Center for Environmental Excellence by AASHTO CENTER HOME  
skip navigation
CEE by AASHTO Home | Compendium Home | Online Compendium Help | Recent Updates | Inquiries | FAQs | State DOT Links
About Best Practices | Comment on Best Practices | Suggest A Best Practice | Volunteer to Vet Best Practices
Printer Friendly Version Print This Page    
« Back to Chapter 7 | Go to Chapter 9 »
Chapter 8 (Revised August 2013)
Winter Operations and Salt, Sand, and Chemical Management
8.4. Stewardship Practices for Reducing Salt, Sand and Chemical Usage

The true cost of winter maintenance products reflects not only the cost of the product and application of material, but also the impacts to infrastructure and the environment. In a survey conducted by Fay et al. (in review) survey respondents were asked to identify mitigation techniques used to reduce the impacts of chloride based deicers on the natural environment. The majority of mitigation techniques provided by survey respondents were proactive measures that can be used to reduce deicer use, and therefore put less deicer into the environment, instead of reactive mitigation technique used to clean-up or remove deicers from the environment once they have been applied. The following proactive measures identified from the survey can be used to reduce salt, sand and chemical usage in winter maintenance practices:

  • Anti-icing consumes much less material than deicing and can be applied prior to storm events to prevent a bond between the ice and pavement from forming, leading to easier plowing.
  • Pre-wetting solid salt and sand reduce bounce and scatter of the product, allows it to work faster, and provides a higher level of service than dry material.
  • RWIS provides information on real-time road conditions, including pavement temperature, and can be used to properly time treatments and determine which treatments should be used.
  • Brine production can support anti-icing and pre-wetting practices.
  • Staff training conducted annually and post storm can aid in preservation of best practices, and allows for sharing of knowledge.
  • Monitoring and keeping records of maintenance activities to improve operations. Consider monitoring chemical usage, application rates and conditions, etc. Monitoring in salt vulnerable areas may be supported by local conservation groups.
  • Pavement temperature can fluctuate significantly depending upon the time of day, degree of cloud cover, subsurface conditions (e.g., frost penetration, moisture presence, thermal retention properties, etc.) and type of pavement. Therefore ongoing monitoring of pavement temperatures is important for good decision-making.
  • Equipment calibration will keep application rates on target and prevent over application of material.
  • Deicing at the right time, in the right place with right amount (the 3 R's) is an effective practice. This can be done utilizing forecasts, available information (e.g., RWIS), and properly calibrated equipment that utilizes technology to apply product on the road surface so that it is not lost via bounce or scatter. All of these principles apply to anti-icing as well.
  • Material storage inside for solid products will prevent loss of product. Return unused product at the end instead heavy "end of beat" applications.

In sensitive areas consider altering application methods and or products, as well as plowing speed. Consider halting side-cast sweeping within 50ft of structures over water, and or placing barriers along water ways or utilizing drainage to move winter maintenance products away from watercourses. Alternative treatments can also be considered, such as only plowing, use of snow fencing, heated pavements, etc.

All of the proactive measures mentioned above are discussed in further detail in this report, as well as additional measures not listed above (e.g., snow fences, effective placement of material). Specific examples of how proactive measures can be used to reduce the amount of product used are provided below.

Iowa DOT Salt Model

The Iowa DOT developed a salt model to allocate salt to garages based on weather conditions and policy usage requirements (personal communication, A. Dunn, October 23, 2012). They needed a way for field personnel to understand how much salt they used. To accomplish this, Iowa DOT developed an algorithm that looked at the past five years of weather information at the garage level, specific parameters from the two closest RWIS stations to each garage, as well as garage lane-miles and LOS, precipitation type and start and end times, snow fall estimates, salt usage, salt ordered and salt received, and actual hours. They used this information to create a salt budget for each garage. Each garage is then allowed a certain amount salt for the season and if they need more they have to justify why to the central office. Any excess salt is kept for the next year.

Iowa DOT also developed an easy-to-use dash board, or user interface, that shows how much salt each garage has used versus what the algorithm predicted. This allows Iowa DOT to track how much salt they have at each garage, how much salt was purchased and how much was used. This newly developed tool is designed to assist field management staff in monitoring their salt usage. The dashboard allows for closer management of resources with outcomes and targets.

Kentucky Department of Highways Salt Matrix and Pre-set Spreader Application Rates

Kentucky Department of Highways developed a salt matrix to reduce application of salt while maintaining the same or better level of service (LOS) (personal communication, M. Williams, October 25, 2012). The developed salt matrices consider the pavement temperature and the heating or cooling trend, the road condition at the time of service, available maintenance strategies/actions, and provide recommended applications for liquid and solid products for both the initial treatment and subsequent treatments for four storm scenarios (light snow, moderate snow, heavy snow, and freezing rain). Reference sheets of the salt matrices are provided to county foremen.

Currently they are using a commercially available pre-wet solid product spreader system that is pre-set at four application rates (200, 400, 600, and 800 lbs/l-m). The drivers then use their judgment and the provided salt matrices to determine which application rate is needed for the given conditions.

Kentucky Department of Highways Participation in the MDSS Pooled Fund Project

The Kentucky Department of Highways is entering into its fourth year as a participant in the Maintenance Decisions Support System (MDSS) pooled fund study. Kentucky Department of Highways tested MDSS in the Lexington and Northern Kentucky areas and is currently expanding its use to include three sample routes within each of its 12 Highway districts. The MDSS tool provides a pavement level weather forecast for specific road segments and can help to identify the probability of bridge frost. In addition, the MDSS program will assist managers with the task of deciding when to activate personnel to respond to an approaching winter event.

To utilizing the MDSS forecasts and treatment recommendations, they will have AVL units on approximately 120 trucks that will transmit plow status, pavement and air temperature, and application rate information.

Washington State DOT and Maine DOT Salt Spreader Slurry Technology

Salt spreader slurry technology is essentially pre-wetting at a high ratio 70/30 percent (solid/liquid). This works out to be approximately 200 lbs/l-m of solid with 9 gal of liquid added. The slurry comes out with an oatmeal consistency. The solid salt grains are extremely saturated with this technique because the liquid is introduced at multiple locations. In the case of WSDOT the commercially available equipment adds liquid at multiple locations in the truck bed using an auger and at the spinner. WSDOT has observed the slurry going into action much quicker on the road, acting immediately and lasting longer on the road (up to 5 days under the right conditions) (personal communication, M. Mills, October 10, 2012).

Maine DOT utilized existing equipment and retrofitted their trucks with a similar system they made in-house (personal communication, B. Burn, October 10, 2012). With the in-house developed product Maine DOT has not been able to achieve the full extent of the commercially available products. Maine DOT published an evaluation of six retrofit designs and determined the approximate cost to be $7500 (2006 US dollars) per unit (Maine DOT, 2007). Maine DOT found an average savings of 7.8% per mile in product, with estimated cost savings of $1329 per unit when comparing the slurry system to control vehicles. Interviewed employees stated that they found the 70/30 slurry mix to out-perform typical pre-wetting methods (liquid application of 6, 8, or 10 gal per ton) while also minimizing bounce and scatter of product. Crews also stated that using a "heavier application on the first application followed by smaller applications", worked best and allowed for the product savings. Crews also stated the importance of getting out early in the storm.

8.4.1 Shifting to Anti-Icing
< back to top >

Over the last two decades, maintenance departments in North America have gradually made two transitions in their snow and ice control strategies. First is the transition from the use of abrasives to the use of more deicers (Staples et al., 2004). Currently, the United States applies approximately 20 million tons of salts each year for winter road maintenance. This is partially owing to the negative impact of abrasives to water quality and aquatic species, air quality, vegetation, and soil and the hidden cost of sanding (e.g., clean-up costs). In addition, the use of abrasives for winter maintenance has financial and environmental implications as it depletes valuable aggregate sources.

In more recent years, there is increased adoption of anti-icing strategy by many highway agencies in the U.S. in place of deicing (O'Keefe and Shi, 2005). According to a survey conducted by Fay et al. (in review), the majority of survey respondents (78%) indicated they have implemented anti-icing as a tool for reducing product application while maintaining or improving LOS. Anti-icing was also identified by the respondents as one of the ten most common practices that have been implemented or modified by their agencies. Anti-icing is defined as "the snow and ice control practice of preventing the formation or development of bonded snow and ice by timely applications of a chemical freezing-point depressant" (Ketcham, 1996). Generally, applications of liquid products mitigate frost and light black-ice events either before or after the weather incident, but are not considered practical for burning through thick snow and ice (Blackburn et al., 2004). The U.S. Federal Highway Administration (FHWA) provides guidance on the application rate of anti-icing materials, including liquids, solids and pre-wetted solids in its "Manual of Practice for an Effective Anti-icing Program" (Ketcham, 1996). This guidance covers four types of storm events: light snow, light snow with periods of moderate or heavy snow, moderate or heavy snow, and frost or black ice (note that guidance for the first three events only applies to solids and pre-wetted solids). For light snow, anti-icers should be applied at rates of 100 to 200 lbs/l-m, whereas light snow with periods of moderate to heavy snow application rates ranged from 100 to 225 lbs/l-m, depending on pavement surface and temperature conditions. Moderate to heavy snow application rates ranged from 100 to 250 lbs/l-m, whereas frost and black ice application rates ranged from 25 to 200 lbs/l-m, depending on pavement surface and temperature conditions. Depending on the pavement temperature, an anti-icer application rate of 65 to 400 lbs/l-m was suggested by Levelton Consultants (2007). Mitchell et al. (2006) conducted a survey of state DOTs regarding their application rates of anti-icers and found the typical application rates ranged between 20 and 65 g/l-m. Within Ohio, the anti-icer application rate ranged between 30 to 49 g/l-m.

A recent study sponsored by Clear Roads (Peterson et al., 2010) synthesized the current practices of during-storm direct liquid applications (DLA) and found DLA to be "a valuable asset for the winter maintenance toolbox". The DLA benefits listed by the synthesis include: reduced application rates, reduced loss of materials, faster post-storm cleanup, quick effect, further prevention of bonding, expanded toolbox, accurate low application rates, reduced corrosion effects, and leveraging proven benefits of liquids.

In practice, most agencies currently take a toolbox approach customized to their local snow and ice control needs as well as funding, staffing, and equipment constraints. Depending on the road weather scenarios, resources available and local rules of practice, departments of transportation (DOTs) use a combination of tools for winter road maintenance and engage in activities ranging from anti-icing, deicing, sanding (including pre-wetting), to mechanical removal (e.g., snowplowing), and snow fencing. Benefits of Shifting to Anti-icing

Anti-icing has proven to be a successful method of maintaining roadways during the winter season. Relative to deicing and sanding, anti-icing leads improved LOS, reduced need for products, and associated cost savings and safety and mobility benefits (Dye et al., 1996; Gilfillan, 2000; Kahl, 2002; Blackburn et al., 2004; Conger, 2005). A study performed in Kamloops, B.C. predicted that the majority of collisions caused by slush, snow, or ice could have been prevented by using a liquid anti-icer (Gilfillan, 1999). Russ et al. (2008) concluded that "if there is forecast winter weather likely to affect driving conditions, it is desired to have some form of salt on the road, preferably in the form of dried brine. If there is no or very little salt residue on the road, pretreatment is recommended, except under the following conditions: pretreatment would be rendered ineffective by weather conditions or (b) blowing snow may make pretreated roads dangerous". Rochelle (2010) evaluated various anti-icers in the laboratory and found that "the presence of chemical, regardless of chemical type, increased the friction of the pavement surface and reduced the shearing temperature as compared to non-chemically treated substrates for all pavement types, all application rates and all storm scenarios". Along with the effective use of deicers for snow and ice control, a successful anti-icing program can lead to substantial environmental benefits. Stewardship Practices to Minimize Anti-icing Materials Application

Anti-icing has emerged as the most commonly used proactive winter maintenance practice. Anti-icing has been shown to improve level of service (LOS); reduce the need for products, abrasives or plowing; and associated cost savings and safety and mobility benefits (Dye et al., 1996; Gilfillan, 2000; Kahl, 2002; Blackburn et al., 2004; Conger, 2005). The side benefit of anti-icing is that it reduces winter maintenance products impacts on the environment, infrastructure and vehicles. Generally speaking anti-icing is an effective practice at reducing application of winter maintenance products relative to other practices, such as deicing, because lower application rates can be used.

