High Nature Value farmland (HNVf) refers to traditional agricultural landscapes which support high levels of biodiversity. HNVf is declining across Europe, through both intensification and abandonment, raising concerns about the conservation of species associated with it. One argument is that such HNV species are better described as being dependent on disturbance, and that management practices on HNVf mimics the effects of natural disturbance processes such as herbivory, fire, severe weather and flooding, which are often suppressed across much of Europe. If true, one innovative approach for conserving HNV species would be to restore natural disturbance processes, as advocated by the "rewilding" conservation movement. I set out to explore the feasibility of this approach, focusing on flooding. Restoration of flood regimes is receiving growing attention, not only for its biodiversity benefits but also to reduce the risk of flooding in downstream urban areas, improve water quality, and increase the amenity value of rivers. I carried out four linked studies. Each followed a broadly similar approach: I selected sampling locations both in floodplain areas and comparable surrounding farmland, I surveyed birds and butterflies as indicator taxa of biodiversity, I carried out habitat mapping, and I estimated disturbance through both field measurements and remotely sensed data. My first study was around the Pripyat river in Ukraine, which has a large floodplain that has had minimal human intervention. I found that several, but not all, HNV species had higher population densities in the floodplain than in the surrounding HNV farmland, and several more had approximately equal population densities in the two habitats. This suggested intact flood regimes can retain some species that are elsewhere restricted to HNVf. My second study was around the river Rhône, France, where a restoration project has increased flow in floodplain channels and improved their connectivity with the main channel. I compared restored and non-restored segments of floodplain channel, as well as farmland and other habitats. While I found significant differences in population densitiesbetween habitat types, few species showed significant differences between restored and non-restored segments, suggesting that restoration had only limited ecological impacts on the wider landscape. My third study investigated a more ambitious restoration project which has reconnected a large area of floodplain to the river Peene in Germany. I found many HNV species had equal or higher population densities in the restored floodplain than in the surrounding unrestored farmland. This demonstrates that it is possible to restore the effect of natural flood disturbance on habitats and species. However, not all HNV species benefitted from flood-disturbed habitats, so clearly other conservation interventions need to be considered. For the final study, pooling data from all three sites I looked for associations between species' traits and environmental variables. My aim was to identify traits which predict species' responses to the environment, and hence which species are most likely to benefit from restoration. However, I did not find any significant associations in my data. In conclusion, I demonstrated that flood-disturbed habitats are suitable for many species traditionally associated with HNV farmland, and further that restoring flood disturbance processes can recreate those habitats. As such, reconnecting rivers and restoring floodplains would be a useful conservation intervention for species threatened by the abandonment or intensification of HNV farmland in Europe. However, other species did not benefit, and require other conservation interventions, perhaps other forms of rewilding, or continued incentivisation of favourable farming practices.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:763855 |
Date | January 2019 |
Creators | Kasoar, Timothy |
Contributors | Balmford, Andrew |
Publisher | University of Cambridge |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Source | https://www.repository.cam.ac.uk/handle/1810/287531 |
Page generated in 0.0014 seconds