Lexington, KY Richmond Road/Athens-Boonesboro Road improvements, part 2
Posted by Isaac Rowe on 20 August 2024 in English.Read part 1 here.
I drive through the new RCUT intersection again today, this time making a GPS trace of my movements. I attempted to make each of the possible turns in the configuration. I did have to make one illegal maneuver, as the left turn north from Athens-Boonesboro does not go anywhere, and, while the light is installed and the turn lane unblocked, there are signs indicating that no turn is possible from that lane. The light is (as far as I can tell), in a semi-permanent red arrow state until the side road is developed. Since the turn lane is divided from the main road by a concrete island, the only way out is to make an impermissible left turn or unsafely reverse back into traffic. I chose to make the left turn. Cones or bollards are probably warranted to keep motorists from accessing the left turn lane to nowhere.
Based on this trace, I attempted to align the nodes for the separated left turn and U-turn lanes to the correct positions. Since the lane begins further back from where the dividing island begins, there is some ambiguity on when to split the lane into a separate line; I chose to break it off roughly where the dividing island begins. This makes for a less visually appealing line, but it does semantically indicate where lane changes are still technically possible.
This matches the approach I saw elsewhere on OSM, in Troy, MI. Apple Maps does the same thing (possibly with OSM data), while Bing Maps maps the entire turn lane as separate. Google Maps has a very interesting mix where one direction breaks away at the dividing island, but the other direction breaks away much further back in the turn lane. Google Street-View imagery from November 2021 shows bollards dividing the entire turn lane, so their system must’ve used that as the dividing point . Bing Maps imagery collected in April 2022 do not show these dividers.