The goal of this lab was to gather data from online sources and import the data into a single geodatabase. This was done largely through a Python script designed to project and clip the data while importing it into the geodatabase. The study area for this lab was Trempealeau County and is a continuation of the previous lab's focus on frac sand mining in the county.
The data came from a variety of online sources. This included DOT railroad data, USGS land cover and DEM data, USDA cropland data, Trempealeau County land records, and USDA NRCS soil data. The Trempealeau County land records contained a geodatabase for the county, which was used as the main geodatabse for the lab.
METHODS
The basic workflow is as follows:
First a workspace had to be created. Since a large amount of data would be downloaded, a temporary working folder was created to store the files until they were processed and put into a geodatabase.
Next, the data was downloaded. First, the USGS data was obtained. This was done by navigating to the USGS National Map Viewer, selecting Trempealeau County, and downloading the National Land Cover and DEM data. Next, the USDA website was used to download the land cover data. Next, the Trempeleau County Land Records geodatabase was downloaded from the website. This would serve as the main geodatabase for the project. Lastly, the USDA NRCS Web Soil Survey data was downloaded from the USDA website.
The data, all kept in the working folder, was then unzipped to allow it to be used. The geodatabase was moved to the project folder to serve as the main geodatabase. The USDA Soil Survey data was then imported to the geodatabase after joining the tabular data into the shapefile using a Microsoft Access macro. Next, the railroad data was processed. Since this shapefile expanded beyond the scope of Trempealeau County, it was clipped to the county and imported to the geodatabase. Importing it into the feature dataset automatically projected it into the correct Trempealeau County projection.
Next, a Python script was written to project, clip, and import the raster data into the geodatabase. This included the NLCD, the USGS DEM, and the NASS data. The Python script is shown in the blog post about scripting, under Script 1. This script was designed to loop through every raster in a given folder and project it, clip it to Trempealeau County, and import it into the geodatabase.
Lastly, the rasters were used to create a map of Trempealeau County in ArcMap. This is shown in the results section.
RESULTS
Shown below in Figure 1 is the map created from the raster files. The legend for the NASS and NLCD maps have a large number of categories, so they are shown separately below in Figure 2.
 |
| Figure 1: Map created from raster data for the county |
 |
| Figure 2: Legends for the map above |
Shown below in figure 3 is a table showing the data quality for each dataset used. This information was obtained from the metadata for each set. Notice there are many values enetered as "NA." This means the information could not be found in the metadata or this did not apply for this dataset.
 |
| Figure 3: Data quality for each dataset used |
CONCLUSION
This lab gave an introduction to basic data gathering from a variety of sources, as well as some Python scripting. The emphasis in this lab was data quality assessment. It is obviously important to know how to download data, but more important than that is being able to assess the quality of the data, otherwise results will be invalid.
No comments:
Post a Comment