| dc.description |
Eurasian beavers (Castor fiber) were extirpated from Great Britain (GB) ca. 400 years ago. Harvested for their meat, pelts and castoreum, their numbers across Eurasia were reduced to a few isolated populations. In recent years, beavers have been reintroduced in GB and their numbers have increased across Europe due to conservation efforts. However, the landscapes that beavers are now returning to have been significantly altered by anthropogenic landuse. This land use change has had hugely detrimental impacts for natural riverine and riparian processes, with respect to their structure and function. Beavers are well known for their industrious behaviours: building dams and lodges, excavating burrows and canals, and felling trees. Beavers therefore act as a significant ecological and hydrological disturbance, creating dynamic, structurally heterogeneous wetland ecosystems. Not only do the impacts of beaver enhance biodiversity through the provision of diverse habitats, but they also help to restore natural hydrological, geomorphic and ecological processes that are all but lost across intensively-farmed, densely populated European landscapes. Beavers may therefore play an important role in restoring this ecosystem function and could potentially help to mitigate the harm caused by anthropogenic landuse. However, we now rely on agriculture and infrastructure; the expansion of beaver populations can consequently result in conflict where their impacts intersect anthropogenic activity. As such, there is a requirement to better understand the impact of beavers on the structure and function of natural processes, to inform their management and conservation. