Spatial design improves efficiency and scalability of seed‐based seagrass restoration

Abstract

Abstract Coastal ecosystem restoration is often ineffective and expensive in practice. As a consequence, upscaling restoration efforts to functionally relevant spatial scales remains one of the largest hurdles for coastal restoration practice. On small scales, restoration success of vegetated ecosystems (i.e. salt marshes and seagrasses) can be amplified by spatial designs that harness positive interactions. However, it remains unknown if positive interactions can be harnessed with seed‐based approaches, that are considered to be more cost‐effective and scalable than traditional shoot‐based restoration methods. Here, we investigated with a full‐factorial seeding experiment if (1) restoration scale (4, 40 and 400 m2) and (2) seeding density (10 and 50 injections/m2) affected multi‐year recruitment efficiency (measured as restored plants/seed injection) of annual eelgrass Zostera marina in the Dutch Wadden Sea. We found that the largest restoration scale (400 m2) increased second‐generation recruitment efficiency by suppressing a sedimentation‐related negative feedback. With increased restoration scale, the inner parts of the restoration plots captured less sediment, which decreased the desiccation stress of the restored eelgrass during low tide. Due to this stress alleviation, plants grew larger and produced more seed‐bearing spathes, which the following year resulted in two and three times higher recruitment efficiency at the largest restoration scale compared to the smaller scales. Moreover, lower seeding density more than doubled second‐generation recruitment efficiency compared to the higher density, supporting recent work showing that the effectiveness of ‘clumped’ spatial designs is context dependent. Synthesis and applications. The efficiency of restoration efforts is seldom taken into account, but can offer restoration projects a valuable metric with which workload, donor material and cost‐requirements can be reduced. We demonstrate that simple modifications to seed‐based coastal restoration designs (e.g. scale and density) can have a substantial impact on recruitment efficiency and multi‐year restoration yields. Thus, optimised restoration designs can strongly contribute to the upscaling potential of coastal ecosystem restoration. However, optimal restoration designs are expected to be strongly context dependent and we therefore argue that investigating optimal designs should be adopted as common practice, providing a crucial steppingstone between ‘proof‐of‐concepts’ and true large‐scale restoration attempts (km2).