The End for LOS in California? State Wants Input on a New Planning Metric

With little fanfare, California is considering a change in how it measures transportation impacts that could herald a major change in environmental law. SB 743, passed and signed into law in September, is a potential game changer because it could completely remove LOS — Level of Service, a measure of car traffic congestion — from the list of tools that must be used to analyze environmental impacts under the California Environmental Quality Act. As the state contemplates a broader, more sustainable metric to use for smarter urban planning, the public is invited to weigh in on what the LOS replacement should look like.

## USA## doesn’t pull punches when describing why many oppose”Level of Service” metrics. Image: Andy Singer

CEQA requires new projects, be they highways or housing units or basketball stadiums, to analyze potential environmental changes created by the proposed project. In copious detail. Water, air, land, noise, plants, animals: any physical aspect of the existing area that might be affected negatively must be analyzed.

For a variety of historical reasons, traffic congestion has crept into this group of environmental impacts under CEQA and become part of the law. Congestion is analyzed by measuring the flow of traffic at intersections (how many vehicles get through in a set amount of time) and grading those intersections on their performance. Planners refer to this as LOS, for Level of Service.

The irony of LOS is that CEQA requires mitigation when projects cause delay to automobile traffic—even if the projects create better conditions for other road users, such as transit riders, bicyclists, or pedestrians. Thus the San Francisco Bike Plan was held up for years because of a lawsuit claiming the city did not take into account the negative effects bike infrastructure would have on LOS.

Streetsblog covered SB 743 as it was passed last year, but at the time we missed a nuance that makes it an even bigger potential change for CEQA and planning. At first read it looked like the LOS provision, tacked onto a bill written to streamline environmental review for a new Sacramento Kings basketball stadium, applied only to areas designated as “Transit Priority Areas,” defined as within a ½ mile of high quality transit. In some places, this covers very large areas: for example, most of San Francisco is so designated because of its dense transit networks. This alone could make a huge difference in the way environmental impact reports are handled for many projects.

Neither Streetsblog nor many advocates monitoring the legislation realized on the first read that the new law gives the Governor’s Office of Planning and Research (OPR) the discretion to come up with a substitute for LOS and apply it throughout the state—not just to urban areas “well served by transit,” but everywhere. And to all projects.

The long-term results of using LOS as a measure of environmental impact have been argued about for years and explained well elsewhere. Removing it from the CEQA process has the potential to profoundly affect the way cities are planned and built. And while some of the larger cities, including Los Angeles and San Francisco, actively pursue the question of whether traffic impact is an appropriate measure of environmental impact (and working on their own substitute measures), not every locale is happy about it.

OPR is asking for early feedback on two items: a draft list of goals it wants the new criteria to meet, and a preliminary list of possible replacement measures for LOS. These are both described in detail in this report, and summarized below. The deadline is this Friday, February 14, and comments can be sent to: Future drafts will incorporate feedback received now, with the goal of preparing a final draft by July 1, 2014.

Below is an explanation of why many people oppose using LOS as a measure to analyze environmental impacts. Streetsblog is also reaching out to municipal leaders who use LOS for a future story explaining why they may not want to remove it entirely from CEQA.

LOS Redux

The way LOS is measured and applied has had unanticipated results—perverse outcomes, in planner speak—that run contrary to other state goals such as emissions reductions and safer streets.

To illustrate, consider a project proposed in a relatively undeveloped area. It might affect traffic flow at a few nearby intersections, but if those are wide, free-flowing intersections currently experiencing little delay (with a current LOS of A or B, say) the new project may well have a minimal effect on overall traffic flow. On the other hand, a similar project in a busier, more built-up area will likely have a bigger effect on traffic flow at a larger number of intersections. This is because 1) in a dense area there are probably more intersections within a set distance and 2) those intersections may already have more delay and more congestion (a lower LOS) to begin with, given already existing land uses.

But which of the two proposed projects will produce more car trips and therefore more traffic, emissions, noise, and safety issues? LOS can’t tell you that; it can only tell you whether nearby traffic is likely to be delayed. It does seem intuitive that a project in a dense area where a lot of people walk, take transit, or ride bikes would produce fewer car trips than the same project in a place where people always only drive—but measuring LOS won’t get you that information.

In addition, it turns out the fixes for the problem—the mitigations required by CEQA—are not so good. If the CEQA analysis shows that a project will lower LOS to an “unacceptable” level, then something must be done about it—and that something could well be a design for a wider road with more lanes and faster traffic. This would solve the Level of Service problem, but a wider, faster road also creates less safe and certainly less comfortable conditions for users who are not in vehicles, including people on bikes or on foot.

And which travelers contribute more in terms of emissions, noise, and the potential for life-threatening collisions?

Thus requiring mitigations because a project reduces the number of cars that can get through a given intersection in a given amount of time can lead to road designs that discourage efficient, active transportation that might actually lower emissions and noise.

Longstanding arguments against using this traffic flow grading system include whether traffic flow is even an environmental impact that needs to be analyzed—while it is true that traffic jams can affect local air quality, emissions are assessed separately under CEQA. Worrying about traffic flow above other considerations has encouraged car-centric development in outlying areas and created wider, faster roads and longer distances for everyone, including people who aren’t in cars.

There are other arguments against using LOS, but the kicker is that a development’s attempts to improve access for people not in cars—by, say, adding a bike lane or widening a sidewalk—could lower the LOS at nearby intersections, thus triggering the need for mitigation. That’s a huge anti-incentive to improving infrastructure for anyone except car drivers.

