Congestion and Auckland Council's Transport Evaluation: What difference will Public Transport ma
In 2011 the Strategic Group in Auckland Council released their evaluation of Transport with different four different city models. One key metric was the % of vehicle kilometers traveled in congestion, shown below.
The analysis suggests that dispersed expansion followed by intensive containment will produce the most congestion. Intensive expansion was ranked third. The least amount of congestion was with the dispersed containment model.
It seemed counter-intuitive to me that the two types of models under the dispersed category would be at the opposite ends of the congestion scoring because they both involve going out, but differing only in the density of the local town and the quantity of them. Surely it would depend on the type of trip - i.e. where the destination was and when it was taken, and the transport capacity and options between different towns. I wanted to check whether my underlying assumption for the dispersed city model category: at first glance that trips are less likely to encounter other trips to the degree that would cause congestion is correct, for it seemed that this was not the case.
I requested the inputs to the ART3 model - the traffic model that Auckland Council uses to model and predict future traffic trends across cars, public transport and active modes - part of which shows the distribution of trip purposes:
What immediately became apparent is home to work trips contribute less than 1 in 5 trips. The remaining 80% of all trips are related to schools, shops, social reasons and almost 50% of all trips are under the category of 'other'! The Preliminary Studies report stated that between two thirds and three quarters of the 'Other' category of trips were "for recreational/social purposes, such as eating or drinking, watching or playing sport etc." Furthermore, the proportion of these two trip types taken in active or PT modes was around 15% in each category.
Public Transport is good at taking a large number of people from one location to another, especially if the main purpose of the trip is to transport a person. When more than one person from a family or social group wants to take a trip, or if the purpose of the trip is to transport something, they are more likely to use a personal vehicle.
The challenge for trips in the 'other' category is it is almost impossible to provide a comprehensive public transport service for them in a city that is well known for its outdoor activities and social life with friends and family.
Returning to the original Evaluation, it would seem possible that a more dispersed model would produce more vehicle kilometers traveled if we assume that where people live is not necessarily where they prefer to play or watch sport and not necessarily where their friends/family live. In that context, two comments are necessary:
If we were to travel in a "Dispersed Expansion" city model, then it is likely that vehicle congestion will increase, because the probability that you will meet other trips increases. However, this conclusion is plausible if the traffic-carrying capacity of the road was not increased to allow more public transport and general traffic.
Given the types of trips that are hard to predict (i.e. the 'Other' category) with respect to the destination and timing, it is important to recognize that public transport is going to play a minor role for that type of destination. Therefore active and car-based choices must be prioritized and the necessary infrastructure funded and built. The increase in PT capacity should be aligned with the types of trips that are more likely to use it: which intuitively is work and education, although it is very interesting that the inputs to the ART3 model suggest otherwise (education and other).
While the City Rail Link will help with work and education trip types, I have the concern that the Dispersed Intensive and Dispersed Expansion Models could be discarded because of the conclusion of increased congestion. It should also be noted that the same model states that
1. Average car speeds across the network will be approximately double average public transport speeds regardless of the city model.
2. It logically follows that access to employment by a predefined SLA across the network by car ranges from 50% to 75% while it ranges from 4% to 6.5% by public transport.
If this is the case under existing transport infrastructure arrangements, then reducing congestion will disproportionately benefit cars. To improve the same metric by public transport will require significant investment. This is not to say that public transport has no role, and it would be possible to improve access to employment by public transport if intensive development are permitted and residents choose that lifestyle. That said, the 6.5% figure quoted above is the best result, and the model predicts this in 2041 under the Intensive Containment scenario.
This might explain Auckland Transport's massive investment into public transport over the last 5 years, and that this will continue for the medium-term future. It is my opinion that investment in public transport be provided for, but caution should be applied as to the outcomes of it, bearing in mind that 5% of all work trips use PT, and 25% for secondary and tertiary education locations. Since the ART3 model was created the proportion of trips by PT will have increased, but even if all work trips used PT, this would impact less than 20% of all trips, and likely significantly less, given that public transport utilization to destinations outside of Auckland's CBD is unlikely to achieve the same utilization as CBD-terminated work trips.
