Working on a proposal to link Wellington and the Wairarapa with a new 110 km/hr highway, I’ve been hearing about punching a tunnel through the Remutaka Range.
Not a good idea.
The Remutaka Road Action Group’s (RRAG) business case calls for reasonably flat grades, with an absolute maximum slope of less than five percent (note 1). In the Remutaka and Tararua Ranges, there are two options: Cut through the top of a ridge, or build a tunnel. Practical routes for the new highway include a southern Tararua crossing with a ridge-top cutting 80 metres deep, and a Remutaka crossing with a 160 metre deep cutting.
Another option crosses the Remutaka Range through a very deep cutting, about 250 metres deep. Excavating this very deep cutting could cost more than building a 1.3 kilometre tunnel. This option closely follows previous proposals from the 1960s and 1990s, both of which had tunnels through the same ridge. When RRAG talks up the new highway, many people think about those earlier tunnel proposals. They assume RRAG is proposing a tunnel, even though two of the three route options do not need tunnels.
The more I look at this project, the stronger my objection to a tunnel.
The existing Remutaka road is the only link between the Wairarapa and the nearest port and airport, and it is major constraint on the Wairarapa region’s economy. The 52,000 people who live in the Wairarapa depend on this road for critical services, including important medical services. The new highway must be open at all times. The existing hill road is winding and dangerous. If the new highway is closed, the existing hill road might be an acceptable detour for some road users, but only in emergencies.
Experience with highway tunnels shows that they are not fit for RRAG’s purpose. Two recent examples highlight the problems: Sydney’s M4 Motorway (opened July 2019), and Auckland’s Waterview Connection (opened July 2017). Both tunnels comply with safety rules applicable in New Zealand and Australia. A tunnel on the new Remutaka Highway would be required to comply with the same, or perhaps more stringent, safety rules.
Both the Waterview and M4 tunnels are frequently closed for maintenance. NZTA advertises that they expect to close the Waterview tunnel for two to three nights a month. Some of my former colleagues were involved in construction and maintenance projects during planned M4 tunnel closures.
Safety
Highway tunnels need frequent maintenance closures because they are jam-packed with complex, fragile systems. Most of these systems are connected with fire safety.
Fire safety starts with the basic design of the tunnel. Like the M4 and Waterview, the Remutaka Highway Tunnel would have a parallel “twin-tube” design (note 2), with one tube for each traffic direction. The roadway tubes are connected every 50 to 100 metres by “cross-passages”. The cross-passages allow people to evacuate from a fire zone to the unaffected tunnel tube, and then out of the tunnel.
A fire detection system based on heat detection triggers the safety systems, stopping traffic and activating evacuation signs, ventilation systems, and so on. Unlike smoke detectors, heat detectors provide a very reliable indication of fire: If a heat detector goes off, there is almost certainly a fire. Somewhere.
The biggest risk to tunnel users is smoke inhalation. So, like evacuation routes in high-rise buildings, today’s highway tunnels have complex ventilation systems designed to blow smoke away from evacuation routes.
Tunnel fires tend to be very hot, with temperatures usually exceeding 1000°C. Also, tunnel fires tend to spread quickly between vehicles. A highway tunnel needs a fire suppression system which can damp things down until firefighters can take over. Usually this is a water deluge system.
This is where things start to get really messy. By “messy”, I mean expensive and unreliable. The deluge pumps can supply water only to a limited number of zones, usually three or four. The automatic control system is programmed to dump water into the zone identified by the heat detection system, and into the adjacent zones (note 3). However, tunnel engineers have found that heat detection systems do not accurately pin-point the fire. Often the first alarm signal comes from a nearby zone. To make sure the deluge hits the right spot, human operators use closed-circuit TV to check, and if necessary over-ride, the automation.
There’s a lot to go wrong.
The tunnel has miles of plumbing, hundreds of valves, several pumps, generators, water tanks, ventilation fans, ventilation dampers, cameras, evacuation signs, sirens, and the list goes on. All this equipment is controlled by hundreds of computers, with battery backed up power supplies, networked over miles of fibre-optic cable. A failure in any part of the system can lead to multiple fatalities. A possum sleeping in a fan, a kaka nesting in a deluge pipe, a rat chewing cable: Anything.
Trying to make sure all this complexity works when it should, highway operators frequently close highway tunnels for testing. But that’s not all. Software needs regular updates. Each item of equipment needs regular replacement (every ten to fifteen years for computers, perhaps 20 to 30 years for pumps, and so on and on). And of course there are the inevitable accidents, out of control vehicles damaging equipment, safety barriers, and so on. Any replacement, maintenance, or upgrade job requires a tunnel closure.
High BCR Reduces Risk
The Remutaka Highway project has a very high benefit cost ratio (BCR), with estimated economic benefits more than seven times the construction cost. This means a small increase in cost will not jeopardise the project’s economic viability.
One very important safety feature we rarely see in highway tunnels is the stopping shoulder. An international tunnel consultant observed that in some tunnels, most crashes are caused by stationary vehicles in the traffic lanes. Stopping shoulders reduce that risk. We have to dig out more rock, but because of this project’s high BCR, widening the tunnel to make room for stopping shoulders won’t ruin its economic viability.
In my opinion, the addition of stopping shoulders does not overcome the safety problem. Even with stopping shoulders, a tunnel would not be as safe as an open cutting.
In the event of a crash in a cutting, seriously injured people can be airlifted from very close to the crash site. With a tunnel, they first need to be taken out into the open. Rescuing crash victims in the confined space of a tunnel is more difficult, even without the heightened risk of fire.
Tunnel crashes are more dangerous, and the victims are less likely to survive even after being rescued. An open cutting is safer.
Reliability
Frequent maintenance closures would be unacceptable for the Remutaka Highway. Closing the Remutaka Highway would force night-time traffic onto the winding, steep, dangerous Remutaka Hill Road. This is unacceptable for some road users. Transferring a patient between Masterton and Lower Hutt or Wellington hospitals over the Remutaka Hill Road, for example, is out of the question. You might as well put the patient inside a washing machine. The only practical option is an expensive air transfer.
Motorway tunnels in Auckland or Sydney do not present this type of problem, because these tunnels run parallel to good quality local roads that can handle displaced night-time tunnel traffic.
Summary
The new Remutaka Highway will overcome a critical constraint on the economy of the Wairarapa and Greater Wellington regions. The highway is designed to be easily upgraded to a true dual-carriage motorway.
Motorways are by far the safest type of road. Remutaka Highway users can expect a big improvement in safety.
Putting a motorway through a tunnel, even with stopping shoulders, undermines the potential safety improvement. In addition, a highway tunnel requires continual monitoring by human operators, increasing the highway’s operating cost.
A highway tunnel will not be reliable, because of frequent closures. Some traffic will be able to detour over the existing Remutaka Hill Road. However, that winding dangerous road is not an option for users such as ambulance operators.
The RRAG business case shows two good routes for a tunnel-free Remutaka Highway.
In my view, putting the new highway through a tunnel would be a mistake.
We are technorg.
Notes
1. Five percent gradient means one metre vertical climb for every 20 metres horizontal distance.
2. I have seen single-tube designs in which part of the tunnel is walled off as an evacuation gallery. However, RRAG has been advised that only a twin-tube design would satisfy current safety requirements in New Zealand.
3. Tunnel fire detection/suppression zones are usually 50 to 100 metres long.