Transcript B. Mills, J. Andrey, S. Tighe, S. Baiz

Weather information and surface
transportation in Canada:
The long and winding road
Brian Mills1,2, Jean Andrey2, Susan Tighe3, and Sarah Baiz3
1Adaptation
& Impacts Research Division, Environment Canada ([email protected])
2Department
of Geography & Environmental Management, University of Waterloo
3Department
of Civil & Environmental Engineering, University of Waterloo
Introduction
•
•
“Pushing the Product”
•
The “Bottom-up”
perspective is a
necessary complement if
not starting point
•
A few examples from
road transport
What happens after the
flush?
Introduction
•
Economic and social activities in Canada are highly
dependent on road surface transportation—by far the
most important mode
Value of Canada-U.S. Trade by Mode (2007)
Total Trade (CA$569,821 million)
Air
5%
Other
14%
Marine
4%
Rail
17%
Source: Transport Canada 2008
Road
60%
Introduction
Maintaining the mobility afforded by the highway system
without compromising safety or environmental quality
requires substantive investments—many of which are
weather-related

Design, construction and maintenance of infrastructure

Operations
Fiscal Year
Local
Provincial/Territorial
Federal
20
06
-0
7
20
05
-0
6
20
04
-0
5
20
03
-0
4
20
02
-0
3
20
01
-0
2
20
00
-0
1
Safety interventions
18000
16000
14000
12000
10000
8000
6000
4000
2000
0
19
99
-0
0

Environmental
Millions CA$

Road Transportation Expenditures by Level of Government
19
98
-9
9
•
Introduction
Andrey 2009
•
Despite numerous interventions, significant risk remains
•
Why? Imperfect decision-making? Role of wx info?
Weather & Climate-related Decisions
at Many Scales
• Drivers
• Public & commercial transport service providers
• Public road authorities
• Road associations
• Construction & maintenance industry
• Vehicle manufacturers
• Vehicle repair industry
• Insurers
• Police/enforcement agencies
• Emergency responders and healthcare industry
• Weather, road weather, and hydromet service providers
Case 1: Weather-related collision risk
•
Robust estimates of the relative risks and social costs
associated with driving in inclement weather are lacking at
the city-region and national scales in Canada.
•
This information is fundamental to design and evaluate
the efficacy of response measures such as the provision
of weather information intended to influence driver
behaviour just before and during a particular trip.
D. Doiran, National Post
Method
•
National Transportation Accident Incident Database
(TRAID) collision data (1984-2000) combined with hourly
and six-hourly records of precipitation (R, S, ZR/ZL,
mixed) for 28 Canadian cities
•
Matched pair event-control analysis conducted producing
~ 36,000 entries
•
Relative risk calculations performed by dividing the sum
of injury collisions/injuries during events by
corresponding counts for controls
•
Further analysis facilitated the development of risk
estimates disaggregated by precipitation type, amount,
injury severity, region, etc.
Results
•
Risk of injury increases by approximately 70% during
precipitation relative to dry seasonal conditions
•
Minimal and minor injuries tend to increase more than do
major and fatal injuries
•
About 200-400 fatalities and several thousand injuries are
attributable to weather-related motor vehicle collisions
each year with an estimated social cost >CA$1 billion
Relative Risk of Different Severities:All 28 Cities
Relative Risk
3.0
2.5
Minimal
2.0
Minor
1.5
Major
1.0
Fatal
0.5
0.0
All
Rain
Andrey et al. 2005, Andrey 2009
Snow
Freezing Rain
Rain mixed
with Snow
Results
•
Comparative analysis using insurance claim data (19992002) for Winnipeg, Manitoba
•
Similar results for incidents (below) and claim costs
except Accident Benefit Costs during snowfall events
(RR=3.33)
Relative Risk (odds ratio and 95% CI)
TRAID and MPI Insurance Comparison for Winnipeg (1999-2002)
3.1
2.9
2.7
2.5
2.3
2.1
1.9
1.7
1.5
1.3
1.1
0.9
0.7
0.5
Rain-TRAID
Mills et al., in press
Rain-Insurance
Snow-TRAID
Snow-Insurance
Total-TRAID
Total-Insurance
Future Research
•
Complete RR analyses using insured loss data and
weather warning/advisory information
•
Develop a prototype collision prediction model that can
incorporate probabilistic weather prediction information
and produce a normalized index
•
Evaluate the effect of this “impact-centric” information on
stated/observed driver response relative to traditional
types of weather and road weather information
Case 2: Seasonal Load Restrictions
on Ontario Highways
•
Secondary roads are often subjected to heavy loads from
agricultural or resource extraction (forestry, mining) operations
•
Where frost penetrates into the subgrade, such highways are
extremely vulnerable to damage caused by brief periods of thawinduced weakening
•
Seasonal load restrictions (SLR) are used by transportation
agencies to reduce/increase the permissible loads carried by
trucks
•
Fixed dates/durations are often used in establishing SLRs which
provides certainty for trucking operations but can lead to
tremendous damage when thaws occur outside of the period
Case 2: Seasonal Load Restrictions
on Ontario Highways
Source: MTO, 2005
Method
•
Empirical model was developed to predict frost and thaw
depths as a function of simple freezing and thawing
indices derived from air temperature
•
Model validated and calibrated against data obtained for 2
winter seasons at 2 instrumented test sites periodically
evaluated for pavement strength using a portable Falling
Weight Deflectometer
Results (Northeast Ontario Site)

