Controversy has developed surrounding the roads staff at the City of Hamilton and whether a technical report illuminating substandard surface conditions on the Red Hill Valley Parkway was intentionally hidden. But other technical reports have been received by the City which need further clarification.

In July of 2017 Nicole O’Reilly, a reporter with the Hamilton Spectator newspaper, wrote an article questioning why there were so many collisions on the Red Hill Valley Parkway (RHVP) in Hamilton, Ontario. Ms. O’Reilly noted that in the “last five years” the RHVP experienced twice as many crashes than the adjoining Lincoln Alexander Parkway (LAP). She also noted that the RHVP carried a lower traffic volume than the LAP thus the collision numbers should have been reversed. This article included a figure using data from the City of Hamilton. The figure used different colours and sizes of circles to help highlight where differences in collision rates and severities existed.

Missing Report & CIMA Reports

In early February, 2919, considerable controversy developed when the City of Hamilton issued an “apology” to the public with respect to the finding of a technical report from Tradewind Scientific (2013). The apology was needed because the Tradewind report had indicated that the surface friction of the RHVP was substandard and it appeared as if someone had attempted to hide the report and its findings. The report was reportedly found by City staff in late 2018. But its finding was not revealed until the apology was issued on February 6, 2019. These facts were reported in February, 2019 articles written by Samantha Craggs, a reporter with CBC News and also by Hamilton Spectator reporter, Matthew Van Dongen. Collision statistics were quoted in these articles of a similar nature to the original article by Ms. O’Reilly in July of 2017.

Subsequently it became known that a roadway safety consulting firm, CIMA, had produced a report in 2015 entitled “Red Hill Valley Parkway Detailed Safety Analysis”. In that report CIMA mentioned that in 2013 they “conducted a safety review of the section of the RHVP between the Dartnall Road and Greenhill Avenue interchanges, providing a series of recommendations to the improve safety”. It would seem logical that the 2015 report and the 2013 safety study would have been requested by the City of Hamilton although that was not explicitly stated.

The 50-page, 2015 CIMA report was extensive. It covered matters such as collision statistics, field investigations and illumination studies. Various countermeasures were also reviewed. Beyond the 50 pages that were the main body, several large Appendices provided additional details such as the spacial distribution of collisions involving wet versus dry surface conditions and median-related collisions according to road surface conditions. Despite all these details there was no mention of any results of surface testing and the results of the Tradewinds Scientific report of 2013. Instead, in a sub-section of the CIMA report entitled “Conduct Pavement Friction Testing” the authors made the following recommendation:

“In order to determine whether low pavement friction may be contributing to collisions (especially wet surface), the City should consider conducting pavement friction tests under normal conditions as well as under typical wet pavement conditions encountered on the RHVP. Special focus should be given to the curves near King Street and the Queenston Road interchanges…”

Thus it would appear unusual that CIMA, which was presumably contracted by the City of Hamilton to conduct a detailed study of the road safety conditions on the RHVP, completed their report in 2015,  yet they appeared to have no knowledge of the fact that Tradewinds Scientific had completed a report on the surface conditions of the RHVP two years earlier, in 2013, presumably for the City of Hamilton.

In a Hamilton Spectator article written by Nicole O’Reilly in July of 2018, it was reported that samples of pavement were being taken from both sides of the RHVP during that month. The article also reported that “smaller samples had also been taken in December of 2017. Part of the motivation for this sampling was because the City was aware that “some people felt the RHVP was slippery”. The current Director of Engineering Services for Hamilton, Gord McGuire, stated that the December 2017 samples were inconclusive, much like earlier testing of the RHVP’s surface in 2015.

The O’Reilly article also indicated that “cat’s eyes” had been installed on the RHVP in 2015 but that those became “ineffective” and would be re-installed at a cost of $100,000. It would seem unusual that such an installation would require such a large scale replacement in less than three years of service.

It then became known that CIMA produced another report on the RHVP dated January, 2019. It is not clear why this additional report was required since detailed analysis and recommendations existed in the 2015 CIMA report. The closest explanation as to why this additional report was needed was provided by in the 2019 CIMA report as follows:

“The present study takes into account the findings and recommendations of the previous studies, in order to confirm or expand the recommendations to reduce roadside related collision frequency and severity.”

