Deicer Usage on Concrete and Asphalt Pavements in Utah

Chase David Thomas, Brigham Young University

Abstract

The objectives of this research were to 1) compile winter maintenance data for the Utah Department of Transportation (UDOT) to directly compare concrete and asphalt pavements with regards to deicer usage and 2) determine if there is a statistical difference in deicer usage on concrete and asphalt pavements. To this end, three data sources were consulted for this research: Material Maintenance Quality Assurance (MMQA) database, UDOT road database, and Google Maps. The final compiled data set prepared for analysis in this research contained deicer quantities by deicer type, pavement surface areas by pavement material type, traffic, longitude, latitude, and elevation data. The deicer data evaluated in this analysis represented the total quantities of each deicer distributed during the 8-year period during which the MMQA database was used by UDOT.Several multiple linear regression analyses were performed to determine if concrete or asphalt pavements required different amounts of deicers, including salt, Redmond salt, brine, wetted salt, magnesium chloride, sand, pre-mix, and wetted pre-mix, during the winter seasons evaluated in this research. Because plow routes were not equal in total pavement area, a variable called “concrete proportion” was created. Similarly, traffic and deicer quantities were divided by total pavement area in lane miles to account for the variation in maintenance station sizes and to allow for direct comparison of the various maintenance stations. After the values of the independent variables were finalized, full and reduced models were created for the total amount of all deicers per lane mile and the amounts of each of the eight individual deicers per lane mile based on the statistical significance of the respective independent variables. A total of 18 regression models were completed for this research.From the results of the statistical analyses, concrete proportion was statistically significant in models for three of the dependent variables, including brine, wetted salt, and wetted pre-mix. However, neither the full nor the reduced regression model prepared for the sum of all deicers had concrete proportion as one of the significant variables. The absence of concrete proportion as an independent variable in these models shows that, on average, after correcting for differences in traffic volume and pavement area, deicer usage in Utah is not affected by pavement type. Therefore, except in areas where applications of brine, wetted salt, and wetted pre-mix are common, winter maintenance costs should not be a factor in the determination of pavement type.