FDi  22.1  4.2 FI i  2.6 TI i
0  i  i0  

TDi  0.494  0.038 FI i  0.675 TI i

FDi  145  0.85 FI i  0.01 TI i
i  i0  

TDi  848  24 FI i  13 TI i
Where:
i
Number of days after the day indexed as day i = 0
i = 0 Day on which TAir first falls below 0ºC
io
Day of transition from Freezing to Thawing season
FDi Depth of frost on day (cm)
TDi Depth of thaw on day (cm)
FIi
Freezing Index value on day (in ºC -days)
TIi
Thawing Index value on day (in ºC -days)
Future Research
•
Repeat analysis and refine models using additional winter
seasons and locations
•
Develop a damage model and SLR/WWP decision
experiment using weather forecast data. Evaluate social
costs and benefits as a function of accuracy.
Case 3: Impacts of climate change on
pavement infrastructure
•
Current and past pavement designs generally assume a static
climate whose variability can be adequately determined from
records of weather conditions which normally span less than 30
years and often less than 10 years
•
Anthropogenic climate change challenges this assumption and
raises the possibility that the frequency, duration or severity of
thermal cracking, rutting, frost heave and thaw weakening may be
altered leading to shifts in pavement deterioration rates if
corrective actions are not taken
Method
•
Mid-century surface temperature and precipitation
scenarios were developed by statistically downscaling
output from the CGCM2A2x and HadCM3B21 climate
experiments for 17 Canadian sites
•
Scenarios were applied to 2 deterioration-relevant
indicators: 1) Performance Grade Asphalt Cement (PGAC)
high and low temperature threshold criteria, and 2) Freezethaw indices
•
Scenarios were applied at 6 sites using the MechanisticEmpirical Pavement Design Guide (MEPDG) model which
simulates life cycle deterioration (developed by the U.S.
NCHRP and AASHTO)
Results
Indicator analysis suggests that low temperature cracking
will become less problematic; structures will freeze later
and thaw earlier with correspondingly shorter freeze
season lengths; and higher extreme in-service pavement
temperatures will raise the potential for rutting.
Design (98% reliability) Minimum/7day Mean Pavement Temperature
(°C)
•
60.0
50.0
40.0
30.0
20.0
10.0
0.0
-10.0
-20.0
-30.0
-40.0
-50.0
Base
CGCM2A2x
HADCM3B21
SuperpaveTmax-98%annual
SuperpaveTmin-98% annual
Ontario RWIS Tmax-98%annual
Ontario RWIS Tmin-98% annual
Results
•
MEPDG analysis suggests that rutting (AC and total) and cracking
(longitudinal and alligator) issues will be exacerbated by climate
change
•
Maintenance, rehabilitation or reconstruction will be required
earlier in the design life
•
Absolute impacts of climate change are closely associated with
the underlying structural, material, and traffic characteristics of a
particular site thus generalizations must be considered with
caution.
10
9
7
6
5
4
3
2
1
97
10
9
12
1
13
3
14
5
15
7
16
9
18
1
19
3
20
5
21
7
22
9
73
85
49
61
25
37
0
1
13
AC Rutting (mm)
8
Month
Baseline
CGCM2A2x scenario
HadCM3B21 scenario
Future Research
•
Repeat MEPDG analysis using the latest AR4 climate
change scenarios, more sophisticated downscaling, and a
greater range of pavement structures and vehicle loads
•
Incorporate municipal distress data and a ravelling
(pothole) indicator into the analysis
•
Examine utility of monthly-seasonal scale forecasts
Further Reading
Andrey, J, B. Mills, D. Unrau, M. Christie and S. Michaels 2005. Toward a
National Assessment of the Travel Risks Associated with Inclement
Weather, ICLR Paper Series, Institute for Catastrophic Loss Reduction,
London, Ontario. 35 pp.
Baiz, S., S. Tighe, C.T. Haas, B. Mills, and M. Perchanok, 2008. Development of
frost and thaw depth predictors for decision making about variable load
restrictions, Transportation Research Record, 2053:1-8.
Mills, B., S.L. Tighe, J. Andrey, J.T. Smith, and K. Huen, 2009. Climate change
implications for flexible pavement design and performance in southern
Canada, Journal of Transportation Engineering, 135(10).
Thank you!