Thus it would appear that the 2019 CIMA report was intended to focus specifically on roadside safety issues. Again, the City of Hamilton would appear to have been the requesting agency for this report however there is no explanation why the City of Hamilton was interested in this specific issue of roadside safety to the degree that a separate and additional report was needed.

Over-Representation of Higher Severity Collisions on the RHVP

The 2015 CIMA report made the following comments regarding collisions on the RHVP:

“The study area experienced a total of 474 collisions during the period from January 1, 2008 to July 23, 2015. The data, broken down by collision severity, in summarized in Figure 2. There were 4 fatal collisions (resulting in 5 fatalities), 205 injury collisions, and 265 Property Damage Only (PDO) collisions.”

These statistics can  be reported in terms of percent of total collisions as follows; fatals = 0.8%, Personal Injury = 43.0% and Property Damage = 55.9 %.

With respect to the 2019 report the following statistics were provided describing the RHVP’s historical collision rates.

“Collision records were provided by the City in digital format for the five-year period between 2013 and 2017. After removing collisions out of scope (e.g. occurring at intersections/ramp terminals) and duplicate records from the data set provided, a total of 939 collisions were reported to occur along the RHVP mainline…”

Individual year collision totals:

2013 – 143

2014 – 128

2015 – 254

2016 – 203

2017 – 211

“The mainline presented 704 (75) property damage only (PDO) collisions, which include self-reported (SR) collisions, 231 (24.6) non-fatal injury collisions; and 4 fatal injury collisions (0.4%).

42% of mainline collision records do not have information about light, environment and road surface conditions. These correspond to self-reported collisions, which do not contain this information”.

Comparing the data reported in the 2015 CIMA report with the 2019 report, there appears to be a contradiction. CIMA indicated (above) that there were 474 collisions between 2008 and July, 2015. Yet in the 2019 CIMA report, when the collisions are added up for the years 2013, 2014 and 2015 (143+128+254) the sum is 525 collisions. While the years being reported in the two reports do not coincide clearly there is something unusual. If we take half of the 254 collisions for half of the year 2015 resulting in 127 collisions, we can then add the collisions for the full years of 2013 (143) and 2014 (128) resulting in a total of 398 collisions. So 398 of the 474 collisions would be accounted for. This would mean that there would have to have been only 76 collisions along the RHVP in the five years between 2008 and 2012, or an average of just 15.2 collisions per year. Rather than accept that there was a sudden and vast increase in the numbers of collisions on the RHVP after the year 2012, it is more likely that these data have been reported incorrectly.

Never-the-less this collision data in the 2015 and 2019 CIMA reports can be compared to historical data from the Province of Ontario as reported in the Ontario Road Safety Annual Report (ORSAR) which are shown in the table below.

Based on provincial trends the 2008-15 RHVP collisions would appear to have higher than average rates of fatal and personal injury collisions. This can be observed from the Provincial percentages showing a progressive drop in fatal collisions such that by 2014 the percentage was 0.22% whereas the RHVP percentage works out to be 0.84%. Similarly provincial percentages for Personal Injury collisions were dropping to 17.6% by 2014 whereas the RHVP percentage works out to 43.25%.

Similarly increased percentages in Fatal and Personal Injury collisions are seen in the 2013-17 RHVP data. The RHVP percentage of Fatal collisions was reported in the CIMA report as 0.4%. In fact, taking into account the decimal points, the actual RHVP percentage of fatals would be 0.426 %. The RHVP Personal Injury percentage was 24.6 %.

In July of 2019 the Ontario Provincial Police (OPP) released statistics for collisions occurring in the first six months of 2019 for roadways patrolled by the OPP. The data indicated that total collisions were 33,839, of which 4272 were Property Damage and 188 were fatals. By inference, there would have been 29,449 Property damage collisions.  These data result in the following percentages of total collisions: Fatals = 0.35%, Personal Injury = 12.62%, and Property Damage = 87.03%. Typically, OPP patrolled highways would be expected to contain the highest rates of fatal and personal injury collisions because they involve rural, two-lane highways which traditionally involve the highest rates of major collisions. Yet the RHVP would appear to have higher rates of serious collisions that the OPP data. So one would question why this is so or whether there is something unusual or inaccurate in the data.

