APPENDIX B - TECHNICAL ISSUES
As part of the Phase Two analysis, detailed investigation has been undertaken by specialist transport consultants to model the effects of various scenarios under consideration for the tolled bridges in Scotland, using the Transport Model for Scotland ( TMfS). TMfS is a multi-modal transport model capable of predicting the effects of transport interventions and able to provide a growth forecast for future years. Results from the modelling are indicative and focus on re-routing and changes of travel mode - so induced trips that may result from changes to the network are not included. The modelling carried out for Phase Two looks at the longer-term impacts of changes to the tolling regimes; short-term impacts in the immediate aftermath of any changes may not mirror the results of the modelling. The opportunity has also been taken to obtain up-to-date information on a number of High Occupancy Vehicle ( HOV) schemes and HOV research projects currently being carried out in the UK and in other parts of the world.
These investigations have been categorised into three separate areas of work,
1. Analysis of Erskine toll scenarios and effects on the surrounding network
2. Analysis of Tay and Forth toll scenarios and interaction with other bridges
3. Investigation into measures to increase car occupancy over the Forth Road Bridge
Individual Reports have been produced for each of these three work studies and the reports are available separately. However, to facilitate understanding of the main issues for the Tolled Bridges Review the following sections summarise each report and reproduce the respective conclusions drawn up by the consultants for each study.
This traffic modelling builds on earlier work carried out in Phase One, also using TMfS. In Phase One, the following four tests were carried out at Erskine, Forth and Tay:
- The status quo;
- Removing tolls;
- Halving tolls;
- Doubling tolls.
As well as those general tests, there were several independent tests carried out -
- A differential tolling scenario of 50p for cars and £1.00 for HGVs (both each way) was modelled for Erskine;
- as TMfS was not sufficiently developed to model toll differentials by time of day during Phase One, a high toll was modelled for Forth and Tay to obtain an indication of sensitivity to peak hour charging. The scenarios modelled (£5 for cars/£7 for HGVs at Forth, £4 for cars/£6 for HGVs at Tay) were chosen following discussions with Bridge and Local Authority Transport officials. A larger increase was not modelled for Erskine as it does not have a congestion problem;
- the effect of one-way tolling at Erskine Bridge was tested.
1. Erskine Bridge
Modelling work has been carried out using the Transport Model for Scotland ( TMfS) to apply different toll levels to the Erskine Bridge and examine the effects on the surrounding road network, with particular attention to the A82 traffic corridor and other major Clyde crossings within the Glasgow Council area.
The consultant was required to use the model to test the following three scenarios for the forecast years of 2006 and 2011.
- maintain the current bridge toll of £0.60 (in each direction for all vehicles)
- remove the toll in both directions for all vehicles
- change the toll to £0.50 for Cars and £1.00 for OGVs (in each direction)
The conclusions from the study were:-
Conclusions (extracted from the consultant's report)
- When tolls are removed from the Erskine Bridge (Test B01 in 2006 and B02 in 2011) there is a large increase in traffic flow in both directions across the bridge itself, approximately 11,000 vehicles per day in each direction. A large proportion of this 'new' traffic crossing the bridge is re-routeing from the Clyde Tunnel which is itself reduced by approximately 2,400 vehicles per day in each direction. The effect of removing the toll on Erskine Bridge has a minor effect on the Kingston Bridge.
- The flow on Erskine Bridge increases slightly when the charge for cars is decreased and goods increased (Test C01 in 2006 and C02 in 2011). In this test, the effect on the Clyde Tunnel and Kingston Bridge is minimal.
- The test that produces the most significant change in results from the Reference Case is the removal of tolls as one would expect (as it is the largest change to the toll regime). The removal of tolls has an interesting effect on route choice. It was found that a higher percentage of traffic using the Erskine Bridge terminates in Glasgow, i.e. swapping their choice of routes from the North bank corridor of the Clyde (A82 and A814 etc) to the South bank corridor (M8 and A8 etc). This in turn provides a degree of congestion relief around the North bank corridor.
- The model output demonstrates that there is no consequent increase in delay in and around the Glasgow area for traffic re-routeing to the South bank of the Clyde (via the M8) from the North bank. This suggests that this traffic may link back to their original route choice closer in to Glasgow or that the redistribution of route choices maintains a status quo of link and junction delays, with no significant change in flow orientation that would cause significant junction delays along the North or South Clyde Corridor. Indeed, the results suggest that it would be more likely that increases in toll would cause delays along the North Bank of the Clyde as the removal of tolls attracts traffic away from the North Bank. This confirmed earlier modelling work done in Phase One.
- The removal of tolls does result in an increase in delay on and around the Erskine Bridge - both North and South Bridgeheads and on the secondary road network.
