Experimental Studies on Geocells and Mat Systems for Stabilization of Unpaved Shoulders and Temporary Roads

Abstract

Geosynthetics have been used to improve the performance of geomaterials, especially when weak soil exists in roadway applications. In this study, two types of geosynthetic materials, geocell and a mat system, were studied for their applications for unpaved roads and shoulders. The study of geocell was focused on its application for unpaved shoulders. The ability of geocell to improve different geomaterials over intermediate strength subgrade and its possible effect on vegetation were investigated. The study of the mat system was focused on investigating the performance of the mat system over soft and intermediate subgrade with different strengths under cyclic loading to simulate temporary roadway conditions. In the study of geocell for the application for unpaved shoulders, six large scale plate loading tests were conducted on a single type of geocell on target 5% CBR subgrade to investigate the benefits of geocell reinforcement on different base course and topsoil combinations. Different base course and topsoil combinations were investigated including: 200-mm thick unreinforced aggregate, 200-mm thick soil-aggregate mixture (50% aggregate and 50% top soil) with and without geocell reinforcement, 200-mm thick geocell-reinforced topsoil, 50-mm thick aggregate over 150-mm soil-aggregate mixture (50% aggregate and 50% top soil), and 50-mm thick top soil over 150-mm thick geocell-reinforced soil-aggregate mixture (50% aggregate and 50% top soil). Earth pressure cells were install at the interface between subgrade and base course to monitor the load distribution. The cyclic plate loading tests showed that geocell effectively reduced the permanent deformation and the geocell-reinforced soil-aggregate mixture slightly outperformed the unreinforced aggregate at the same thickness. The plate loading tests also suggested the topsoil cover resulted in large permanent deformations. A one-year long outdoor field vegetation test was conducted on base courses with different combinations of aggregate and topsoil including: 200-mm thick unreinforced topsoil, 200-mm thick soil-aggregate mixture (50% aggregate and 50% topsoil), 50-mm thick aggregate over 150-mm soil-aggregate mixture (50% aggregate and 50% topsoil), and 50-mm thick topsoil over 150-mm reinforced soil-aggregate mixture (50% aggregate and 50% topsoil) to investigate the possible effect of geocell on shoulder vegetation established mainly by tall fescue grass and perennial ryegrass. One control (unreinforced) section and one geocell-reinforced section were prepared for each base course combination with a surface area of 1.5 m by 1.5 m. During the one-year test period, soil moisture temperature and volumetric moisture content were monitored. Weather data, such as precipitation and air temperature, were obtain from the nearby weather station at the Lawrence airport. Vegetation growth was evaluated by grass leaf blade length, root length, and grass density. Vegetation biomass was obtained at the end of the test. The test results showed no definite evidence of geocell influencing the vegetation in unpaved shoulders. In the study of the mat system, six large-scale cyclic plate loading tests were conducted on a single type of polyethylene mat system with anchorage to study its performance over soft and intermediate subgrade with the CBR ranging from 1% to 4%. For the comparison purposes, test sections with and without the mat system were prepared and evaluated. For the test section with 1% CBR subgrade, an aggregate base course was used for the test section without the mat system to enable the cyclic plate loading test. The size of the mat system under the investigation was 1.92 m by 1.92 m. A cyclic load at the magnitude of 40 kN was applied through a 300 mm diameter loading plate with a thin rubber pad to simulate a vehicle tire. Earth pressure cells were installed at the interface between the mat system and the subgrade to monitor the load distribution. Loading plate displacements were measured by the displacement transducer inside the actuator. Test results concluded that the mat system was more effective over the intermediate subgrade than the soft subgrade and when large permanent deformations were allowed.