A clue indicating the quality of the collision data described in the CIMA reports can be taken from the following comment made in the 2019 CIMA report:

“For all of these factors, 42% of the mainline collisions and 32% of the ramp collisions have “unknown” values. These correspond to self-reported collisions, which do not contain this information.”

What this wording is saying is that 42% of the RHVP collision data from the City of Hamilton is based on self-reported collisions. Self-reported collisions are those where the involved drivers attend a collision reporting centre and fill out answers to a set of questions describing how the collision occurred. While it is often claimed that a police representative at the reporting centre can reconstruct how such a collision occurred from those responses the reality is quite different. Even when experts are retained to examine all the physical evidence and conduct detailed analyses there is still substantial disagreement amongst them. This is so even though these experts are typically examining higher-severity collisions that result in substantial physical evidence that can be used in the analysis. When collisions occur that are of a lower severity, such as those at a Self-Reporting Centre, there is less physical evidence and it is therefore more difficult to reconstruct. As no official, independent entity visits the collision site there is no certainty whether the events being described by the involved drivers match the physical evidence. Vehicle damage photographed at the police reporting centre is of some assistance but it is insufficient in most instances to identify key elements of a collision such as the energy or momentum that was involved.

Additional collision data for Hamilton, Ontario were reported in a CHCH-TV article in February 6, 2019. CHCH reported that:

“Between 2013-2017 the total number of collisions steadily increased from 7533 to 8802. However, collisions resulting in injuries dropped 13% between 2016-2017. Fatal collisions hovered between 14-16 per year. 31% of those fatals were single vehicle collisions and 44% of all fatals were due to the driver losing control.”

Once again we can use this data to examine the percentage of fatal collisions with respect to total collisions for the City of Hamilton. This results in 0.17%. This is lower than the Provincial average of 0.22% (noted above) but far lower than the percentage of 0.84% for the RHVP.

Furthermore the last portion of the quote (“31% of those fatal were single vehicle collisions and 44% of all fatals were due to the driver losing control”) reveals additional concerns. Gorski Consulting has been involved in a research study approaching 10 years now where a comparison has been conducted at a specific site in London, Ontario between the number of collisions officially reported in police data versus the actual number of collisions and loss-of-control incidents that have occurred based on the physical evidence that has been documented at the site. The results of this study indicate that at least 80% of such collisions and loss-of-control incidents are not reported in official police data. And, due to limitations of our procedures, it is not possible to capture all of those collisions and incidents such that the actual number of unreported events could be substantially higher than 80%.

When combining these facts the concerns should be obvious. Contradictions appear to exist between the numbers of collisions on the RHVP reported in the two CIMA reports. Self-reported collisions that constitute a large part of the data. A large number of collisions that are not accounted for in the police data.  What this says is that there is an unreasonable belief by many researchers and experts in the accuracy and precision of the collision data collected in police files.  Reports quote this data as if it is highly accurate, without informing readers of the potential inaccuracies that may exist.

Curve Radii on the RHVP

In its discussion of the curve radii on the RHVP, CIMA’s 2019 report quoted recommendations taken from the 2017 version of Transportation Association of Canada (TAC) manual manual entitled “Geometric Design Guide for Canadian Roads”. The following text and Table 12 are taken from the CIMA report:

Superelevation is designed into a curve to create a centripetal force toward the inside of the curve that is developed from the earth’s gravitation pull due to the banking of the cross-slope of the road surface. By this method the frictional force between the tires road surface is assisted by this banking in pulling the vehicle toward the inside of the curve. Obviously  a curve designed with no superelevation will be much different than one with the maximum of 6% superelevation. With no superelevation the frictional force between the tires and road surface is the only force available to move a vehicle in a curved motion through the curve. Thus it is important to know the characteristics of the road surface and whether it can provide a reasonable level of friction. But it is also important to find out what superelevation exists in order to obtain a complete picture of the safety of a curve. Furthermore, the introduction of superelevation must be done in controlled manner. Without inquiry a superelevation may exist that does not confirm with standard design and this could contribute to collision causation.

