Soil compaction and the effect on infiltration in urban green environments : A study based on field measurements and HYDRUS 1D modelling

Abstract: The consequences of recent flooding and extreme rain events have highlighted the importance of proper urban planning and preventative measures for storm water management. As cities become more urbanized the significance of permeable surfaces such as parks and other urban green spaces increases which infiltrate the water into the ground. Agricultural research has for many years emphasized the effect of compaction on soil parameters and how, not only the crop yield reduces but also how the infiltration decreases. This thesis aims to study how the infiltration rate, bulk density and soil resistance changes with compaction through field experiments where a vehicle is let to roll over an urban green area. The thesis will also simulate rainfall over five theoretical soils that can be found in urban environments exposed to compaction to determine what significance compaction has on surface runoff. The modelling software HYDRUS-1D will be used so simulate rain fall events on the different soils. The rain events simulated will be based on the five hyetographs that best represent Sweden’s rain events, based on historical data. A CDS rain will be simulated as well. They will be simulated for a 2, 10 and 100 year return period. A literature study will also be conducted to determine how relevant freeze-thaw cycles are to the soil parameters. It is since previously known that freeze-thaw cycles can improve aggregate stability, increase soil particle fragmentation which can lead to less soil penetration resistance and even partially return the soil conditions to those prior to compaction, but the process does not extend to layers beyond 40 cm. The field experiment results showed a clear decrease in infiltration rate with increasing number of vehicle passes. There was no clear correlation between bulk density and the number of vehicle passes. This result is attributed to the relatively light weight of the vehicle used as well as the heterogeneity of the soil. The cone penetration measurements showed an increasing resistance with increasing number of vehicle passes for only one of the three measured sites, with the most resistance being measured in a pathway on the green area. The insignificant results of one of the two other sites are attributed to wet weather conditions and unknown underlying material. The HYDRUS 1D simulations showed that a higher sand content mitigates the effects of soil compaction and leads to less runoff. The soil classified as sand (93% sand) had no runoff, the loamy sand (80% sand) had mild runoff. When comparing a sandy loam (60% sand) and a clay soil it is concluded that the sandy loam is more sensitive to soil compaction as more compaction leads to more runoff compared to the non-compacted scenario. The clay soil has little variation between the compaction scenarios but has generally more surface runoff in total. Soil texture therefor affects the surface runoff more than soil compaction. Most amount of runoff was generated by the two hyetographs which had a late peak intensity, most likely due to the soil already being saturated when the peak occurs. The runoff also increases with the return period of the rain event for both the hyetographs and the CDS rain.