An Exploratory Study on Precipitation Behaviour on a Porous Asphalt Highway

Abstract

Porous asphalt roads are renowned for their improvements of driver safety. Their high content of interconnected voids allows for an easy infiltration of water in the asphalt layer. This property limits the volume of standing water on the asphalt, reducing the risk of aquaplaning and increasing driver visibility by limiting the splash and spray effect. Other advantages include noise reduction and filtering of the pollutants of runoff water. As such, the use of porous asphalt has become more common across the world in the past 50 years. Currently about 90%-95% of Dutch highways is constructed using a toplayer of porous asphalt or Zeer Open Asfalt Beton (ZOAB) according to Dutch nomenclature.

While porous asphalt is known for its improved hydraulic performance, very little is known about the runoff response from its application on the highway. Rijkswaterstaat, the Dutch government agency responsible for managing the countries highway, always used a percentage of 20%-40% of precipitation
that will occur as runoff in internal documentation. However, these values need to be verified over a prolonged period of time using a continuous timeseries of recorded runoff. Secondly, in recent years the objective of Rijkswaterstaat has become to infiltrate as much runoff as possible in the verge next to the road in order to save costs in construction and maintenance of sewage as well as decrease the burden of the pollutants and peak discharge on the local environment. This has resulted in a need for a better understanding of the behaviour of precipitation on a porous asphalt highway under different conditions.

This study investigated the relation between the precipitation and the following runoff response on a 50-meter stretch of highway in the Netherlands. Furthermore, it explored the overall in-situ dynamics after a precipitation event. This was done through water balances and a multivariate analysis using a principal component analysis and a linear discriminant analysis. Finally, a linear reservoir model was applied to measured runoff responses.

Overall, this study has shown the potential of the linear reservoir model to accurately describe an in-situ runoff response. It has shown that this model is also applicable when there are multiple successive precipitation peaks during a single precipitation event. However, the total runoff volume during an event
can not yet be predicted. Furthermore, the water balances are not closing during the winter of 2024-2025. As such, a more thorough understanding of the evaporation dynamics on the highway is required at the relevant timescales. Lastly, this study has provided a statistical indication that traffic could play
a role in these dynamics as well.