Suitability of two sensor technologies to monitor spatio-temporal patterns of soil moisture on extensive green roofs

Abstract

Green roofs have the ability to mitigate stormwater runoff and to reintegrate wild-life and biodiversity in urban areas. However, they constitute complex ecosystems that still encompass grey areas. For instance, there has been only few research conducted on the accuracy of sensors such as capacitive EC-5 or thermic PlantCare, given the specific composition of green roofs substrates. In addition, green roofs often present spatial heterogeneity such as differences in substrate depths or partially shaded areas. This heterogeneity is expected to impact SWC and consequently plants development. Hence, the objectives of this thesis was to test both types of sensors in green roof substrate and to monitor - with abiotic parameters and vegetation coverage - two plots on an extensive green roofs with two different depth and shadowed areas. Firstly, PlantCare appeared as the most suitable sensor for green roofs application, given EC-5 dysfunctions due to poor contact between the sensors and the substrate. Secondly, the difference in depth has an influence in SWC, since the deepest plots presented a higher SWC during the entire experiment. As a consequence, a disparity in vegetation coverage was noticed, as the deepest plot presented more than 20 % of plant coverage while its neighbour had only 5 %. Finally, the shadowed area presented a higher coverage of plants than the rest of the plot. However, the behaviour of SWC was more complicated to seize at this scale due to the presence of plants and the small volume of influence of Plantcare.