- The most significant diversion of vehicles occurs at Erskine Bridge itself (as expected). By the time traffic has re-routed as a consequence of the toll removal, the changes around Kingston Bridge (both increases and decreases depending upon time of day and direction of travel) are generally insignificant. Any predicted increases are estimated to be less than 100 vehicles per hour (although this may also be due to capacity restraint effects on Kingston Bridge).
- When the charge for crossing the Erskine Bridge is amended (to £0.50 for Cars and £1.00 for OGVs) as in tests C01 and C02 (for 2006 and 2011 respectively), a minor reduction in Annual Average Daily Revenue occurs, despite a slight increase in traffic utilising the Erskine Bridge. This is predominantly down to the make up of traffic using the bridge - more cars (paying less tolls) and less OGVs (paying higher tolls) use the bridge which nets out at a reduction in toll revenue.
- In the case of the removal of tolls, the percentage OGVs using Erskine bridge reduces from around 9% to around 5%.
- There are insignificant effects to changes in destination and mode choice within these Erskine Tests.
The following three maps show traffic congestion projected to 2011 during the AM peak on a typical day in Glasgow city centre and the surrounding area under the current tolling regime (Figure 1); the congestion in the same area if the tolls are removed (Figure 2); and the difference in congestion levels between the two scenarios (Figure 3).
Figure 1. Congestion in 2011 AM Peak: Current Tolling Regime.
Figure 2. Congestion in 2011 AM Peak: When Tolls are removed on Erskine.
Figure 3. Changes in Congestion in 2011 AM Peak: When Tolls are removed on Erskine.
For further details of this work refer to the full report entitled " Tolled Bridges Study: Phase Two - Erskine Bridge", July 2005. Prepared for the Scottish Executive by Consultants MVA. This is available on request.
2. Tay and Forth Road Bridges - Additional Work
In a similar manner, TMfS modelling has been applied to the Tay Road Bridge to investigate the effects of alterations to the tolling arrangements. In addition, the interaction of the Forth and Tay bridges has been considered with regard to through traffic through Fife for the scenario when both bridges are tolled one way, in the same direction. General observations have also been made regarding the interaction with other major bridges and the effect on toll revenue.
The specific tests that have been undertaken are as follows:
- Reference case - the status quo at all bridges.
- Test 1 - Two-way tolling on all bridges. On the Tay Road Bridge the toll is £0.40 for cars and £1.00 for goods vehicles in both directions. On the Forth Road Bridge the charge is £0.50 for cars and £1.00 for goods vehicles in both directions. Erskine Bridge remains unchanged.
- Test 2 - The toll plaza on the Tay Road Bridge is moved to the south end of the bridge and tolled in the northbound direction. All toll charges remain the same as they are at present.
- Test 3 - Tolls are removed from Tay, Forth and Erskine. The effects of removing the tolls at Erskine are examined in detail at section 1 above.
The conclusions from the study are:-
Test 1 - Conclusions
- The Tay and Forth Road Bridges both experience an increase in flow in the current toll direction as the toll decreases to accommodate the two-way toll. In the previously untolled direction, both bridges note a reduction in flow as the cost of using the bridge in this direction increases.
- The move to two-way tolling has knock on effects on both the Kincardine Bridges. There is also an effect on Friarton Bridge on the M90 (and to a lesser extent the M9 at Stirling). The flow on Kincardine Bridge increases significantly Southbound due to the introduction of the Southbound tolls on the Forth Road Bridge, but the flow decreases Northbound. Similarly Friarton Bridge flows increase Northbound and decrease Southbound as the Tay Road Bridge becomes tolled in both directions.
- The principal traveller response to the changes in toll regime is to change route choice. Mode and destination choice effects of altering the existing one-way toll regime are negligible.
- As a result of the introduction of two-way tolling both Forth and Tay Road Bridges experience a net drop in CO2 emissions, but Kincardine Bridge and Friarton Bridge both experience increases.
- The revenue generated as a result of two-way tolling has little effect on the Tay Road Bridge, but results in a slight increase in tolls over the Forth Road Bridge.
Test 2 - Conclusions
- Changing the toll on the Tay Road Bridge to become a Northbound toll rather than a Southbound toll decreases the Annual Average Daily Traffic ( AADT) flow Northbound on the bridge by around 15%, whilst there is an increase in AADT traffic Southbound over the bridge of around 19%. The overall two-way AADT flow on the Bridge is however largely unchanged.
- There are impacts that this tolling strategy would have on the Road network in Fife. Both the A91 and the A92 will note increases in the amount of Southbound traffic on them, whilst experiencing a decrease in Northbound flow.
- The principal traveller response to the changes in toll regime is to change route choice. Mode and destination choice effects are negligible.
- There are slight decreases in emissions of CO2 on the Tay Road Bridge and the M90 Southbound, but there are increased emissions on the M90 Northbound.
