New Zealand has more than 1,000 roundabouts and, following the British example, has historically used them as a means of intersection control for both low-speed urban and high-speed rural and highway applications.
However, in recent decades design engineers have been challenged to better provide for pedestrian and particularly cyclist users, which are perceived as being disadvantaged. Concerns for these vulnerable road users are at least partially responsible for the increased proliferation of traffic-signal-controlled intersections in many New Zealand cities.
In May 2012 the New Zealand Transport Agency (NZTA) published a report titled “Improved Multi-Lane Roundabout Designs for Urban Areas” (NZTA 2012). This was a two-year research project that was mainly motivated by the need to respond to the design challenge of providing for vulnerable road users and also by the apparent need to better quantify safety advantages/disadvantages between roundabouts and traffic-signal-controlled intersections. The report reviews international design practice, makes some recommended legal changes in the transport code for the New Zealand government to consider and includes some preliminary design guidance, which is expected to be universally applicable. Similar to precedents from the U.S., a roundabout first-type policy also is recommended for adoption in New Zealand, and this is based on safety reasons.
The research was divided into five main topics:
- Compare safety between multilane roundabouts and traffic signals for all types of road users;
- Research and evaluate options for pedestrian facilities at multilane roundabouts;
- Research and evaluate the use of vertical deflection devices at main road roundabouts;
- Research and evaluate current guidelines for sightlines at roundabouts; and
- Evaluate the turbo-roundabout from the Netherlands for feasibility in New Zealand.
This research found ample evidence to demonstrate that a well-designed roundabout should in general experience significantly fewer vehicle injury crashes than if the intersection were controlled by traffic signals, and especially those involving serious and fatal injury. Safety performance for a given location will depend on factors including traffic volumes, number of legs and traffic lanes, and whether or not there is adequate speed control for all directions.
Affirming the relevance to New Zealand, an analysis of 40 urban intersections in the Auckland region undertaken as part of this research demonstrated a 47% reduction in vehicle-occupant injuries—and many overseas studies show larger savings than this. Thus it was concluded that in order to reduce nationwide injury crash statistics at urban intersections, roundabouts should be the preferred choice over traffic signals—particularly so for intersections with four arms or more, which is where vehicle conflict speeds can be higher.
The safety and amenity of cyclists at multilane roundabouts was found to justify more attention by designers, as evidence does indicate that these users can not be insignificantly affected. However, measures to either reduce vehicle entry speed or physically separate cyclists from vehicle traffic are expected to substantially address these concerns, as the majority of cyclist crashes at roundabouts involve the circulating cyclist being hit by a driver entering the roundabout. A new type of low-speed multilane roundabout called the C-Roundabout has been developed in New Zealand and successfully improves safety for cyclists by reducing driver speeds to around 30 km/hr, and signalized roundabouts in the United Kingdom also have improved cyclist safety.
This research did not conclusively find a significant difference in safety performance for pedestrians between multilane roundabouts and traffic signals. Some of the nationwide statistics inferred that traffic signals may be presenting more safety problems for pedestrians than roundabouts in New Zealand, but further research is required to elaborate on this.
Zebra, Hawk and Pelican
The research demonstrated that a well-designed multilane roundabout should be able to accommodate pedestrians relatively safely. However, design of the pedestrian crossing point can be a critical factor at multilane crossings and particularly where vehicle speeds are higher than around 30 km/hr.
Zebra-crossing-type facilities, where pedestrians have priority over vehicle traffic, offer the greatest mobility to able-bodied pedestrians, although they can have some disadvantages to visually impaired users. A review of zebra crossings at multilane crossing points in Auckland, New Zealand, demonstrated that they can be relatively safe if located less than 20 m from the roundabout, mainly due to the lower vehicle speeds near circulating lanes. However, zebra crossings at multilane locations where vehicle speeds are higher can invariably experience safety problems, especially if vehicles regularly queue over the crossing. Additional safety measures to compensate include raised platforms, staggered island arrangements or active warning devices such as flashing signs or flashing road studs.