OPR’s Draft Replacement Goals and Metrics

Under SB 743, a traffic impact measure chosen to replace LOS must help reduce greenhouse gases as well as promote multimodal transportation and diverse land uses.

OPR has drafted an expanded list of potential goals and is looking for feedback on them. They include: maximum environmental benefit and minimum environmental harm; efficient use of limited fiscal resources; social equity, including low-cost access to destinations, livable communities, and minimizing traffic impacts; maximum health benefits associated with active transportation and minimum adverse health effects from vehicle emissions, collisions, and noise; simplicity and clarity; consistency with other state policies such as greenhouse gas reduction and complete streets; and efficiency of the overall transportation network not just for cars, but for all users.

The OPR report asks: are these the right objectives? Are there other objectives that ought to be included?

Then there’s a preliminary list of alternatives to LOS. Each one of these deserves a separate discussion too lengthy to cover here. So far they include: vehicle miles traveled (per capita or per person); automobile trips generated; multi-modal level of service; fuel use; motor vehicle hours traveled; or a presumption in particular defined areas that any single development will not produce significant regional traffic impacts.

A series of open questions at the end of the OPR report make for some thought-provoking reading, including questions about the effect of parking, which wasn’t addressed here, although it does appear in SB 743.

Look for more discussion of LOS and OPR’s efforts in the next few days.

  • True Freedom

    Vehicle Miles Traveled (VMT) is a poor metric for autos, because it does not account for the “quality” of that mile. One mile traveled on an open, free flowing route is a “better” mile than one traveled in stop and go/ congested areas.

  • Juan Matute

    It’s all about the environmental implications under CEQA, and air quality is analyzed separately from transportation impacts. Transportation impact analysis under CEQA is supposed to assess the non-social aspects of additional load on the transportation system. Level of service assessed the social aspects of loading. In a state that’s looking for aggressive reductions in greenhouse gas emissions, VMT per trip or Auto Trips Generated per trip are the best metrics.

  • EastBayer

    What about new vehicle trips induced?

  • I don’t think it’s something the state is looking at, but it’s not a bad idea.

  • thielges

    Oh man this is a great opportunity to change LOS to incorporate other transportation modes. Traditional auto-centric LOS only considers the delay of cars through streets. So the first thing to fix is to replace the measurement of car (as in the piece of equipment) delays and replace it with the delay of people. This would give heavier weight to transit modes that carry multiple people as opposed to single driver autos. For example this would open the door to reconfigure signals and crosswalks at busy intersections to allow more people to pass with less delay. It also makes it more politically feasible to implement true transit priority signalling.

    The other key factor is to factor in is access to transportation networks. Today any driver can drive on any street. There are no terrifying “double black diamond” (in this frustrated skier’s parlance :-) streets that would deter the average driver from going from point A to B. We have successfully dumbed down our street network so anyone in a car can use them.

    Compare that to the cyclist’s perspective where just about any journey of more than a few miles requires traversing a narrow busy street, gnarly interchange, or other scary street feature that requires nerves of steel. The number one factor that keeps a larger population of people from cycling is that they feel unsafe and uncomfortable on our existing streets. In other words our street network is inaccessible to all but the most die hard cyclists.

    LOS rates streets and intersections on a scale of A down to F where A is great and F is unacceptable. If we rated our streets on whether they are accessible to the average person on a bike then pretty much the entire state would be rated “F”.

  • Juan Matute

    The idea is to normalize whatever metric against a denominator such as all trips, which, in the case of VMT, incorporates regional locational efficiency as well as mode split. In the case of auto trips generated / total trips, it only incorporates mode split, but not regional locational efficiency, an accessibility measure.

  • Wanderer

    It’s good to get rid of LOS. But most of the metrics proposed–with the exception of Multimodal LOS–don’t provide any way to measure the impact of development or roadway projects on transit travel time. So projects could go through that slow down transit, making it less attractive and undermining the intent of the legislation. This needs to be remedied.

  • Juan Matute

    Transit travel times are not an environmental impact under CEQA. Air emissions and noise resulting from congested travel are, but longer or more variable travel times on their own are not.

    The next effort needs to be to reform the law enabling congestion management programs to make way for a systemic approach to prioritize transit service. Assessing transit at the “project” level (project in CEQA terms) is an incrementalist strategy that’s not going to get the bus anywhere.

  • Wanderer

    I’ve seen a few jurisdictions treat transit travel time as a CEQA impact (e.g. San Francisco) and I’ve seen others treat it as a “non-CEQA impact” within EIRs.

    I’m all for making congestion management programs more transit-oriented. But since that’s likely to take a while, I think it’s important to fight against degradation of existing services. If the existing bus routes become slower and slower they’re not going to be attractive to passengers or to politicians making funding decisions.

  • Jerard Wright

    The devil is in the details.

    Two things that I wonder; How this impacts regional transit projects as they will need that LOS for potential federal grant dollars? And how this can sound good for some smaller project like a bike lane but can create major impact not for motorists but for transit buses which are carrying more people per vehicle?

    For example a poorly planned and conceived road diet along a major bus transit corridor has the potential to seriously impact transit service by imposing further schedule delays due to congestion and thus further limiting and inhibiting the multi-modal potential of such an arterial and sacrificing one mode in replace of the other.

  • KillMoto

    We should never ever spend another penny widening roads for the sake of air quality. Instead, we need to incentivize the adaptation of cars that emit nothing when stopped: hybrids, pure electrics, and traditional petrol cars with automatic start/stop.