical summary for each layer; compare several pairs as you did at last step), what are the differences due to? (Hint: Please link Step1 and question 1 with the current question; consider how sampling strategy and resolution affect information contents.)

3.Comparing these contour outputs (Hint: using several pairs as you did before; you are encouraged to make a few maps to illustrate the differences; the maps do not have to be fancy but appropriate symbolization (color and size, ...) will help the comparison). What are the differences due to? (Hint: Please link the previous steps and questions with the current question)

4.Comparing these contour outputs (using several pairs as you did before; you are encouraged to make a few maps to illustrate the differences). What are the differences due to? (Hint: Please link the previous steps and questions with the current question).

5.Comparing the contour maps derived from the original versus resampled data (You can arrange a few appropriate pairs for comparisons; you are encouraged to make a few maps to illustrate the differences). What observations can be drawn through these comparisons?

6.By working through this lab exercise, you should have a good sense about the spatial variability of environmental data and how sampling strategy and resolution affect data. Write a few paragraphs to discuss your insights and their implications for GISbased environmental analysis.