For anti-icing, decision processes can be unique to each maintenance division or department. Russ et al. (2007) developed a decision tree for liquid anti-icing for the Ohio DOT, which aimed to help maintenance supervisors consider a number of factors, including: current road and weather conditions, the availability of maintenance personnel and the best treatment strategy. Russ et al. (2008) concluded that "if there is forecast winter weather likely to affect driving conditions… (and) there is no or very little salt residue on the road, pretreatment is recommended, except under the following conditions: (a) pretreatment would be rendered ineffective by weather conditions or (b) blowing snow may make pretreated roads dangerous". In contrast, the Montana DOT uses just-in-time anti-icing. Under this strategy, rather than relying on a forecast, treatment of roadways does not begin until the maintenance agency identifies visual signs that a weather event is approaching such as moisture and temperatures drop, at which time crews will begin deploying anti-icing trucks. When used appropriately, this may avoid extra cost, reduce exposure of equipment and infrastructure to corrosion, reduce wastes, and maintain good public relations (Conger, 2005). This strategy, however, may pose a challenge for maintenance sheds in rural areas where the maintenance personnel are in charge of extended roadway segments.

Anti-icing applications do not perform satisfactorily below 20° F, as the liquid chemical can freeze if the pavement temperature is too low. In addition, anti-icing before a rain to sleet event can wash the chemical off the road (Blackburn et al., 2004). To avoid slippery pavement, agencies should not use anti-icing as a strategy and should particularly avoid the use of MgCl2, if there is blowing snow or the air temperature is above freezing and rising and the relative humidity is high (O'Keefe and Shi, 2005) due to the hygroscopic nature of MgCl2 . Also, caution must be used if anti-icing applications are made after extended dry spells, as they may lead to "a slippery emulsion to be formed by chemical and the contaminants that have built up on the roadway" (Leggett and Scouts, 2001).

The application rate of anti-icing products may significantly affect the skid resistance on asphalt pavement. In some cases, too high an application rate can lead to a reduction in the coefficient of friction of pavement. For instance, field tests were conducted in Michigan, which revealed that doubling the application rate of a commercially available liquid calcium chloride product from 30 gal/lane-mi (70 L/lane-km) to 60 gal/lane-mi led to a drop of average friction coefficient from 0.52 to 0.43. Note that the friction coefficient for dry and wet pavements averaged at 0.72 and 0.62 respectively (Leggett, 2001). The chemical residue remaining on the pavement can lead to slipperiness by drawing water to the road surface. There have been a few reported instances of this occurring across the Snowbelt region; however, not all cases are traced back to winter maintenance products. In some cases, slipperiness is the perception of the driver and sometimes it is caused by other contaminants on the roadway. In a few instances, it is a result of chemical application, but this only happens when dilution has occurred and refreeze is possible (Leggett, 1999). Applying liquids in streams spaced at least 8 inches apart will help minimize slipperiness in rare cases by not wetting the entire pavement surface at once (Blackburn et al., 2008).

Reductions in application of anti-icing materials can be observed from the these practices if the winter maintenance product can remain on the road longer and if less product is lost from plowing, vehicle splash off the road, or due to bounce and scatter.

8.4.2 Road Weather Information System (RWIS)
< back to top >

Weather information is central to the success of winter maintenance operations. Knowing when to apply treatment materials, what materials to apply, the correct amount that should be used and the locations where materials are needed all rely on timely and accurate weather information to guide decisions. Having access to such information can greatly assist in effective winter maintenance operations, but having the right data can be a challenge. For example, National Weather Service data, while widely available via the internet, is not typically highway-specific and does not lend itself to making real-time predictions. Other internet-based weather data sources present similar limitations and lack the capability of providing highway-specific information such as pavement temperature and condition status. However, Road Weather Information Systems (RWIS) are a tool that addresses these shortcomings and provides decision makers with real-time information specific to the roadway itself.

RWIS uses pavement and atmospheric sensors, Closed Circuit Television (CCTV) and communications to collect real-time roadway and weather data and provide it to agency personnel at garages, offices and in the field. RWIS stations are distributed throughout the roadway network, particularly at sites where weather and roadway conditions are of concern (e.g., mountain passes), to provide weather and roadway conditions data continuously through weather events.

To date, a good deal of research has been completed that has shown RWIS to be an effective and beneficial tool, providing material savings, improved level of service and reduced labor costs The following sections provide an overview of RWIS, the benefits it provides, and its location in the field, among other aspects.

What Are Road Weather Information Systems (RWIS)?

Road Weather Information Systems (RWIS) are networks of stations that collect various atmospheric (Environmental Sensing Stations, or ESS), pavement surface, sub-surface and video data (optional) to provide weather information for a specific site along a roadway. Aside from these sensors, RWIS is also comprised of communications, processing and display components. Collectively, RWIS stations provide information to managers that support decision-making with respect to deicing chemical applications, anti-icing strategies, materials and staff, and equipment scheduling and optimization.

A single RWIS site, which can consist of a number of different sensors both above and below ground, is a remote processing unit (RPU) station. Such a site typically consists of atmospheric condition sensors and CCTV cameras mounted to a tower, with pavement and subsurface sensors embedded in the pavement and below ground. A data processing unit and communications hardware are also present at the site, located in a weather-hardened enclosure (cabinet). Sensor data are processed on site and sent to a central processing unit (CPU) for storage, transmission to other workstations, or accessed directly when a need for records arises.

A key aspect of RWIS is its data processing and display capabilities, which provide maintenance staff with tools and information necessary to support improved decision making. As the technology has evolved, RWIS data has reached a point where it can be accessed from nearly any desktop workstation in an organization, we well as from remote terminals (including home computers) via the internet (provided the user has authorized access to the website or software). This provides the opportunity for decision making to be either centralized or decentralized, depending on the particular needs of an agency. Regardless, access to real time weather and condition information allows for better treatment decisions to be made, resulting in timelier winter maintenance activities to occur.

Benefits of RWIS

RWIS has been well documented through studies such as the NCHRP Synthesis 344: Winter Highway Operations and the FHWA Test and Evaluation Project 28: Anti-icing Technology, Field Evaluation Report (NCHRP, 2005; Ketcham et al., 1998). Consequently, the benefits it provides have been thoroughly identified and documented by a number of different studies and sources.

