Lab 2 - Data Quality: Standards

In this week's lesson we learned about various spatial data quality standards. The two most widely used (and the ones we focused on) are National Map Accuracy Standards (NMAS), and National Standards for Spatial Data Accuracy (NSSDA). There are certain metrics that are used when measuring positional accuracy within these standards, for example: Root Mean Square Error (RMSE); 68th, 90th, and/or 95th percentile.

For our lab assignment, we were tasked to determine the horizontal positional accuracy and quality of data for two road network datasets using the NSSDA protocol. The basic steps to complete this accuracy assessment are as follows:
  1. Make certain that all datasets have the same projection.
  2. Using Network Analyst, build a basic network for each of the road network datasets to be able to generate the "junctions" of intersections.
  3. Manually select a minimum of 20 sample points at intersections on the two provided road network datasets, while using the created junctions as guides. Must select intersections that meet the following criteria:
    • minimum of 20% in each quadrant within the study area (can use the Create Fishnet tool)
    • greater than 10% of diameter apart
    • select "good" intersections (T-sections are great, however, no merging lanes, dead-ends, or cul de sacs)
  4. Create a new point feature class for what you believe to be the truest, most accurate intersection locations based on your visual interpretation using the provided aerial photography.
  5. Add XY coordinates to all 3 point shapefiles of intersections.
  6. Calculate the metrics using the horizontal accuracy statistic Excel worksheet, as provided in the Positional Accuracy Handbook.
  7. Prepare a formal NSSDA accuracy statement.
The screenshot below depicts the locations of my test point samplings, along with one of the road networks, and the equally-sized quadrants within the study area boundary.


My results were as follow:

City (ABQ) Accuracy Statement: Using the National Standard for Spatial Data Accuracy, the data set tested 15.08 feet horizontal accuracy at 95% confidence level.

StreetMap USA Accuracy Statement: Using the National Standard for Spatial Data Accuracy, the data set tested 10,322.04 feet horizontal accuracy at 95% confidence level.

Comments