The 2015 CIMA report recommended that the City of Hamilton conduct surface friction testing thus the CIMA authors did not have that data. Its 2019 report also did not specify what the specific superelevation was at the 420 metre curve south of King Street. Furthermore there was no indication from CIMA whether they conducted testing of how the superelevation was incorporated into the curves of the RHVP. Knowledge of these facts is crucial in determining whether the curves along the RHVP provide a reasonably safe environment. To say that the 420-metre-radius was compatible with the design speed says nothing if there was insufficient surface friction, if the curve did not have a sufficient superelevation, or if that superelevation was developed improperly. This is the crucial information that is missing from the CIMA reports.

In a footnote of the 2019 CIMA report it was indicated that in their 2015 report incorrectly reported that the Design Speed for the RHVP was 100 km/h. Their explanation for this error was as follows:

“The 2015 study report indicates 15% to 22% of drivers at or exceeding the design speed of the road (110 km/h) in the northbound and southbound direction, respectively. At the time, the design speed information had not been provided to CIMA and was assumed as 110 km/h based on the operating speeds.”

This represents an unusual and important failure. Knowing the Design Speed of the RHVP was critical to CIMA’s evaluation of the road’s safety. Simply assuming a Design Speed without obtaining further information about this crucial element would have misinformed the City of Hamilton in a substantial way. The assumption that was made based on the observed operating speeds shows a lack of understanding that a four-lane, controlled-access expressway like the RHVP does not look much different to an average driver than any 400 series highway in Ontario and thus, without police enforcement and monitoring, drivers would tend to travel at similar speeds to what they would on those 400 series highways. Assuming Design Speed based on this observation of operating speed is difficult to comprehend.

Differences In Character of the Lincoln Alexander and Hill Valley Parkways Could Explain Collision Frequencies

Road surface friction is not the only factor that may explain why there appear to be more collisions on the RHVP versus the LAP. Differences in the geometry of these roadways could also be a contributing factor. The LAP is essentially straight and level. In contrast the RHVP contains significant horizontal curves while also crossing through the elevation change that is the Niagara escarpment. These facts should create differences in the demand for tire force as vehicles travel through each Parkway.

There is a low demand for tire force when a vehicle is travelling in a straight line at constant speed. And there is a high level of tire force available on a new and dry surface.  A well-maintained, newer vehicle with good tires and suspension will also be helpful in maintaining a vehicle’s stability. However these ideal conditions will not always exist. The ideal friction in the range of 0.8g of a typical, new, asphalt surface could be reduced substantially when the surface becomes wet, to a range of 0.5g or lower. Snow cover might reduce that friction to a range of 0.2g and icy conditions might bring it to values as low as 0.05g. Driver actions such as changing lanes, braking or accelerating need tire force to complete those actions and that demand may challenge the available limits at times of low surface friction. When vehicles travel down the Niagara escarpment of the RHVP they need to counter the acceleration provided by the earth’s gravitational pull by applying braking. This braking is not needed when travelling along the generally level geometry of the LAP. Similarly, when vehicles climb up the slope of the RHVP tire force is also needed to counter the earth’s gravitational pull, thus drivers need  to apply their accelerator pedals more so than travelling along the LAP. These differences in the geometry of the RHVP and LAP need to be acknowledged and taken into account when discussing differences in the collision histories of the two Parkways.

ET-Plus Guardrail Terminals Along RHVP Warned as Safety Risk

The 2019 CIMA report finally acknowledged something that official agencies have failed to acknowledge with respect to the ET-Plus guardrail terminals that exist along most highways in Ontario and which are the dominant installations along the RHVP. Without naming the apparatus the CIMA report referred to the terminals with the “4-inch” wide channels as being substandard, as shown in the photo below, taken from the 2019 CIMA report. There are no other terminals along the RHVP with 4-inch wide channels other than the ET-Plus terminals.

CIMA is the first to admit that guardrail terminals such as the ET-Plus are substandard.

This demonstrates the difficulty in conducting a fully independent and public study of roadside installations that could be dangerous to the public. The manufacturers of the ET-Plus have been involved in a well-publicized, legal battle in the U.S. over claims that the ET-Plus is unsafe and, at times, harpoons striking vehicles rather than allowing the terminal to ride along the rail as it is shown to do in controlled testing. Large companies with the legal clout to commence civil suits over any negative pubicity with respect to their products make it difficult for sufficient information to be brought forward to the public so that an impetus can be generated to request an independent investigation into potential safety issues. Intense lobbying and misinformation from all sides makes it difficult to come to a clear understanding whether installations such as the ET-Plus should be removed from all highways or whether they are similar in their failures to most other fixtures on the roadside.