Pedestrian signals near roundabouts are a viable alternative to zebra crossings, but pedestrian wait times need to be set low enough to reduce the jaywalking that will otherwise occur and in turn may compromise pedestrian safety. Staggered signalized crossing arrangements can reduce disruption to vehicles as crossing times are shorter for each direction, and “Hawk” or “Pelican” crossings as used internationally can reduce disruption to traffic flow with apparently minimal compromise to pedestrian safety.
It also was found that lane arrow markings, which cause uneven vehicle queuing on roundabout approaches, can have safety implications for pedestrians, so they should be applied with due consideration where there are multilane zebra crossings or pedestrian signals (especially if there is much jaywalking).
The premise for this topic was that given vertical-deflection devices at roundabouts are beneficial for pedestrian and cyclist safety due to the reduction of vehicle speeds, there should be some justification why they can’t be used more often on main road roundabouts worldwide. In some cities such as Malmo, Sweden, they are more often being used for this purpose.
This research has identified that the most likely adverse effect of any significance would be some additional noise generated from some heavy vehicles as they traverse the device (e.g., lightly laden trucks with three axles or more and mechanical leaf-spring suspension, or two-axle trucks if driven at excessive speed). Although there are other potentially adverse effects, including delays to emergency vehicles, vehicle occupant discomfort, fatigue damage to heavy vehicles, traffic diversion and vibration damage to adjacent buildings or structures, all of these were found to be usually of minor nature and therefore of minor significance. For any proposed installation, the safety benefits of a vertical-deflection device should be objectively weighed up against these potential adverse effects. For example, noise effects could feasibly be assessed by a review of truck volumes by type, time of day and proximity to sensitive land-use activities.
The main options are raised speed platforms, speed humps and speed cushions, and in general the higher profile the vertical-deflection device (and depending upon gradient of approach ramps), the greater the speed-reduction effect. The use of vertical-deflection devices at roundabouts need not be limited to only pedestrian and cyclist safety considerations—they offer an economic alternative to vehicle deflection as a means of vehicle speed control, which might otherwise be costly in terms of land-take.
A look at sightlines
Sightlines of opposing vehicles to the right (to the left in the U.S.) can influence driver speed through a roundabout, sometimes more so than vehicle-path geometry. It was concluded from this research that sightline restrictions can potentially be used as a means of reducing driver speed through a roundabout, but they do need to be applied with due caution as safety can potentially be compromised.
Excessive sightlines of opposing vehicles can encourage higher than desirable driver entry speeds at a roundabout and can subsequently increase crash types, including loss-of-control, rear-end and entering-vs.-circulating vehicles (particularly with two-wheeled users, who are less visible). In the United Kingdom, visibility barriers have been successfully used to address loss-of-control and rear-end crashes at higher speed rural locations, and British design guidelines suggest this measure as an optional treatment.
However, one significant finding from this research is that if sightlines to the right are too restrictive relative to the speed of opposing vehicles, then entering-vs.-circulating vehicles can potentially increase. The crash history of a particular roundabout in Auckland where sightlines were severely restricted did demonstrate this—there was a relatively high chance of an incident occurring as drivers needed to virtually stop at the limit line in order to adequately make a decision whether or not to enter the intersection.
The turbo-roundabout is a generic series of designs developed in the Netherlands, and since 1997 there have been around 70 installed in that country. Their main design elements are a spiral lane arrangement with mountable lane dividers on the circulating carriageway and approaches, which lend to slower vehicle speeds and fewer sideswipe crashes. Evidence suggests it has superior safety and capacity performance to a conventional multilane roundabout with a similar number of lanes, although it may be relatively expensive to construct in terms of land-take and additional costs such as lane dividers, delineation and signage.
It was recommended that an example be built and evaluated in New Zealand, including a comparison of safety and capacity with or without the mountable lane dividers. R&B