The Strategic Highway Research Program (SHRP) sponsored research in the early 1990s examined the potential benefits of improved weather information (Boselly et al., 1993a, 1993b). The study analyzed the potential cost-effectiveness of adopting improved weather information (including RWIS and tailored forecasting services), which used a simulation model based on data from three U.S. cities. It indicated that the use of RWIS technologies can improve the efficiency and effectiveness as well as reduce the costs of highway winter maintenance practices.

An NCHRP study documented the cost savings and benefits of RWIS technology, finding that the primary benefit offered was safer travel for motorists (Boselly, 2001). Other benefits included:

  • Improved Level of Service
    • Safer travel; improved driver information; help for local agencies, public service functions (through sharing data).
  • Cost Savings
    • Save agency money; reduce staff/equipment requirements; reduce use of salt (also environmental benefits); reduced patrolling.
  • Maintenance Response to Information.
    • Get the right equipment and materials at the right place at the right time; assist with crew scheduling; increased efficiency; implement better response strategies; helps make maintenance more effective; helps with decisions to "do nothing" when appropriate.
  • Indirect Benefits
    • Shorter travel times; reduced accident rates; reduced workplace absenteeism; less disruption of emergency services.
  • Other Benefits
    • Reduced wear on equipment and bridges; help in paving operation planning (other than winter maintenance); assistance in avalanche risk assessment.

In addition, the report states that based on the literature/agencies reviewed, a benefit-cost ratio of between 1.1 and 5.0 may be expected from RWIS systems, depending on the application (Boselly, 2001).

A study performed for Caltrans noted that past research on RWIS data indicated that when integrated with other data sources, such data resulted in cost savings through reduced crashes, lower insurance premiums, improved snow removal efforts, and reduced congestion (Laskey et al., 2006). Earlier work for Caltrans in 2002 showed that agency personnel perceived RWIS as useful, especially for snow/ice control and traveler information (Ballard et al., 2002; Ballard, 2004).

Strong and Shi performed a cost-benefit analysis of the use of customized weather service for winter maintenance activities. The research applied an artificial neural network model that included winter maintenance data from dozens of Utah DOT (UDOT) maintenance sheds for 2004-2005 to estimate the effectiveness of UDOT's weather information program (Strong and Shi, 2008). This program employed weather data from UDOT's 48 RWIS stations, as well as forecasts from four staff meteorologists. The benefit-cost ratio of the UDOT weather operations program was determined to be 11.0, based on material cost savings alone.

Ye et al. (2009) evaluated the effects of weather information on winter maintenance costs by using the methodologies of sensitivity analysis and artificial neural network. As part of this work, the researchers performed a cost-benefit analysis for the states of Iowa, Nevada, and Michigan to determine whether the use of accurate weather information provided by RWIS was effective in reducing winter maintenance costs (Ye et al., 2009). The direct benefits were found to outweigh the costs of using weather information (costs of the entire system), producing benefit-cost ratios of 1.8 (Iowa) and 3.2 (Nevada) [N].

Boon and Cluett discussed the use of RWIS in enabling proactive maintenance practices in Washington State (Boon and Cluett, 2002). The work identified a number of benefits associated with various winter maintenance practices, including:

  • Anti-icing - Lower labor and material costs; higher level of service; travel time savings, and improved mobility; improved safety; reduced equipment use hours and cost; reduced sand cleanup; less environmental impact; road surfaces returned to bare and wet more quickly.
  • Routine Patrolling - Reduced equipment use hours and cost; improved labor productivity.
  • Allocation of Resources - Increased labor productivity; reduced weekend and night shift work; improved employee satisfaction; reduced maintenance backlog; more timely road maintenance; reduced labor pay hours; overall higher level of service; more effective labor assignments.
  • Traveler information - Better prepared drivers; safer travel behavior; reduced travel during poor conditions; fewer crashes, injuries, fatalities and property damage;
  • Increased customer satisfaction and political support; improved mobility; safer, more reliable access.

Furthermore, the authors briefly discuss the results of previous studies on RWIS that had developed benefit-cost ratios. These ratios ranged from 1.4 to 5.0.

McKeever et al. (1998) developed a life cycle cost-benefit model for RWIS using a site in Abilene, Texas. Direct costs of RWIS used as inputs included initial capital purchase. Direct savings (benefits) included reduced winter maintenance costs (patrol, labor, equipment and materials), while indirect savings included reduced liability risk, accidents, pollution, and travel costs. It was found through the model that a savings of approximately $923,000 could be accrued over a 50 year period.

Strong and Fay (2007) found that Alaska's benefits from RWIS usage included: reduced staff overtime, less misdirected staff time, fewer wasted materials and equipment, and improved roadway level of service (LOS) (Strong and Fay, 2007). However, the RWIS network appeared to be achieving only some, but not all, of these benefits statewide. The researchers concluded that the full benefits of RWIS would be achievable only if there was a concerted effort toward proactive winter maintenance, namely anti-icing.

Ye and Strong discussed the potential benefits of weather information in winter maintenance. Specifically, the researchers identified secondary applications of RWIS information (enhanced signal timing, weather-responsive operations, etc.), including the costs and benefits associated with them (Ye and Strong, 2009). These included:

  • Enhanced traffic signal timing
    • Reduced vehicle delay; less fuel consumption.
  • Weather-responsive operations
    • Improved accessibility of information; ability to coordinate and pool resources; improved cost effectiveness; improved public safety; more efficient evaluation activities; more timely and accurate information provided to the traveling public; better prepared TMC operators to address adverse weather.
  • Dynamic warning systems
    • Reduced vehicle crashes; reduced crash severity; more comfortable driving.
  • Anti-icing spraying system
    • Reduced vehicle crashes; reduced crash severity.
  • Traveler information
    • Safer and more comfortable driving; better trip planning for travelers.
  • Intelligent Transportation System (ITS) Nodes
    • Provide power and communications system to other technologies.
  • Network expansion
    • Provide road weather information to interested agencies/entities; share data with other states; help calibrate weather models.

Experience with RWIS in Iowa indicated that three primary benefits could be achieved. These included improved maintenance of ice free roadways, reduced labor costs, and reduced chemical usage (FHWA, 2009a). By providing accurate information on current conditions, faster and more efficient responses to storm events could be made.

