Controlling soil erosion in a changing climate: evaluating suitable plant species in grassed waterway design.

Soil erosion is a global problem which needs mitigating due to the on-site and off- site impacts it causes. Soil erosion is set to become an even greater problem due to climate change. Climate change is likely to increase the intensity, frequency and duration of precipitation events. This change in precipitation will increase flow erosivity and thus increase the chance of soil detachment. Grass-based erosion mitigation features will have to be able to withstand a higher volume of water as runoff volumes will increase due to climate change. An increased surface runoff rate will increase sediment transport capacity leading to more soil erosion when coupled with an increased detachment rate therefore solutions for the future need to be researched. Grass-based erosion mitigation features such as swales, buffer strips and grassed water ways (GWWs) have been shown to be effective. In this study, Festulolium Bx511 (F2), Festulolium cv Prior (F1) and a mixture of Festuca rubra and Lolium perenne (C) were used in mixtures and monocultures to investigate their efficacy in mitigating erosion. Experiment 1 used growth rooms under different climatic conditions, a summer scenario (22°C) and an autumn scenario (15°C). There were also different rainfall scenarios, drought (No rainfall), normal (100 % rainfall based on average rainfall (1981 – 2010) average rainfall (1981 – 2010) data from the Met Office) and excess (200 % of average rainfall (1981 – 2010) based on data from the Met Office) to see how they would affect the plant traits needed for erosion control. For summer establishment conditions the normal rainfall value was 49.2 mm, and the excess rainfall was 98.4 mm. For autumn establishment conditions the normal rainfall was 81.9 mm, and the excess value was 163.8 mm. A plant trait ranking system was devised, the species which showed promise were taken forward and used within hydraulic flume experiments to assess actual soil erosion mitigation potential. Plant traits linked to erosion control include both above ground (% cover, plant height, number of stems, number of tillers, stem diameter (mm), stem area density (mm² mm-²), above ground biomass (g) and below ground traits (root total length (cm), root total surface area (cm² ), root diameter (mm) and total root length (cm) of ≤0.25 mm diameter. Climate change is likely to change how grass plant traits are manifest due to the differing climatic conditions. Therefore, any solutions currently promoted that utilise grass monocultures and mixtures for erosion mitigation features such as GWWs may need to be revised to mitigate for climate change. Conclusions from Experiment 1 include that species selection for soil erosion control features such as GWWs must consider potential rainfall and temperature conditions during the grass establishment for optimal erosion control. There were, however, two species combinations which could be considered as year-round candidates, Fest_1+Fest_2+C and C. Experiment 2 was a hydraulic flume experiment where the inflow rates used were 0.2 – 1.4 l s¯¹. Significant differences in the following plant traits; number of stems, number of tillers, stem diameter (mm), stem area density (mm² mm-²), total root length (≤0.25 mm ⌀), total root surface area (cm ² ), and root diameter (mm) were observed between different treatments. Conv had a significantly higher number of stems as compared to all other experimental treatments. Fest_1 had a significantly higher number of tillers, stem diameter and stem area density as compared to all other treatments. Fest_1+Fest_2+C had a significantly higher total root length (≤0.25 mm ⌀) as compared to Conv. Fest_1+2 had a significantly higher total root surface area than the Fest_1 and Fest_1+Fest_2+C experimental treatments. Fest_1+Fest_2 had a significantly higher root diameter as compared to the Fest_1+Fest_2+C experimental treatment. However, significant differences did not manifest in sediment concentration. In conclusion, it did not matter if grass species monocultures or mixtures were used as there were no significant differences in sediment concentration between the experimental grass treatments. Experiment 3 was also a hydraulic flume experiment where the inflow rates used were 0.2 – 2.6 l s¯¹. In this experiment there was a lowered seeding rate (L) and a recommended seeding rate used (N). There were significant differences in plant traits and also in sediment concentration. The critical thresholds for the Environment Agency (EA) major event classification of 1000 mg l¯¹ to be reached were determined for Experiment 3. There were several experimental grass treatments which did not breach the limit set out by the EA (Conv N, Fest_1+2 L, Fest_1 N and Conv L). In conclusion the Conv L, Fest_1 N and Fest_1+2 L species treatments should be recommended for farmers for use in soil erosion mitigation features such as grassed waterways. This thesis was made possible thanks to the soils training and research studentship centre for doctoral training (STARS CDT). It was funded by Biotechnology and Biological Sciences Research Council (BBSRC) and Natural Environment Research Council (NERC), Grant/Award Number: NE-R010218-1.

PhD in Environment and Agrifood