As an example, a photo taken by a Hamilton Spectator newspaper reporter showed a situation where an ET-Plus terminal was struck on the RHVP near Barton Street. The rail buckled and folded over such that the folded end of the rail penetrated into the driver’s door of the striking vehicle. This is the kind of mechanism where past fatalities have occurred. Fortunately the collision was not severe enough to cause the rail to penetrate further into vehicle’s occupant space but there was no information about the driver’s injuries. However, with a slightly higher severity the consequences could have been substantially more serious.

In a March 8, 2016 field study of the RJVP Gorski Consulting documented a number of terminals that were impacted, damaged and remained unrepaired. A common mechanism of failure with the ET-Plus appears to be that, when the terminal is struck, the rear edge of the terminal comes down on the top edge of the rail and collapses it. This collapse deforms the rail before it has a chance to pass through the narrow throat of the terminal. It is the passing of the rail through the throat that causes its controlled flattening or “ribboning” thus dissipating the collision energy. The collapse of the top edge of the rail before reaching the throat causes the rail to jam in the throat of the terminal instead of passing through it. Once the rail is no longer able to pass through the throat the terminal is not able to ride on the rail. This mechanism has not been previously identified in the public domain and further studies should have been made by any entity using this terminal on a public highway. Some of the  instances damage of terminals on the RHVP from March 8, 2016 are shown in the photos below.

Example of a damaged ET-Plus terminal in the median of the RHVP near Queenston Road. This photo was taken on March 8, 2016. The terminal has been dislocated even though there has been a minimal length of rail that passed through it.

Example of an ET-Plus terminal that was struck on the RHVP near Greenhill Ave. While this was a mild impact it caused a buckling of the top end of the guardrail which is suspected to be the mechanism of failure in many similar impacts.

This view of the damaged ET-Plus terminal near Greenhill Ave shows the buckled area of the top of the guardrail which becomes jammed.

This close-up view of the measured extent of the buckling of the top edge of the rail. When buckled this railing has difficulty passing through the throat of the terminal.

The 2019 CIMA report recommended that the ET-Plus terminals be replaced by Mash Sequential Kinking Terminals (MSKT). CIMA has not provided any explanation why this replacement is necessary. Although it may be valid and would appear to match concerns observed in Gorski Consulting field studies, such a recommendation requires further explanation by CIMA.

Gorski Consulting Testing Along the RHVP

Gorski Consulting has conducted recent testing along the RHVP and the LAP and this has been reported in several news articles uploaded to News page of the Gorski Consulting website. The Gorski testing involved driving through the RHVP while documenting the extent of motion caused to a test vehicle. The results of the testing showed that the LAP produced more vehicle motion than the RHVP. This generally indicated that the surface  of the LAP was more uneven than the RHVP. This point needs to be considered when discussing the need to conduct re-surfacing of the RHVP.

It should be recognized that the surface friction testing in the Tradewind report is different from the testing performed by Gorski Consulting. The Gorski Consulting data does not evaluate the force generated between a vehicle’s tires and road surface. The Gorski Consulting testing demonstrates how disturbances are created in the motion of a test vehicle that could affect the tire force. It needs to be emphasized that, regardless of the friction available when the full weight is loaded on the tires, when the mass of a vehicle is lifted the tire force drops and the effective friction drops. This drop in tire force can be irregular and chaotic when bumps, depressions, patches, etc. are encountered in a random fashion. At times it may be predominant only at one tire, one axle, one side of the vehicle or it may involve all four tires at the same instance. This change in tire force from motion can be combined with other factors such a braking, acceleration or changing lanes which require tire force to complete their effects. All these influences cause further differences in tire force. Further problems can exist from changes in certain roadway features such as superelevation or curve radius.  Thus, while road surface friction is an important matter, it cannot be viewed in isolation as the only factor that matters in collision causation. The Gorski Consulting data shows that the irregularities of the road surface can be relevant in causing collisions particularly in conditions (wet, snow-covered or icy surfaces) where the typically high tire forces under dry conditions are reduced.