The Transportation Association of Canada highlighted a number of the benefits of RWIS, including the following:

  • Provides an understanding of pavement temperatures and trends that can improve the accuracy of decision making.
  • Pavement and subsurface sensors can provide data for developing trends and aid in forecasting.
  • Provides temperature monitors to aid in determining treatments based on pavement status (wet/dry), freeze point of the solution on the road, presence of and concentration (for some deicers), as well as subsurface temperature.
  • Data can aid in helping staff make better decisions regarding snow and ice control.
  • Salt use optimization is achieved by more accurate deployment of equipment and application (TAC, 2003).

In addition to the literature discussed in this section, the current (2012) update of this chapter has conducted a survey of winter maintenance professionals. The following are the survey responses from this survey synthesis that pertain to weather forecasting and RWIS:

  • Currently have 18 RWIS sites with sensors that give the chemical factor and thus help us to avoid the over application of salt when we do have a winter storm event (South Carolina DOT).
  • Currently have 49 fixed road weather stations and more than 140 mobile road weather stations are made available to decision makers. Training sessions were held and others will be held again over the coming years so that users can use the devices to their full potential (Quebec, Canada).
  • Have reduced chloride use by storm intensity/duration forecasting (Utah DOT).
  • Road weather stations were implemented first as tools to help decision makers to better plan operations. If the operations are better planned, it is possible to reduce the amount of road salts spread by using other technologies, other materials or simply to spread the right amount at the proper time (Quebec, Canada).

RWIS Selection, Sitting, Use, and Maintenance, Connection to Snow and Ice Control Materials and Methods and Use of Friction Indicators to Minimize Chemical Usage

The location of RWIS stations is crucial in ensuring their successful use in winter maintenance operations (FHWA, 2013). Strategically placed RWIS stations provide forecasts that are 90% to 95% accurate, a rate which improves with the addition of more stations and better technology (FHWA, 2009b). Specific information regarding RWIS, its selection, procurement, siting, use, maintenance, and calibration can be obtained in the two-volume SHRP report Road Weather Information Systems Volume 1: Research Report and Road Weather Information Systems Volume 2: Implementation Guide (Boselly et al., 1993a, 1993b). These documents provide background information on RWIS sensors and equipment, sensor siting, communications needs, equipment siting, and steps to implementing each station. The reader should be aware however, that many sensor technologies and communications mechanism have evolved in the years since these reports were written (1993), although much of the information provided, particularly with respect to locating RWIS sites, remains practical and valid.

Fleege et al. (2006) provide best practices and practical guidelines to ensure the reliable operation of RWIS sensors in the field in NCHRP Web-only Document 87, Test Methods for Evaluating Field Performance of RWIS Sensors. This information will assist maintenance staff in maintaining RWIS equipment to ensure accurate data is measured and reported by the system.

Once RWIS sites have been deployed, the information they provide must be put to use. The National Cooperative Highway Research Program (NCHRP) Report 526, Guidelines for Snow and Ice Control Materials and Methods, provides guidance related to the selection of appropriate treatment strategies and tactics for specific winter storm conditions (Blackburn et al., 2004). RWIS data directly compliments this effort by providing the timelier storm and pavement condition data needed to drive the different strategies and tactics outlined in the report. When combined with the information outlined in NCHRP Report 577, Guidelines for the Selection of Snow and Ice Control Materials to Mitigate Environmental Impacts, RWIS data can help agencies develop treatment strategies that minimize material usage, maximize efficiency, produce better pavement conditions and minimize impacts on the environment (Levelton Consultants, 2007).

Improved friction is a central aspect of winter maintenance, and RWIS data can assist in improving pavement friction conditions by guiding treatment strategies. NCHRP Web Document 53, Feasibility of Using Friction Indicators to Improve Winter Maintenance Operations and Mobility, provides insights on the use of friction data with RWIS data to aid in decision making during storms (Al-Qadi, 2002).

Training in the use of RWIS and its applications to maintenance is also a critical aspect that must be considered. Mitchell et al. (2006) as part of a larger effort examining snow and ice removal operations in Ohio, found that regular training and refresher courses for maintenance staff, particularly at the county level, were essential in the use of RWIS. Training was viewed as a way to obtain more and effective utilization of RWIS information and for counties to share best practices.

8.4.3 Pavement Sensors and Thermal Mapping
< back to top >

Pavement sensors monitor surface conditions and can serve as a data input for warning systems (e.g., ice warning). They can provide information for use in producing forecasts of pavement conditions based on temperature measurements. Pavement sensors can also measure chemical concentrations, providing information that can help determine when additional treatments should be made. Many available sensors also can measure the freezing point of the solution present on the sensor, providing insight into when diluted products may be likely to refreeze. Collectively, the information generated by pavement sensors can be used in monitoring conditions and planning and carrying out treatment strategies.

Sensors intended to measure roadway weather conditions can be of two general types, invasive and noninvasive. The invasive sensors, more commonly known as in-pavement sensors, are installed in the pavement level with the road surface and can use many different sensing technologies in determining roadway conditions. Noninvasive sensors are typically installed on the roadside or somewhere over the road surface and use non-contact means of monitoring weather surface conditions.

In-pavement sensors use the dielectric characteristics of the condition present on the surface of the sensors to determine if the road is dry, damp, wet, or icy and to detect the presence of deicers (Schedler, 2009). The conductivity of the liquid present on a sensor can be used to determine the freezing point temperature of that liquid; this type of sensor is known as a passive sensor. An active sensor is one that actively heats or cools itself to directly measure the freezing point of a liquid present on the sensor (Jonsson, 2010). Some in-pavement sensors also utilize microwave radar to measure the depth of water present on the sensor (Schedler, 2009).

Non-invasive sensors can be mounted apart from the road surface, either above the roadway or at the roadside, or handheld, and use spectroscopic methods, thermal radiation, and infrared radar methods to determine surface weather conditions from a distance. Infrared radiation and thermal radiation are measured to determine the temperature of the road surface. Spectroscopic measures are used to determine roadway surface conditions (dry, damp, wet, ice, etc.) remotely. Friction metrics can be calculated from the temperature and surface conditions (Jonsson, 2010). Inexpensive handheld salinity measuring devices are also available.

