A new modeling study performed by UC Davis shows in fact many of those fears are simply unfounded. They find capacity with a two lane road unchanged and throughput improving significantly in the westbound direction and only slightly slower in the eastbound direction–seven seconds added for the entire drive from L Street to A Street. The results of this model also show other improvements that lend themselves well to other goals in the city.
Fehr and Peers Study From Five Years Ago
The first problem according to Steve Tracy, a local resident in Old North Davis with professional expertise in street design, was a misuse of the information from the model run. While this study used a fixed model which is common in corridor studies, one must be careful not to use the outputs from that type of model as if they were from an equilibrium model. That type of model is more realistic in some situations, because it better approximates how traffic actually flows through a traffic system.
Said Mr. Tracy:
“An equilibrium model is like water–if you back it up in any one channel it will overflow and go find other ways to get downhill. If you keep backing up all the channels, eventually you reach a kind of equilibrium state where none of the water can get downhill and it’s all a level playing field. That’s the way the real traffic system works.”
Under these conditions, drivers experiment until they find the best route to their destination. In a modeling system, it seeks an equilibrium point, a point “where no driver within the model can improve their travel time to their destination with a different route.”
However, the Fehr and Peers study used a fixed model based on rigid traffic counts. Every car is counted as it enters the corridor whether through Fifth Street or one of the side streets and every movement at the intersection is counted.
“The modeling exercise that’s conducted from that point on is absolutely rigid. Those drivers may not reassign themselves to such streets that might work better.”
This fixed type of model was appropriately used to study traffic flow on the main corridor, in this case Fifth Street. But is was not appropriate to assume that back up on the side streets would occur in the real world. Steve Tracy suggested that some of the misrepresentation of the model outputs may have been intentional.
“Part of that back up was due to the fact that they had manipulate the inputs on the model, making drivers in the computer model unwilling to merge into a gap in traffic that most of us would take. That backed up the very low percentage of the cars that were on the side streets at the non-signalized intersections.”
Thus cars were not willing to merge between gaps in traffic that in the real world, cars would take. Moreover the rigidity of the model prevents drivers from seeking better routes.
“In the real world, they are going to try another street, preferably one with a signal where there is plenty of capacity and it will work better. As soon as they find a path that works for them, they’ll give up on E Street. But this was a rigid model so those cars could not go anywhere else. To call that out as a flaw in the proposed design is an absolute misuse of the outputs of the model.”
Another flaw with the study is that when they originally ran the vehicle counts, they did so on a Wednesday evening during Farmer’s Market. As Mr. Tracy described, this was an absolute violation of the protocols in the traffic engineering manuals. Those say counts should not be taken during a special event unless traffic from that event is the sole purpose of the model exercise. Later, additional counts were taken on a different evening.
New UC Davis Model
This study has vast improvements over the previous study. It starts by taking the traffic counts used in Fehr and Peers work. However, one of the key flaws of the Fehr and Peers work was also that they did not do bicycle counts, per instructions from City staff. “At great expense we had guys in lawn chairs with clickers counting vehicles, and they were instructed to ignore bicycles when bike lanes were a key component of this whole exercise.”
Without accurate bicycle counts to reply upon, Feng and Ying simply made the assumption that the bicycle count would comprise ten percent of the vehicle count.
In the Fehr and Peers model, even when bicycle lanes were present, they had right-turning motor vehicles slowing down in the through traffic lane at the intersections, and then slowly turning right to depart the street. This causes unrealistic back up and slowdown on the streets.
“The UC Davis model is much more realistic in terms of how cars behave–they behave like they do in Davis and like you are required to under state law.”
Steve Tracy continued:
“The motor vehicles in this model, unlike Fehr and Peers first pass at this, do merge into the bike lane before they turn right. In fact you are required under the California Motor Vehicle code to do just that if there is a bike lane present to clear the traffic lane to through traffic.”
The UCD model overall is much more realistic in terms of how it deals with the motor vehicle flow and the obstacles that a vehicle would find through the corridor.
“If the random assignment of bicycles through the corridor has the bicycles block the turn lane, then the cars have to wait in the UCD model. So the presence of bicycles impedes vehicle flow in this model which again is realistic, it’s how we have to do it in Davis.”
