It has always been one of our goals to stay ahead of the coming changes at the NextGen 911 and FHWA level, so after much statewide coordination and feedback, AGRC is officially adopting a new statewide roads data model. The two main drivers for this update are the Next Generation 911 GIS requirements and the Federal Highway Administration’s All Roads Network Of Linear-referenced Data (ARNOLD) reporting requirements for state DOTs.

AGRC will officially adopt this new roads data model on September 13th, 2017. The purpose of this blog post is to give our users a heads-up on the coming changes and to allow the necessary time to adjust your internal workflows, beforehand. After that date, we will continue to provide downloadable data (and an SDE layer) in the older model, until March 21st, 2018. However, we encourage you to adopt this newer model sooner than later, as it will be the official statewide roads data model. There will be a subsequent blog post in September, before the official change.

Download the New Data

If you obtain the statewide roads layer from our website’s Roads and Highways System page, you will now notice the new data model available for download as Shapefile and File Geodatabase. Similarly, if you are accessing the roads layer via our SDE connection (only available inside the State network) you will temporarily notice a statewide layer named Roads_NextGen. This NextGen layer will replace the Roads layer on September 13th, 2017. At that time, the current SDE layer named Roads will be renamed to RoadsODM (RoadsOldDataModel). For both the downloadable data and the SDE layer, the older data model will be maintained and kept current until March 21, 2018.

Further Details of the ETL

Spatial data for this model was directly derived from the former data model. During the ETL, all spatial fields were reassigned via point-in-polygon queries (from SGID data layers), based on the segment’s midpoint. In order to capture the correct polygon boundaries for fields with right and left values (ie: ZIPCODE_R; ZIPCODE_L), a 50-foot offset (from the segment’s midpoint) was applied for each side. The FULLNAME field was also recalculated, based on a field concatenation formula. Read about the ETL process for a detailed look into our process, showing a side-by-side comparison of what fields were pushed where, as well as what fields are new or deprecated (note: see the sheet titled, SideBySideComp).

The Take Away

  • AGRC will officially adopt this new data model on September 13th, 2017
  • AGRC will maintain and publish both models until March 21sth, 2018
  • The new data model is based on the NextGen911 and FHA’s ARNOLD requirements

Please do not hesitate to contact me if you have any questions or concerns about the new data model or this transition. If you regularly you use this layer, we encourage you to adopt this new model within your own agency. Let me know if you are planning on doing so and are looking for guidance or assistance with the ETL process.

Follow AGRC on Twitter to stay up-to-date on this transition, as well as other GIS related stuff.