- The Annual Average Daily Revenue ( AADR) generated in this test decreases by just under £1,000 in both modelled years.
Test 3 - Conclusions
- The removal of tolls from all the tolled bridges gives rise to large increases in flows across each (previously tolled) bridge.
- As the most sensitive traveller behaviour to the removal of tolls is to change route, as a result of the increase in flow at the Forth and Tay, there are consequent reductions in flow on some of the alternative routes. However, these reductions are not large enough to substantiate the increases; additional traffic is attracted to the untolled bridges as a result of mode choice (to change from PT to Car), destination choice (for example, as a long term consequence to move house or place of work for Home Based Work travellers, to move shopping/leisure locations for Home Based/Non Home Based Other travellers) or induced traffic.
- As a result of these increases in flow, there are large increases in Congestion and Emissions (both NOx and CO2) over each of the Bridges. This results in decreases in congestion and emissions elsewhere in the road network.
- For Forth and Tay, the noticeable increases in Congestion and Emissions occur on the A91 in Fife, and on the Edinburgh City Bypass. The main decreases of note are on Friarton Bridge, the A90 near Dundee, and around Kincardine Bridge.
- There is no toll revenue generated as a result of this test.
The following maps show headline results of this modelling for Forth and Tay - the congestion experienced at both bridges projected to 2011 during the AM peak on a typical day under the current regimes (Figures 4 & 7 respectively); the congestion that modelling suggests would be experienced if tolls were removed (Figures 5 & 8 respectively); and the difference in congestion levels between the two scenarios (Figures 6 & 9 respectively). Two additional maps for Tay show the congestion levels that modelling suggests would be experienced if the toll booths were moved to the south side of the bridge (Figure 10), and the difference between those congestion levels and the congestion experienced with the toll booths in their current position (Figure 11).
Figure 4. Congestion in 2011 AM Peak: Current Tolling Regime.
Figure 5. Congestion in 2011 AM Peak: When Tolls are removed on bridges.
Figure 6. Changes in Congestion in 2011 AM Peak: When Tolls are removed on bridges.
Figure 7. Congestion in 2011 AM Peak: Current Tolling Regime.
Figure 8. Congestion in 2011 AM Peak: When Tolls are removed on bridges.
Figure 9. Changes in Congestion in 2011 AM Peak: When Tolls are removed on bridges.
Figure 10. Congestion in 2011 AM Peak: When toll booths are moved to the south side of the bridge and toll collected northbound.
Figure 11. Changes in Congestion in 2011 AM Peak: When Tay is tolled northbound and toll booths situated on the south side of the bridge.
For further details of this work refer to the full report entitled " Tolled Bridges Study: Phase Two - Additional Work", August 2005. Prepared for the Scottish Executive by Consultants MVA.
3. Car Occupancy over the Forth Road Bridge
The primary aim of this study is to investigate possible measures that can be developed to increase car occupancy over the Forth Road Bridge, particularly during the peak periods. A large number of schemes under consideration around the world, and in many cases under successful operation, have been examined with the specific intention to promote application of the methodology to the Forth Road Bridge.
The main objectives of the study were as follows:
- Literature Review of potentially relevant schemes/applications including website information and advice from authorities such as the International Bridge Tunnel and Turnpike Association ( IBTTA).
- Review of historic schemes and proposals relating to car occupancy that have been considered for the Forth Road Bridge with comments on their applicability, e.g. "Forth TRIP: Radical Measures Study" and "Forth Road Bridge Short Term Measures Study".
- Discuss practical application of HOV lanes concept for the Forth Road Bridge.
- Discuss practical aspects of differential tolling and the concept of higher toll charges for SOVs. Highlight any problems for implementation using the existing plaza layout and investigate the potential of electronic tolling methods with regard to HOVs and SOVs.
- Investigate any similar bridge scenarios in the world ( i.e. two lane dual carriageway) where a workable scheme to reduce single occupancy car usage has been successful and discuss applicability to the Forth Road Bridge.
- Any other observations, general recommendations and advice on cost effective solutions.
A summary of the report of these findings has been taken from the report and is reproduced below.
Conclusions (extracted from the consultant's report)
- The Forth Road Bridge presently operates at capacity during peak times with an average daily traffic flow of approximately 65,800 vehicles in 2004, of which some 91% were cars and light goods vehicles and 6% were heavy goods vehicles over 3500 kg. The remaining vehicles comprised motorcycles, buses and exempt vehicles, with a negligible number of escorted vehicles.
- During peak periods, approximately 70% of vehicles crossing the bridge are single occupancy vehicles.
- The results from the research presented in this report indicate that demand exceeded bridge capacity on approximately 270 days during 2003 and notes that traffic volumes are projected by FETA to increase by between 2% and 3% per annum over the next 5 years.