Results
“An eastbound trip during the peak hour is slightly slower–it’s a matter of a few seconds. Average travel time for through vehicles is 144 seconds with the existing street and 151 for the other one. So it’s seven seconds slower in a two-and-a-half minute trip.”
However, that slight difference is more than made up by the vast difference in the westbound traffic flow which is about one minute quicker with the two-lane street with turn pockets.
Said Mr. Tracy:
“Feng and Ying have tried to play with the signal timing to adjust that, and that’s about as close as they can get in difference. They couldn’t ever quite get the eastbound flow to be better with the two lane street.”
However, given the fact that it is only seven seconds slower and the westbound flow is far improved, overall the average travel time through the corridor is a full 20 to 25 seconds faster. And speed of course is not the only consideration. Remember the purpose of the redesign is not just speed, but capacity, traffic flow, safety, and also the ability to accommodate bicyclists and pedestrians. So if they can get the traffic flow to be comparable to the current design, the other factors more than outweigh the very small eastbound slowdown.
As you can see from this chart, flow is improved as the number of stops reduces.
Here are the modeling simulations, the first shows the status quo and the second is the redesign.
Status Quo Model
Fifth Street Redesign
One of the key features in redesigning the eastbound direction is how the intersection at B Street redistributes the flow. Currently, according to the traffic counts, half the traffic headed eastbound in the evening rush leaves Fifth Street at B Street. Of those vehicles, 43% go south on B Street towards the freeway while seven percent head north towards 8th Street. This is the logical point to reduce the through lane count from two in each direction to one.
The redesign would make the right lane into a dedicated right turn lane with a sign that indicates it is a right turn only. According to Steve Tracy, this represents another problem with the Fehr and Peers model where they failed to account or adjust for lane preference. Thus many of the cars would enter Fifth Street in the wrong lane and change lanes at the last second. That’s not a realistic assumption about how traffic actually flows.
“If you stand at A Street and watch, you’ll see people preselecting the lane that they want to be in. They dialed that down on the model. So that cars didn’t come into the model at A Street with most of them in the lane that they wanted to be in. You could see in the model and in the outputs that that’s what was going on. Half of the drivers between University Avenue and B Street were trying to get into the other lane because suddenly they realized they were in the wrong lane. In the Fehr and Peers model that created a lot of back up like this irrespective of the signal or what’s going on.”
In the real world, if you enter Fifth Street from the campus, you are going to stay in the right lane if you know you want to turn right at B Street. Of course, on your first or second trip, you may not realize that you have to do this, but this again goes back to the fluid model where actual drivers adjust their route to optimize their travel time.
He said, “Even in a fixed model like this one, you can set that lane preference so that cars turning left are in the left lane and cars turning right are in the right lane as they cross A Street.”
Greenhouse Gas Reduction
Steve Tracy argues one of the huge benefits of the Fifth Street Redesign as shown in the model results is a drastic reduction of idling cars. These are cars that are basically sitting in traffic, motor running, and not moving. He said, “That’s pure greenhouse gas emissions to no benefit at all.”
In this model, that drops 50% in the two lane street.
He argues that unlike theoretical programs that cities employ where there’s kind of a tenuous link to what we’re going to do, this is a very real reduction in GHG. “Here’s a thing that’s not tenuous at all, it’s directly attacking the amount of time idling cars are sitting stopped in traffic.”
Steve Tracy wanted very much to thank Dr. Michael Xing along with Ying Zeng and Feng Xio for their work.
“This is a $100,000 gift to the community. That’s what was poured into the Fehr and Peers study five years ago.”
He continued:
“These guys gave us a more state of the art, more accurate model. It shows that there’s no negative impact on traffic volumes.”
Several weeks ago, the DDBA President Jennifer Anderson presented the Davis City Council with a petition of 400 signatures from businesses opposing the Fifth Street Redesign. Supporters of the proposed design, which is from the adopted General Plan, have a petition that has been circulated just over a week and now has double that number of signatures.
- If you want to sign that petition, click here.
—David M. Greenwald reporting
I have lived and worked 24/7 on the 400 block of E/D Streets since 1995, and I strongly believe that the proposals by Steve Tracy will greatly benefit residents and businesses alike.
