Understanding the capacity of unimproved grassland to deliver natural flood management

Abstract

Natural flood management (NFM) is an emerging form of land management which works with natural processes to reduce flood risk. The evidence basis for NFM schemes is immature but is growing, with wetland unimproved grassland being a particularly understudied ecosystem in need of evidence-based research. The research presented in this thesis aimed to assess the extent to which unimproved grassland, a formerly extensive habitat that has been greatly reduced by intensive agriculture, could provide NFM benefits. This study specifically focused on unimproved grassland in southwest England known as Culm grassland; comprising of purple moor grass (Molinia caerulea) and rush pasture (Juncus effusus). Improved grassland (Lolium perenne) fields were used as a comparison/control as the grassland species which now dominates North-west European agricultural landscapes.

The study was divided into four objectives spanning a range of spatial scales to gain a multi-scale understanding of unimproved grassland hydrology. Objective 1 focused upon soil and water table properties using soil sampling and water table monitoring via dipwells. Results showed that M. caerulea and J. effusus dominated fields had on average significantly less compacted soil and slowly released water after rainfall events, compared to the more compacted and flashier L. perenne dominated sites. The research was continued in Objective 2 which used a rainfall simulator to mimic a 40 mm/hr rainfall event over plots of the three grassland types. Results showed a significant relationship between soil compaction and when saturation was reached, but volume of runoff was likely influenced by vegetation density within each plot, as no clear relationship between grassland type and overland flow was found. Objective 3 quantified surface flow pathway length in a field of M. caerulea and L. perenne using an unpiloted aerial vehicle and structure from motion photogrammetry to produce a detailed surface model through which surface flow pathways could be measured. The M. caerulea field had on average 1.4 times greater flow pathway length than the L. perenne control field and a rougher round/vegetation surface which could delay and store surface water. All three field experiments fed into Objective 4: sub-catchment rainfall-runoff response modelling of the Upper Tamar. Scenarios of 0, 10, 20 and 30% unimproved grassland restoration were modelled to assess potential impacts on within river flood hydrographs. Results showed peak flow and peak volume were reduced with increasing unimproved grassland extent. The four objectives of this study showed that unimproved grassland restoration has a place in the NFM toolkit, and that unimproved grassland can deliver other environmental benefits such as greater carbon storage and a diverse habitat for wildlife.