From horticultural greening to urban ecological restoration: How to favour herbaceous biodiversity when de-sealing

Growing demand for nature in cities and the United Nations Decade on Ecosystem Restoration are driving urban restoration efforts. However, practical guidance on how to implement these projects effectively remains limited. This study investigates the effects of reusing excavated soil, sowing strategies, and site location along an urban-rural gradient on herbaceous establishment in urban environments after de-sealing. Based on a mesocosm experiment, we tested three sowing types (no sowing, sowing a mixture of dicotyledonous species, and sowing a mixture of dicotyledonous species and grasses), two soil types (enriched topsoil and excavated poor coarse materials), and an urban-rural gradient. The development of the vegetation was monitored during 12 surveys carried out over the first two years of growth. The results showed that enriched topsoil and sowing (especially with grasses) boosted vegetation cover and species richness. However, the effects were limited after one growing season. Mesocosms with no sowing exhibited more heterogeneous plant communities and supported a wider range of spontaneous species, indicating that low-intervention approaches can promote urban biodiversity while reducing financial and resource investments. Mesocosms containing a high proportion of poor, coarse excavated materials tend to have a more heterogeneous composition. In contrast, the location of the site had a limited influence on outcome of the vegetation. Our results highlight the potential of spontaneous colonisation of nutrient-poor, coarse, excavated materials as a sustainable alternative to conventional urban greening practices. These results provide urban planners with practical guidance, supporting low-input restoration approaches that balance biodiversity conservation with resource efficiency.