Comparing lab-obtained to simulation-obtained saturated hydraulic conductivity of Leca material used in green roof design

Abstract

Climate change comes with higher intensity precipitation and the need for stormwater management increases. Green roofs are low-impact, space-efficient, and can help with stormwater management on-site. For non-vegetated green roofs, detention is the major contributor. The detention of a substrate is dependent on hydraulic conductivity. Saturated hydraulic conductivity can be found by a lab-based permeameter test, but if one wants more information more tests need to be conducted. Another method is to use computer tomography (CT) and making a structural model in 3D. With correct image processing one can simulate the fluid flow and hence get the saturated hydraulic conductivity, and other properties like tortuosity, using open source codes. In this thesis both methods are used, to get results for saturated hydraulic conductivity for four different samples of LECA aggregates. Different image processing elements are looked into, such as binary process, threshold, cropping and region of interest. The results show that simulation-obtained results are elevated by 1-2 orders of magnitude over the lab-based results. No sample was simulated at its true representative element volume, and the results are therefore approximations. But overall the results are in the range of other results described in the literature.