Tried to sign the petition, but can’t unless I donate money. Hmmmmmmm….
Anon. Yes, you will be able to sign without making a donation. Many of us were also stalled at first by that feature of the website that hosts the petition. Just submit your signature and any comments you wish to make, then close the window when the donation request comes up. Thanks. ST.
Just leave it alone. Eight St is for the bikes. Fifth st is for vehicles.
I drove down 5th St but could not find the Target store. ????
No offense Toasty, but did you actually read the report or watch the simulations? Why would you leave it alone if you could speed up the drive times, reduce stop times, and make it safer? The road redesign is not merely about bicycles, in fact, that’s just one small component.
“I drove down 5th St but could not find the Target store.”
You’re not too bright are you? Either that or you’re just not as funny as you thought you were.
You just didn’t go far enough, Bill. It’s in West Sac.
The study says it will increase vehicle delay time on the side streets by 34.71%. Incredibly it appears that the extra exhaust from these idling vehicles was not included in the fantastic claims of reduced emissions from the road diet plan.
So, we are talking about spending (how much?) to satisfy the main goals of improving bicycle route connectivity, pedestrian safety at crossing, and reducing vehicle speeds. Meanwhile we create long lines of idling cars on B, C, D, E, F and G streets waiting patiently for fewer opportunities to turn onto the new and improved one-lane Fifth Street. I assume that since cars are delayed 34.7%, bicycles and pedestrians will also be delayed. I can’t believe that anyone living on the side streets would support a plan that increases the amount of idling traffic in front of their house.
As a driver, bicycler and pedestrian in this area, my primary complaint is the difficulty turning on to Fifth Street or crossing it safely at C, D and E streets. Granted it can sometimes be daunting with four lanes of traffic. However, big gaps open as waves of cars pass through. The study authors admit the density of traffic on Fifth Street will increase “a lot” with the road diet plan. It will stack up more delayed cars spewing exhaust up and down the side streets… which are currently more charming, safe and pleasant for bikes and pedestrians. It will cause more delays for those needing to cross Fifth or turn on to it.
Am I missing something here?
“Incredibly it appears that the extra exhaust from these idling vehicles was not included in the fantastic claims of reduced emissions from the road diet plan.”
Jeff: remember you are talking about less than ten percent of the cars there, and most of those will end up on the signalized streets.
Jeff. David is correct. Actually, it is only 7 percent of all the vehicles in the corridor that are in the traffic counts northbound at the unsignaled intersections. That is C, D, E, I, J, and K combined. And remember, corridor studies like this one are conducted with a fixed model, where drivers are unable to seek better routes. In the real world, some of those driver will go to the signals, which allow them faster travel onto or across the corridor. Those delays that are shown in the model will go down. Steve Tracy.
“…remember you are talking about less than ten percent of the cars there, and most of those will end up on the signalized streets.”
Ok, fair enough on the emissions. Although 10% is 10%. However, I’m not sure I like the idea of more congestion on these side streets. These quaint little neighborhood streets are the veins and are reasonably void of cars (except when Farmers Market is in session) and Fifth Street is our smaller scale Watt-like traffic artery. I am all for improvements for bikes and pedestrians. And I really like thinking about how much nicer Fifth Street will be made to look with this design. However, when I think about the money spent and the tradeoffs for pushing more traffic onto these side streets, it doesn’t pencil out for me. Maybe this is only my personal view, but I don’t mind driving on Fifth as it is today, but I tend to avoid it as a biker and pedestrian for common sense safety reasons… just like I avoid other higher-traffic streets. If we go forward with this road diet plan, I can envision increased irritation for all as they experience additional delays trying to break through the single-file chain of cars traversing east and west on Fifth. I think there are probably less expensive things we can do to improve bike and pedestrian safety without causing more congestion on the side streets.
Question: where will the funding for this come from?
David, I think there is a problem with the top image file on this blog. I keep getting errors with IE. I can open it with Firefox, but the image displays slowly. I don’t think is my problem because it happens on multiple computers.