In addition to pavement sensors, thermal mapping, or thermography, can also be used to determine thermal profiles of road surfaces. This approach can be used to infer pavement temperatures between sensor locations, as well as for forecasting temperatures at given points. Thermal mapping can be accomplished by using inexpensive hand-held radiometers or vehicle-mounted sensors specifically designed to measure pavement temperature. Thermal measurements are typically made in the early morning to ensure that there is minimal change of temperature during the measurement process. Data from thermal mapping has been used to forecast pavement temperatures in locations without RWIS stations, to site RPU stations, and to aid in the development of treatment strategies.

8.4.4 Infrared Thermometers (IRTs)
< back to top >

Pavement temperature has a greater influence on deicer performance than air temperature. Thus, pavement temperature should be considered when deciding salt application rates. Air temperature is readily available, and has some influence on pavement temperature. Pavement temperature models exist, but are difficult to develop and calibrate and are often only regionally applicable (Sato et al., 2004; Adams et al., 2004; Fu et al., 2008). Infrared thermometers are an easy, quick and generally accurate tool for measuring pavement temperature. Handheld and mobile (vehicle mounted) versions are available. Controlled laboratory and field tests of mobile pavement temperature sensors found the average error was 1.4°F, they were more accurate on concrete pavements and the presence of salt did not affect the temperature readings (SRF Consulting, 2005).

8.4.5 Road Surface Traction/Friction Measurements
< back to top >

Information about the real-time friction level of the road surface can be used in decisions to plow and/or apply materials and, if used properly, can prevent over application. However, friction measurements are not often routinely collected on streets and highways during winter storms because equipment is either expensive (such as friction trailers) and/or may have limitations for operation in traffic (some equipment requires heavy braking or reduced speeds). Friction trailers provide physical friction measurements by an instrumented "fifth wheel." Recent research has indicated some potentially promising strategies for determining friction with standard passenger vehicles equipped with 1) traction control systems, or 2) GPS and motion sensors, to back-calculate friction based on vehicular motion (Al-Qadi et al., 2002; Takahashi et al., 2004; Nakatsuji et al., 2007). However, these remain a topic of research and widespread implementation is still a ways off. There are several non-contact optical sensors that measure the amount of ice, snow and water on the pavement using spectroscopy and provide an estimate of the friction. There are a variety of sensors available, some of which can be mounted on vehicles to collect friction during patrols or mounted at RWIS sites. See section for a case study on the use of non-contact optical friction measurements to assess winter maintenance product performance.

Friction measurements have been used extensively at airports and a summary of these techniques can be found in (Fay et al., 2010).

8.4.6 Residual Chemical Measurement
< back to top >

Salinity sensors can be used to monitor the residual salt concentrations on the road surface, helping maintenance managers make educated decisions related to chemical reapplication (Ye et al., 2012). Potential advantages of using on-vehicle salinity sensors include monitoring the concentration of the solution on a road surface along entire stretches of roadways, which allows for more accurate application rates, and integrating measurements from salinity sensors to automatic spreader controls to apply the right amount of product in the right place. There are two general methods for on-vehicle salinity sensors, measuring the conductivity of collected tire splash (Garrick et al., 2002a, 2002b) and the refractive index of an aqueous solution atop of the road surface (Iwata et al., 2004; Mexico Company, 2010), respectively.

In addition, Atkins highways and transportation (2007) identified two non-contact methods of measuring salinity on pavement: laser-induced fluorescence (LIF) (Hammond et al., 2007) and laser-induced breakdown spectroscopy (LIBS) (Nail and Kumar, 2004). Both LIF and LIBS have been successfully used in architectural, defense and medical applications. They could offer the potential for development into viable traffic-speed residual salt measurement techniques. However, some important issues need to be addressed before these techniques are used in monitoring residual salt concentrations on pavement. These include operator and road user safety related to exposure of the laser discharge (in particular with LIBS), and the need for the device to measure salinity on both wet and dry surfaces.

According to Mitchell et al. (2004), "an effective anti-icing program requires prediction and estimation of the amount, type, and timing of chemicals needed for the expected precipitation event while compensating for time and traffic decay of the chemical on the highway surface". As such, they investigated the decay of deicer residuals on several pavement types in Ohio and used field and laboratory data to predict the amount of residual chemical on three types of pavement as a function of time and traffic and to predict the required application rate for the given precipitation conditions. Lysbakken and Norem (2011) and Blomqvist et al. (2011) developed models to account for processes and factors that define the change of salt quantity on road surfaces after salt application.

A recent case study investigated the longevity of inhibitors and the performance of corrosion-inhibited anti-icing products after pavement application during winter storms (Shi et al., 2012b). The field operational tests included the daily sampling of anti-icer residuals on the pavement for seven days after anti-icer application for a black ice event, a man-made snow event, and a natural snow event, respectively. Subsequently, multiple analytical methods were used to examine the properties of pavement-collected samples in the laboratory. It was found that more than 62% of the inhibitor in the CaCl2-based anti-icer and more than 20% of the chlorides (especially for the MgCl2 - and NaCl-based anti-icers) remain on the pavement four days after the application of liquid anti-icers to treat black ice. The longevity of chlorides and inhibitors on the pavement after anti-icer application can vary greatly depending on the pavement temperature, the amount of precipitation, etc.

8.4.7 Nowcasting
< back to top >

Nowcasting refers to the use of real-time data for short-term forecasting. It relies on the rapid transmittal of data from RWIS installations, radar, patrols, and any other information source for making a judgment of the probable weather and pavement condition/temperature over the next hour or two. Nowcasting is a tool that can be used for decision making, such as when to call in personnel, mobilization, and timing. Needs may vary among sites, therefore the frequency of weather information updating required for a nowcast will also vary with the site. Nowcasts can be provided by a weather service or performed by the maintenance manager. Specially trained maintenance managers in some highway agencies already perform this duty using the necessary information available from a variety of sources. (Ketcham et al., 1996)

8.4.8 Road Weather Management Decision Support
< back to top > Clarus Initiative

The Clarus Initiative is a joint effort of the US DOT ITS Joint Program Office and the FHWA Road Weather Management Program. The goal of the Clarus Initiative is to develop and demonstrate an integrated surface transportation weather observation data management system, and to create a Nationwide Surface Transportation Weather Observing and Forecasting System (FHWA, 2011). The system is designed to enable public agencies to more accurately assess weather and pavement conditions, and their impacts on operations. As of 2009, 46 state and local US, and Canadian provincial agencies are participating the program and contributing data.

