NCTCOG Multi-year Building Footprints

Documentation

The NCTCOG Research and Information Services (RIS) department has developed a regional planimetric
building footprint layer using footprints digitized from aerials flown between 2007 and 2021.
Attribution includes land use data, zip code, footprint year and elevation data for those features that
have corresponding LiDAR or autocorrelated surface data.

Enhancement Data
The footprint layer can be joined with additional data to suit your analysis purposes. The NCTCOG
Regional Data Center has numerous layers (some of which have been listed below) that can used to
enhance the footprint layer.

To download data the data for free, visit: https://data-nctcoggis.opendata.arcgis.com/search

• Census
o Census Blocks (2020)
o Census Tracts (2020)
o Environmental Justice Index
o Urban Areas
• Transportation
o Mobility 2045 Update Level of Congestion
o Mobility 2045 Update High-Speed Rail
o Mobility 2045 Update Recommended Rail Transit
• Boundaries
o Texas Congressional Districts
o Independent School Districts
• Demographics
o 2021 Daytime Population (5-Year ACS)
o 2022 NCTCOG Population Estimates (City)
o 2045 NCTCOG Demographic Forecast (Block Group)
• Hydrology
o NHD Flowline
• Regions (HU2)
• Environment
o Parks (2020)
o Soils
• Features
o Development Monitoring in North Central Texas
o Features
o Employers
Additional potential data sources include:

NCT 9-1-1

US Census Bureau

Texas National Resources Information System (TNRIS)

Federal Emergency Management Association (FEMA)

Texas Department of Transportation (TXDOT)

Housing and Urban Development (HUD)

Methodology
The following steps were taken to ensure the most accurate data layer possible:

1. Aggregated all digitized footprint layers into one geodatabase and identified the most current
layer for all available areas.
 2. Merged latest footprints into layers by year and cleaned up overlaps and duplication. This
resulted in 9 individual layers, one for each year.
3. Ran adjacent tool on footprints that had been split previously using the NCTCOG index grid and
spliced them back together. This was so that each footprint was complete, not partial,
representation of the actual footprint.
4. Added a year field for all available years: 2007, 2009, 2011, 2013, 2015, 2017, 2019, 2020 &
2021.
5. Removed structures that were under 500 sq. ft. to cut down on carports and other structures
that aren’t truly “buildings”.
6. Added LiDAR elevation statistics to layers that had matching year LiDAR: 2015, 2017, 2018,
2019, 2020 and 2021using the following ESRI methodology:
a. Created 2ft buffer of footprint layer
b. Created <2ft buffer of footprint layer
c. Mosaiced DEM & DSM layers for each LIDAR year
d. Created random points (200) from the 2ft buffer and <2ft buffer layers
e. Add elevation value to random point files from DEM and DSM mosaic layer
f. Build statistics for all z values (minimum, maximum, median, mean, and standard
deviation) for both DEM and DSM random points layer
g. Joined each Z Statistics table back to DEM and DSM 2ft<2ft building layer
h. Created points from those 2ft DEM/<2ft DSM polygon layers and spatially joined the
points back to original building footprint layer
i. Calculate building volumes using elevation data
j. Calculate the building height based upon the difference between DSM and DEM
elevation value
k. QC percentage of building heights based upon known building heights from city
websites and city personnel and compare to VRICON data
7. Add elevation statistics to 2007 footprint layer from Auto-correlated Surface data using the
above steps.
8. Merge all building footprint layers together into one multi-year layer
9. Populate all footprints with land use information
10. Populate all footprints with zip code information
11. Create final schema for layer with the following data dictionary:

Field Alias Description

YEAR YEAR year
DEM_FREQ DEM_FREQUENCY count of points within buffered footprint at DEM level (bottom)
DEM_MEAN DEM_MEAN_Z average vertical measure (z-value) in DEM points
DEM_MIN DEM_MIN_Z minimum vertical measure (z-value) in DEM points
DEM_MAX DEM_MAX_Z maximum vertical measure (z-value) in DEM points
DEM_STD DEM_STD_Z std dev of vertical measures (z-values) in DEM points
DEM_MED DEM_MEDIAN_Z median vertical measure (z-value) in DEM points
DSM_FREQ DSM_FREQUENCY count of points within buffered footprint at DSM level (top)
DSM_MEAN DSM_MEAN_Z average vertical measure (z-value) in DSM points
 DSM_MIN DSM_MIN_Z minimum vertical measure (z-value) in DSM points
DSM_MAX DSM_MAX_Z maximum vertical measure (z-value) in DSM points
DSM_STD DSM_STD_Z std dev of vertical measures (z-values) in DSM points
DSM_MED DSM_MEDIAN_Z median vertical measure (z-value) in DSM points
MAX_DIFF MAX_DSM_DEM diff between max z-value of DSM and max z-value of DEM
MEAN_DIFF MEAN_DSM_DEM diff between mean z-value of DSM and mean z-value of DEM
MAX_VOL MAX_DSM_DEM_VOL volume using diff between max DSM and max DEM
MEAN_VOL MEAN_DSM_DEM_VOL volume using diff between mean DSM and mean DEM
AREA_SQFT AREA_SQFT footprint area (SQFT) calculated using field calculator
COG_LU COG_LU COG land use number
CATEGORY CATEGORY COG land use description
LU_YEAR LU_YEAR Land use year joined to footprint layer: 2010, 2015 or 2020
ZIP ZIP Zip code

12. Load final layer into new clean schema

13. Remove structures that are less that 500sft
14. Remove building height statistics that are < 4ft
15. QC final data layer

Important Notes

1. Planimetric footprints were manually drawn using 6” orthophotography so there may be some
human error in the data. After the initial creation, entities purchasing planimetric data either
opt to have their data updated or drawn new. The year of the data is either the year it was
drawn (if brand new) or most recently updated. For updates, the vendor adds new data and
removes/changes features that have been notably altered according to the latest
orthophotography. Most of the source data remains unaltered or adjusted in preparing these
layers. Any limitations or inaccuracies in the original data are likely to be reflected in these
layers, either partially or fully.
2. LiDAR is not captured by a sensor and is highly accurate. Due to the man-made nature of the
planimetric data and the machine-made nature of the LiDAR, there may be some discrepancy in
the building elevation if the footprint was not aligned properly. For this and other reasons, using
the footprint layer for macro, not micro level analysis, is recommended.
3. Auto-correlated surface data is similar to LiDAR data and is used to identify elevation and terrain
information. However, it is not as accurate as LiDAR, especially in heavily vegetated areas and
should be used for macro analysis only. For more information on ACS data vs. LiDAR, visit
https://www.asprs.org/a/conference-archive/ottawa07/ottawa_proceedings/16.pdf
4. These data were prepared using indirect methods for purposes of general information or
reference and are not a substitute for a detailed analysis of a particular area or structure based
on primary data sources such as engineering drawings, surveys, or field studies.
5. These data were prepared using standard methods as applied by the North Central Texas
Council of Governments. The resulting data are likely to differ from similar layers produced by
other sources using other data or other methods.
 6. While care has been taken to evaluate the overall quality of the data through multiple QC
procedures, no particular level of accuracy in either the source data or in the final product is
guaranteed. Like all data derived through application of various process and indirect methods,
these data are subject to a variety of issues and limitations. The user alone is qualified to
determine the usefulness and appropriateness of the data, including both the geography and
the attributes, for any given application.

Disclaimer of Warranty and Limitation of Liability

There are various ways to generate building footprints and gather associated attributes. The North
Central Texas Council of Governments (NCTCOG) has selected a method that facilitates a relatively
efficient and consistent process using resources available at the agency. The attributed building
footprints layers were developed for general, regional planning activities. They have not been evaluated
for other uses and are not for engineering purposes. They can differ from results of detailed analysis for
specific areas or from similar products derived through different processes. This document along with
the data referenced are presented “as is” without warranty of any kind, express or implied, including
warranties of merchantability and fitness for a particular purpose and no such warranties can be
assumed. NCTCOG makes no guarantee regarding accuracy, completeness, currency, or reliability.
Responsibility for the use of this document and any referenced data lies solely with the user. In no event
shall NCTCOG be liable for any claim of loss or other damages, either direct or indirect, arising from use
of data or any related materials generated or distributed by NCTCOG