Had to drive around and avoid three ladies on bicyles traveling down 5th Street yesterday, between B and D Street. We need this road diet, for the safety of bikes and pedestrians, and for the safety and emotional well being of drivers. It would not bother me a bit if implementing the road diet resulting in a car taking a bit longer on side streets, or increased GHG emissions slightly, if it will make it safer for everyone. To me, the overriding concern is SAFETY. To me, this is a no brainer.
Jeff. You have raised several issues in your last comment, so I will try to parse them out and address each one.
1. Yes, 10% is 10%, but the most recent traffic counts show the side street traffic at intersections on both side of the tracks without signals is only 7%. I believe it is actually lower, because the numbers from I, J, and K Street submitted for modeling are suspiciously even. I don’t think they reflect actual counts. Including both northbound and southbound traffic on those three streets, the consultant magically “observed” five numbers that came out to even tens, and the sixth an even five. I think just the five zeros that show in the data are a ten-thousand to one probability. So they pulled these numbers out of their….
2. There will not be an unusual amount of congestion on these six side streets, because those drivers will go to routes that work better, until equilibrium is reached. They will probably settle on streets with signals, that in the model have little backup. This is the difference between the fluid state of the real world, and the rigid state of the model that is used for corridor studies. Even so, of the six streets with no signals, if you count both northbound and southbound vehicles, only northbound E Street carries more than 100 cars (146) in the peak hour. And that number is less than 2 1/2 cars each minute, and then only for the peak evening rush hour. This is a situation that occurs for only five hours a week, only the 30 weeks when UCD is in full session. This is less than 2% of the hours in a year.
3. Both the City’s consultant and the UCD engineering school ran models that demonstrate through traffic will flow BETTER with the proposed design. What more could we ask for a “Watt-like traffic artery” than better efficiency?
4. You may be in a position where you can easily avoid Fifth Street, but as comments on the petition that is being circulated show, many of us can not avoid Fifth Street except with great inconvenience.
5. You repeated this argument about “additional delays” so I’ll repeat the response. There will be very minor time added to an eastbound trip through the corridor (7 seconds) and a huge time savings in a westbound trip (nearly a minute). The overall time savings is huge. How will that increase irritation?
6. The funding will come from SACOG, and is federal money all parties are eager to get out into the economy and working. What better project than one that will improve safety for all users of the street, improve traffic flow, and reduce greenhouse gas emissions?
Steve Tracy.
Steve:
Thanks for the detailed reply. My issue about delays is for bikes, pedestrians and cars crossing or turning on to Fifth Street. I don’t dispute the study’s conclusion that time differences are negligible or slightly better between the existing and new plans. However, the study SAYS there will by a 34.71% time delay for these side streets. Any more delay is irritating… including the need to reroute… which does not improve the delay time.
You also seemed to miss my point about avoiding Fifth Street. I only avoid it to cross as a biker or pedestrian during the busiest times. Otherwise Fifth Street is not a problem for me. The road diet plan appears to make it more difficult for a pedestrian or bike to cross from C, D and E since we are increasing auto density on Fifth and reducing the gaps in traffic that allow safe crossing. For autos, it also appears to make it more difficult to cross or make a left or even a right turn from these three side streets. I assume this is where the additional delays manifest.
“The funding will come from SACOG, and is federal money all parties are eager to get out into the economy and working. What better project than one that will improve safety for all users of the street, improve traffic flow, and reduce greenhouse gas emissions?”
This comment helped me identify my real issue here. It is spending more public money on non-essential public works projects. Nothing in the report that I could find made commitments that improved safety would be realized. The time improvement are negligible. Additional side street delays will occur. The claims of reducing greenhouse gas emissions are out of context for this type of project; and again do not justify the spending of millions of public money. If approved, I suppose I would be happier to bike down Fifth Street, and it will certainly beautify the street. However, I would rather see this money used in the core area to improve parking and beautify the streets because this would result in an economic benefit to the merchants and the city.
