Mapping, GIS, and Spatial Data
This domain tests spatial data accuracy, datums, and the limits of GIS as boundary evidence. The exam expects you to read an accuracy statement critically and to choose a projection appropriate to the measurement.
Accuracy standards
- Horizontal positional accuracy is commonly reported as RMSE and converted to a 95% confidence value using 1.7308 × RMSE_r. Vertical 95% accuracy uses 1.96 × RMSE_z.
- Report the statistic you actually computed: RMSE is not the same as 95% accuracy.
- Metadata, per FGDC or ISO 19115, records source, datum, accuracy, and processing.
Projections and data models
- State Plane coordinates are appropriate for accurate distance and area computation in DC. Web Mercator preserves shape for web display but distorts distance and area and is unsuitable for measurement.
- A parcel fabric is a cadastral index, not a survey. The recorded plat and deed control.
- Topology finds gaps, overlaps, and dangling edges; it does not validate legal boundaries.
DC data context
- DC GIS layers are maintained through OCTO, including the Master Address Repository for addresses. Address points, orthoimagery, LiDAR, and parcel layers are reference data.
- Orthoimagery has its own accuracy and offset characteristics, especially near tall buildings.
Worked example
Four checkpoints give radial residuals of 0.4, 0.6, 0.3, and 0.5 ft. RMSE_r = √[(0.16 + 0.36 + 0.09 + 0.25)/4] = √0.215 = 0.464 ft. Horizontal 95% accuracy = 1.7308 × 0.464 = 0.80 ft. The dataset meets a 1-ft 95% specification, but quoting 0.46 ft as the 95% value would understate the error.
Traps that cost marks
- Presenting RMSE as 95% accuracy.
- Measuring area or distance in Web Mercator.
- Treating a GIS parcel layer as a boundary survey.
- Mixing NAD27 and NAD83 without transformation.
- Ignoring metadata or assuming orthoimagery is free of building lean.