Through the Road Weather Management Program, FHWA has sponsored research investigating the use of vehicles as meteorological sensor platforms (FHWA, 2011). Work by the National Center for Atmospheric Research (NCAR) assessed the feasibility of using this data to improve surface transportation weather observations, predictions and road condition hazard analysis and predictions. Research in this area is ongoing. Another example of using vehicle data to support road weather management is the Connected Vehicle (CV) program funded through the Research and Innovative Technology Administration (RITA). Testing conducted in 2009 found that vehicles accurately measure temperature, but are not as accurate at measuring barometric pressure (Chapman et al., 2010). At this time the future direction of this program in unknown. MDSS

Maintenance Decision Support System (MDSS) is a decision support tool that integrates relevant road weather forecasts, coded maintenance rules of practice, and maintenance resource data to provide winter maintenance managers with recommended road treatment strategies (FHWA, 2011). MDSS is an integrated software application that provides users with real-time road treatment guidance for each maintenance route. Additional information on the development of MDSS and its evolution can be found in Deployment of Maintenance Decision Support Systems for Winter Operations (Pisano et al., 2006). MDSS has developed into an automated decision support tool for winter road maintenance managers, but also serves to provide forecasts, weather predictions, reports on observed weather and road conditions, serves as a training tool, and is a management support system that can be used year round (McClellan et al., 2009).

Over 20 states have implemented MDSS to some extent. Indiana DOT documented the progression of limited use of MDSS to statewide implementation in the 2008 - 2009 winter season (McClellan et al., 2009). With the statewide implementation INDOT paid careful attention to understanding the changes needed in the management strategy of MDSS and their winter operations, as well as how this change would impact INDOT. They found demonstration of MDSS to all levels of INDOT personnel, allowed for easier understanding of the value of MDSS as a business practice and to the financial bottom line. Training for employee's occurred prior the winter season and a detailed management plan was developed for all aspects of implementation. Throughout the season INDOT proactively resolved issues within INDOT and with the forecast provider. With the well-orchestrated implementation in the first year of statewide implementation, INDOT was able to use 228,470 less tons of salt, for a savings of $12,108,910 and 58,274 less overtime compensation hours, for a savings of $1,359,591. When normalized for varying winter conditions the total combined saving for salt use and overtime compensation hours was $10,957,672.

Additional observations were, MDSS proved to be a powerful management tool, considerable time was spent refining the forecasts with the vendor, and it took time for the management to become comfortable using MDSS but eventually many used it as a planning tool instead of reacting to winter conditions (McClellan et al., 2009). Continued training and exposure to MDSS was needed to ensure the success of the program.

A cost-benefit analysis of MDSS use over two winter seasons 2007 - 2009 in the City and County of Denver, Colorado found savings that exceeded the costs of the system while maintaining the LOS (Cluett and Gopalakrishna, 2009). The saving from the use of MDSS were realized from more effective tactical crew deployment decisions, rather than using the recommended treatments mode of MDSS.

A cost-benefit analysis of MDSS implemented in New Hampshire, Minnesota, and Colorado identified benefits as reduced material use, and improved safety and mobility and identified significant cost savings (Ye et al., 2009c). The benefits were found to outweigh the costs associated with the technology in all three states, with benefit-cost ratios ranging from 1.33 to 8.67 due to varying conditions and uses of resources. Identified intangible benefits and needs were:

  • MDSS provides a quantitative evaluation of performance measures.
  • MDSS can be used as training tool.
  • Outcomes associated with changes in rules of practice can be evaluated through MDSS.
  • Use of MDSS requires a quality weather forecast.
  • The quality of the recommendations from MDSS is dependent on properly sited, maintained and reliable ESS.
  • Less use of maintenance vehicles.
  • Consistent/seamless treatment of roads among maintenance sheds.
8.4.9 Weather Forecasts and Information Services
< back to top >

In order to facilitate best practices in winter maintenance operations, it is crucial to obtain and utilize accurate weather information. Otherwise, the consequences could be: excessive use of deicers and materials, failure to respond in a timely manner to a storm event (resulting in greater crash risk and user delay), unplanned use of overtime staffing, etc. Improvements in weather information can help in all stages of winter storm response, including pre-, during and post-storm. When considering the choice between spatially or temporally improved forecasts, Fu et al. (2009) found that improved spatial resolution of forecast data will provide greater expected benefit to service levels.

Boselly et al. (1993) analyzed the potential cost-effectiveness of adopting improved weather information, including road weather information systems (RWIS) and tailored forecasting services. The simulation results indicated that the use of RWIS technologies can improve the efficiency and effectiveness as well as reduce the costs of highway winter maintenance practices. A 2007 survey of winter maintenance personnel (Ye et al., 2009a) found that free weather information sources, private-sector weather providers, and RWIS were the most widely used weather information sources. Air temperature, wind, and the type and amount of precipitation were primary parameters of current and forecast weather conditions, whereas road weather elements (e.g., pavement temperature, bridge temperature, and pavement conditions) were also widely used in winter maintenance. Three case studies at the State level collectively showed that winter maintenance costs decreased as the use of weather information increased or its accuracy improved (Ye et al., 2009a). A survey conducted for this project found the majority of the respondents (75%, n=24) identified "utilizing additional information sources (road weather forecasts, RWIS, maintenance decision support systems - MDSS, pavement and/or vehicle sensors, etc.)" as an effort their agency made in the last 10 years to reduce the amount of winter maintenance products applied during winter maintenance operations while maintaining the same or better LOS.

Anti-icing is more sensitive to weather conditions than other winter maintenance practices, since it is a proactive practice that is sensitive to pavement temperature, dilution, and other factors (Blackburn et al., 2004). Near-real-time weather and road condition information and accurate weather forecasts are critical to the success of an anti-icing program; as such information will guide the timing and amount of product needed for pro-active operations (Shi et al., 2007; Ye et al., 2009a, 2009b). To this end, reliable, micro-scale models for the forecasting of localized weather (or road surface temperature) should be established and a network of weather stations should be in place to enable, validate and refine the models. For instance, the Utah DOT features a Weather Operations program that provides reliable, site-specific local weather forecasts for the highway maintenance staff, which promotes the adoption of anti-icing practices by winter maintenance managers and crews (Shi et al., 2007).