Jeff. There is much more detail available on the benefits of the redesign, as experienced in other communities that have employed this exact solution to dangerous 4-lane streets. Please go to http://www.oldnorthdavis.net and click on the link to “Fifth Street Redesign Resources” to see a presentation I put together over the last few months. Let’s make this clear, this is not a “non-essential” project. It is one that will address a critical safety issue. And not just safety for bicyclists and pedestrians, but for the occupants of vehicles as well. Fully ten percent of all the accidents in this whole town occur on less than 4,000 feet of a single street–Fifth/Russell between A and L Streets. 5th Street has a significantly higher accident RATE (corrected for traffic volume) than other similar streets in town. Near the end of the presentation you will see data from other communities that document the huge reduction in accidents that is common with the road diet solution.
I will be repeating things here, but this 34% increase in side street delay is a function of the type of model used. It is the right model for this circumstance, but it is intended to evaluate main line (5th Street) traffic, not side streets. In the real world, those drivers suffering delays on the side street will experiment with other routes, and settle on those that are FASTER. This type of model does not allow that reassignment.
This isn’t just about pedestrians running across when gaps in the traffic appear. It is about designing the street so drivers are more likely to stop. Under California law, drivers must yield to pedestrians, but only when they are actually off the curb and in the crosswalk. A crosswalk, by the way, is the path connecting two corners, even if it isn’t painted. Two problems exist with our current design: First, for years Public Works staff has resisted painting crosswalks, which help drivers see that pedestrians may be present. Second, only by stepping off the curb in front of moving traffic does a pedestrian crossing Fifth Street actually enter the crosswalk and trigger their pedestrian status where drivers are supposed to yield. Not many of us are foolish enough to do this. Also, if we eventually get landscaped medians, the “nose” of that median provides a mid-crossing refuge for pedestrians. This also allows vehicles to stop in two groups. As the street sits now, EVERY vehicle approaching the crosswalk, both lanes in both directions, must yield when a pedestrian is in the crosswalk. A raised median splits the yield zone in two, legally, so until the pedestrian gets to the mid-point of the street cars on the other side can proceed. This reduces vehicle delay and idling time. With the proposed design and only one lane in each direction, all of this is of course even simpler for the pedestrian since the first car that stops holds up the whole line. As the street sits now, drivers often change lanes to avoid stopping, and zoom past the vulnerable pedestrians in the crosswalk. Very dangerous. Please also ponder the fatality risk chart in the presentation.
As to expense, the American Automobile Association (one of the country’s largest auto insurers) recently conducted a study on the full cost of automobile accidents. That study concluded that the average injury accident costs $68,000 in 2005 dollars. figure $70,000 now for a round number. Those costs include property damage, personal medical costs, loss of work, and the public agency costs (fire, police, courts, etc.). This does not factor in the hard to quantify personal trauma associated with these accidents. (A small child was slightly injured in an accident at noontime today right in front of the fire station.) There have been over 100 injury accidents in the 5th Street corridor in the five years that Public Works staff has resisted making this improvement. It is a reasonable expectation that a third to one half of those accidents might not have occurred had the street been fixed when the issue first surfaced. This calculates to a total cost of two to three million dollars for accidents that could have been avoided.
Another way to take a look at this is that fully ten percent of the effort our police and fire departments spend responding to automobile accidents in this whole town is spent on 5th Street. Cutting these accidents significantly would free them up to do other important things.
Finally, isn’t one of the most important functions of government to protect the welfare of the people? Here is a case where serious safety issues are present because of an outdated street design. A case where the fix is known and proven, where no group will suffer any delay or reduction in services with the fix. As one resident recently stated, it isn’t a matter of we should do this, but under these circumstances we MUST do it.
Thanks for your attention.
Steve Tracy
Steve:
First, I think this is a great post. Thanks for putting in the time.
I will take a look at the link to your presentation.
“Another way to take a look at this is that fully ten percent of the effort our police and fire departments spend responding to automobile accidents in this whole town is spent on 5th Street. Cutting these accidents significantly would free them up to do other important things.”
I wasn’t up to speed (no pun intended) on the metrics for traffic accidents and fatalities on Fifth Street. I can be convinced to support the road diet plan based on this type of analysis.