- Previous analyses have confirmed that simply removing the tolls will not increase bridge capacity as two lanes on the bridge can accommodate 3,600 veh/hour and the bridge's seven manual tollbooths can process up to 520/hr each, which equates to a total of 3640 veh/hour. As a result, a strategy to encourage high occupancy vehicles ( HOVs) on the bridge or increase public transport usage seems the most likely solution to address the delays and congestion experienced by road users.
- HOV lanes have become widely adopted in the United States since they were introduced in the 1970s. There are over 125 HOV projects in 30 cities, covering over 2,500 lane-miles and carrying more than 3 million commuters a day.
- The possibility of using HOV lanes on the Forth Road Bridge is potentially attractive because of the opportunities it offers to reduce the volume of single occupancy vehicles crossing the bridge during peak periods. However, results from the research presented in this report indicate that there are only a few bridges in the world that incorporate an HOV facility, including two bridges in Canada and seven in the United States, of which two have a similar traffic demand to that on the Forth Road Bridge. The Dumbarton Bridge Toll Plaza and the Coronado Bridge Toll Plaza, both in California, have average daily traffic flows of approximately 61,000 vehicles and 68,000 vehicles respectively. However, these bridges differ from the Forth Road Bridge in that the Dumbarton Bridge has three lanes in each direction and the Coronado Bridge has two lanes in each direction and a reversible median HOV lane.
- As the Forth Road Bridge is limited to two lanes in each direction, with no hard shoulders, developing an HOV lane would require the physical separation of the two lanes over the bridge to create one HOV lane and one general purpose lane and the direction of HOVs through the northbound HOV tollbooths to access the dedicated HOV lanes.
- The provision of a dedicated HOV lane in addition to general purpose lanes can be a cost effective solution on a land based roadway where sufficient land is available to widen or upgrade a road. However, due to the limited road space available on the Forth Road Bridge, the results of previous studies have concluded that this is not technically viable.
- In the United States, a new congestion management technique has been developed using high occupancy toll ( HOT) lanes which combine the HOV approach with pricing strategies to allow SOVs access to HOV lanes by paying a toll. These are limited access: normally barrier separated roadway lanes that provide free or reduced cost access to qualifying HOVs and also provide access to other paying vehicles which do not satisfy the passenger occupancy requirements. By using price and occupancy restrictions to manage the number of vehicles travelling on these lanes, HOT lanes can maintain volumes consistent with uncongested levels of service even during peak periods, and eliminate the "unused lane syndrome" normally associated with HOV schemes.
- The "unused lane syndrome" is a characteristic that can exist when HOV lanes are introduced on roads where the volume of HOVs is significantly lower than the volume of SOVs. In this case, the HOV lane can be relatively lightly trafficked or under-utilised compared to the SOV lane.
- The Forth Road Bridge does not presently exhibit this characteristic because traffic is free to use both lanes. However, if an HOV lane were introduced on the Forth Road Bridge, setting aside an existing lane for this purpose, it is feasible that the "unused lane syndrome" would appear with a correspondingly long queue in the SOV lane. Results from the Phase 1 study indicate that the exclusion of SOV traffic from the dedicated and more lightly trafficked HOV lane could generate queues in the general purpose lane of more than 7km.
- A strategy based solely on occupancy pricing on the general purpose lanes through the introduction of differential charges at the toll collection point would assist in discouraging single occupancy vehicles on the Forth Road Bridge, where road users have the option of car sharing, whilst balancing traffic demand across the available lanes.
- However, in the United States, HOT lanes, which are still at the experimental stage, are considered to be the next development of the HOV lane operation where SOVs can choose to pay a higher toll to use an HOV facility. In the case of the Forth Road Bridge, this could apply where HOV toll booths are used by SOVs to save time by avoiding delays at the SOV toll booths.
- Implementing an HOT scheme on the Forth Road Bridge could benefit from an electronic toll collection ( ETC) and traffic information system to facilitate variable, real time toll pricing of non- HOV vehicles. Hence different charges could apply to SOVs and HOVs depending on vehicle occupancy and time of travel. All relevant information could be clearly displayed on variable message signs on the approach to the toll collection points.
- An integrated ETC and vehicle occupancy detection monitor could enforce the HOT lane to deter potential violators. Such a system would monitor vehicle occupancy and toll pricing and record when incomplete or anomalous transactions occur at the tollbooths.
- It should be noted that the information presented in this report represents an overview of the possible measures to increase car occupancy on the Forth Road Bridge and that more detailed work, including investigating operating conditions on the approach to the toll booths, would be required to assess the extent to which implementation of these measures would be successful.
For further details of this work refer to the full report entitled " Overview of Measures to Increase Car Occupancy over the Forth Road Bridge", August 2005. Prepared for the Scottish Executive by Consultants Scott Wilson (Scotland) Ltd..