Russ et al. (2008) identified real-time Doppler as a helpful tool for providing weather information in a timely manner. Mesons, also referred to as mesonets, regional networks of weather information that integrate observational data from a variety of sources, have also been identified as a useful tool (Strong et al., 2010). Mesons aim to provide a more comprehensive and accurate picture of current weather conditions. These data management systems are expected to maximize availability and utility of road weather observations and facilitate more accurate, route-specific forecasting of road weather conditions. Weather service customized for winter maintenance operations can be extremely valuable. Shi et al. (2007) examined the labor and materials cost for winter maintenance in the 2004-05 season for 77 Utah DOT sheds and established an artificial neural network model to treat the shed winter maintenance cost as a function of UDOT weather service usage, evaluation of UDOT weather service, level-of-maintenance, seasonal vehicle-miles traveled, anti-icing level, and winter severity index. The model estimated the value and additional saving potential of the UDOT customized weather service to be 11 to 25% and 4 to 10% of the UDOT labor and materials cost for winter maintenance, respectively. It was also estimated that the risk of using the worst weather service providers to be 58 to 131% of the UDOT labor and materials cost for winter maintenance.

8.4.10 Traffic Information
< back to top >

PTraffic is a critical consideration in winter maintenance, as vehicles have a direct impact on the condition of the roadway during and after a storm event. From a winter maintenance perspective, roadway surfaces are affected by vehicles in a number of ways, both negatively and positively. Vehicle tires can compact or abrade snow, slush and ice. Vehicles can also displace/disperse snow, ice, slush and treatment materials, all of which impact the effectiveness of winter maintenance operations. All of these can be considered negative impacts on winter maintenance. The positive impacts of vehicles include engine heat and exhaust providing warmth to melt snow, slush or ice, aiding in restoring bare pavement.

All of these impacts must be considered when planning and conducting all winter maintenance operations, but particularly anti-icing. Consequently, traffic information should be taken into consideration during the planning of upcoming winter maintenance operations. Specifically, traffic rates and associated variations over a 24 hour period (particularly for a day of week similar to a storm event that is being planned for) should be consulted to understand when and where traffic may be at different times and what treatments might be most beneficial to address those traffic volumes. Traffic information is typically electronically available in the form of continuous counts collected by Automatic Traffic Recorders (ATRs) that are strategically located along different highways by state (and occasionally local) highway agencies.

8.4.11 Patrols
< back to top >

There is no substitute for visual observation of weather conditions and conditions on and near the pavement surface by trained maintenance personnel through the use of patrols. The use of real-time cameras and sharing of real-time information with AVL technology and RWIS can aid maintenance staff in the task of performing patrols by providing information before going out on the road, automatically documenting conditions and information, with the potential to reduce the need for some patrols which can lead to - saving time, money, wear and tear on vehicles, and keeping maintenance personnel safe by keeping them off the roads in inclement weather.

8.4.12 Drift Control/Snow fences
< back to top >

A properly designed and placed snow fence can be a cost-effective tool for snow and ice control for highway segments where the abundance of wind leads to blowing and drifting snow on the winter roadways. For such locations, there is a considerable risk of closing the roadways or requiring excessive plowing and chemical usage. The use of primary drift control technique (snow fence) to minimize the amount of snow blowing onto the roadway will provide a number of benefits to the public and landowners. Some of the benefits include:

  • reducing blowing/drifting snow on roadways
  • storing snow at low cost
  • creating safer travel condition
  • reducing the need for snow and ice control products (TAC, 2003).

There are two types of snow fence to manage drifting snow, permanent and temporary. Tabler (1991) found a single row of taller snow fence (at least 8 ft tall) traps more snow, more effectively improves driver visibility, costs less and requires less land than multiple shorter fences.
Snow fence has been proven to be a low-cost mitigation method to prevent blowing snow related accidents. It also helps by reducing maintenance costs and wear-and-tear on the winter maintenance equipment (Wyoming DOT, 2009)

Costs for snow fence can be as little as $1.39 per ft2 of fence (Tabler, 2005). Data from a Wyoming study shows that "storing snow with snow fences costs three cents a ton over the 25-year life of the fence, (relative) to three dollars a ton for moving it" (Tabler, 1991). In the 1970s, the Wyoming DOT reduced snow and ice removal costs by more than one-third on a 45-mile stretch of I-80 where fences were installed. The fences had been remarkably effective in preventing drift formation over the 20 years since installation (Tabler, 1991). Living Snow Fences

Living snow fences are barriers made of trees, shrubs and native grasses along roadsides. When properly designed and placed, living snow fences trap snow as it blows across the ground, piling it up before it reaches roads. Living snow fences prevent big snowdrifts, maintain mobility and visibility, allow for less product and fuel used to keep roads clear, provide habitat for wildlife, control soil erosion, improve water quality, reduce spring-time flooding, and sequester carbon (Leaner and Greener, 2012). "A living snow fence offers benefits worth $17 for every $1 spent" or a cost:benefit ratio of 17:1 (Leaner and Greener, 2012). Living snow fences can also be constructed of corn stalks, or other agricultural products. Agreements have been made with farmers to not remove corn stalks near roads so they may act as annual snow fences (Wyatt et al., 2012).

< back to top >
Continue to Section 8.5 »
Table of Contents
Chapter 8
Winter Operations and Salt, Sand, and Chemical Management
8.0 Introduction
8.1 Selecting Snow and Ice Control Materials to Mitigate Environmental Impacts
8.2 Reducing Sand Usage and Managing Traction Materials
8.3 Strategic Planning for Reduced Salt Usage
8.4 Stewardship Practices for Reducing Salt, Sand and Chemical Usage
8.5 Precision Application to Manage and Reduce Chemical Applications
8.6 Monitoring and Recordkeeping
8.7 Winter Operations Facilities Management
8.8 Training for Salt Management and Winter Maintenance Operations
  Appendix A - Acronyms
Lists: Examples | Tables | Figures
Website Problems Report content errors and/or website problems
PDF Document Download Adobe Acrobat Reader