However, my answer to your question “Finally, isn’t one of the most important functions of government to protect the welfare of the people?” is “yes, but with intelligent cost-benefit discretion applied”. Our desire to protect the welfare of the people seems to a significant driving force behind government overspending. We have a civic responsibility to support government spending to keep our roadways safe, but we cannot afford to eliminate too many natural risks. Watt Avenue in Sacramento is an extreme example where auto accidents frequently occur, and fatalities frequently occur from pedestrians trying to cross where it is not safe. The intersection of Watt and Fair Oaks at one time was known to be the most dangerous in Sacramento. Changes have been made (lanes added actually… as well as other changes) and I think this is no longer the case. However, it is still a dangerous place to walk or drive, and anyone would be foolish to ride their bike on Watt.
In Davis, the same group opining for downtown road diet measures is also part of the group agitating to prevent peripheral shopping development. Despite our best efforts to prevent it, Davis has grown in population and geography. The lack of peripheral shopping alternatives forces more people living in the periphery to drive downtown. This increases traffic and this has a natural tendency to increase the number and frequency of traffic accidents. From one perspective, the road diet plan seems like another forced battle of core area residence that naturally want less traffic and their friends that envision Davis as being a small European village, with the downtown businesses and the rest of us that live in the more common and real world.
But back to the safety issue… for what it is worth, I will throw my support to a road diet plan that addresses a significant safety problem as long as we have some fact-based assurance that the changes will improve things commensurate with the costs. Maybe you can save me some time and point me to these facts.
Jeff. Please take a look at the slides near the end of the presentation from Slide #25 to Slide #43. Wherever I could get data, I have included it. No data does not mean I omitted information that doesn’t support our case. There just hasn’t been systematic information gathering in a lot of the communities where this design was used. They just did it, and they liked the result. Very telling is the fact that in all the examples I have heard of, only one redesign has been completely reversed–I believe either Missoula or Billings. This was due to unending political pressure, not because the redesign failed in any way. (One of the lead opponents of that redesign was later killed by an automobile while crossing that very street.) Also, one in San Leandro was partially undone a couple of years ago when one lane was put back for a portion of the street. Again due to political pressure, not traffic issues.
Please inspect Slide #30 in the presentation, and you will see that of all the quantitative and qualitative criteria selected to evaluate this road diet in Orlando, only one did not come out positive. And that is the merchants–opposed to the change from the beginning–reporting negatively on their impression of their customers satisfaction with the new design. Yet those same customers reported satisfaction. Nobody ever calls the Mayor and says they had a great drive on a redesigned street, but the opponents often can not see the positive evidence right in front of their eyes.
Interesting you should mention Watt and Fair Oaks. I led the transportation planning team for the Sacramento County Planning Department for nearly 15 years, and heard that claim often about Sunrise and Fair Oaks. It was usually fueled by a Sacramento Bee article. Then we would deploy a UCD intern to get the numbers, and correct the mistaken impression. Once we adjusted for the VOLUME of vehicles through the intersection, the accident RATE was nowhere near the highest in the County.
We have attempted to do this same kind of analysis in Davis, as shown on Slide #10. This compares the 2007 accident data (the last full year I had at the time) on 5th Street and a few other high-volume streets in Davis with only two through lanes. The traffic volumes vary some, and the lengths a bit more, but even the quickest math-in-your-head shows that the other streets have accidents rates that are just a fraction of 5th Street’s. It doesn’t have to be like this.
Finally, I don’t think this has anything at all to do with Target. I have been on the Board of the Old North Davis Neighborhood Association for years. We never took a position either way on Target. I’ve also been involved in Davis Bicycles! since the beginning, but that group wasn’t even formed when the Target vote was held. This redesign is in the adopted General Plan because the idea came up in the deliberations of the broad-based citizens committee overseeing the preparation of the Mobility Element in the 1990s. I was also on that committee, and can guarantee we never discussed a Target in Davis. If you look at the addresses of the people signing the petition, you will see that those from outlying ares far outnumber core area residents. We all drive on this street, and we all face a higher level of risk because of the outdated design. It doesn’t have anything to do with where we like to shop.
Again, Jeff, thanks for your polite discourse on this critical issue.
Steve Tracy.
Steve:
Thanks for the info. I will take a look. I appreciate your time and attention and hope we both might at least provide some food for thought. My late, great mother always reminded me “Don’t attribute to malice that which can be attributed to the rational pursuit of self-interest or simple ignorance.” I think the first is rare, the second should be accepted and the third can be corrected if we are open minded.