Points, Figures, and the Survey That Won't Import Cleanly
A surveyor delivers a field book for a 12-acre parcel with 340 points, 22 figures, and a description key set. The points import, but the figures come in as disconnected line segments, and the point elevations are all zero. The description key set maps codes like TRE, FL, and EP to layers and styles. The field book uses a coordinate system in US survey feet, but the drawing is set to international feet. The surveyor's note says the elevations were reduced in the field and stored in the point's Z value, not in the description. A junior technician suggests re-importing with the description key set turned off to see if that fixes the figures.
1. Diagnose the three symptoms - zero elevations, broken figures, and the unit mismatch - and explain which are related and which are independent.
Read the worked answer
Zero elevations: the field book stores reduced elevations in the point Z value, but the import is reading a description-based elevation or defaulting to zero. This is an import setting issue - the point file format must map the Z column to elevation. It is independent of the figures and the units.
Broken figures: figures are built from point connectivity defined by the field book's figure commands (begin/continue/end) and the description key set's figure prefix mapping. If the description key set does not have figure prefixes configured, or the codes do not match, the figure commands are not recognized and each shot becomes an isolated segment. This is a description-key/field-book configuration issue, independent of elevation.
Unit mismatch: the drawing is in international feet while the field book is in US survey feet. The difference is about 2 parts per million - roughly 0.02 ft per 10,000 ft. It will not explain 340 zero elevations, but it will cause small systematic coordinate errors that matter for boundary work.
So: elevations and figures are two separate configuration failures; the unit mismatch is a third, smaller issue. Turning off the description key set would make the figures worse, not better.
2. The technician wants to disable the description key set. Explain why that is the wrong move and what you would do instead, step by step.
Read the worked answer
Disabling the description key set removes the very thing that assigns layers, styles, and - critically - figure prefixes to the codes. Without it, TRE, FL, and EP become plain text descriptions with no figure behavior, so the segments stay broken and you lose the layer/style mapping too. It treats a symptom by removing the mechanism.
Instead:
- Fix the point file format so the Z column maps to elevation. Re-import and confirm a sample of points shows the field-reduced elevations, not zero.
- Audit the description key set for figure prefix entries. Each figure code (TRE, FL, EP) needs a matching figure prefix so the begin/continue/end commands are honored.
- Check the field book's figure commands - confirm the surveyor used the expected syntax and that codes match exactly, including case.
- Reconcile units by setting the drawing's coordinate system to US survey feet, or transforming the import, before any boundary work.
- Re-import and verify that figures close and elevations are nonzero on a spot-check of 10 points.
Do the elevation fix first because it is the simplest to verify independently.
3. After a clean import, how would you confirm the figures and points are trustworthy enough to build a surface, and what would make you reject the data?
Read the worked answer
Confirmation steps:
- Closure check: each closed figure should return to its start point within a tight tolerance (a few hundredths of a foot). Open figures should connect end to end.
- Elevation range check: compare the imported Z values against the surveyor's field notes for a sample. A flat or zero range means the import failed.
- Unit check: measure a known boundary line and compare to the deed or survey record. A 2-ppm error is small but should be resolved before boundary reliance.
- Figure-to-point consistency: every figure vertex should coincide with a point; orphan vertices indicate a code mismatch.
- Duplicate and stray point check: look for coincident points with different elevations, which usually signal a re-shot or a coding error.
Reject the data if closed figures do not close within tolerance, if elevations are missing or implausible, if the unit mismatch is unresolved for boundary work, or if the questionable shots the crew flagged are not reconciled. Do not build a surface on unreconciled figures - the surface will inherit every error.