NYCPlanning / data-engineering

Primary repository for NYC DCP's Data Engineering team
14 stars 0 forks source link

Peculiar Zoning Tax Lot QA results #753

Closed jackrosacker closed 1 month ago

jackrosacker commented 2 months ago

Summary

GIS Team ran zoning tax lot QA today and ran into four lots with abnormal results that may be worth inspecting. All four lots show tabular zoning changes, but no associated changes in the zoning overlay. In all four cases, the changes are an improvement, and increase the accuracy of the zoning values, but it is unclear how the erroneous data existed before without being caught, and why the latest ZTL/PLUTO build is adjusting the values now.

For example: BBL 2032440145 flipped the ZD1/ZD2 values, because previously the zd that covered the lesser portion of the lot was incorrectly listed as the ZD1.

Relevant Action run

Example data

bblnew zd1new zd2new zd3new zd4new co1new zd1prev zd2prev zd3prev zd4prev co1prev
1013980061 R8B C1-8X       R8B        
2032440145 M1-1 M3-1       M3-1 M1-1      
3035870001 R6         R6 R7-2     C2-4
4027740024 R5B R4-1     C1-2 R5B       C1-2

This is not urgent. We're passing ZTL QA since all four of these "errors" are improving data quality, but creating this issue anyway since they may indicate a blind spot in our process.

fvankrieken commented 2 months ago

Zola currently shows 2032440145 as more in an M3-1, implying that's likely a real change (at least from versions currently on zola - so from Pluto 24v1 at least, or January/Feb ZTL

image

fvankrieken commented 2 months ago

So a dtm lot change?

fvankrieken commented 2 months ago

Looks like it merged with the lot above it (sorry for ugly projection/colors in qgis)

image

fvankrieken commented 1 month ago

Due to issues around our publishing, latest was being compared to outdated ztl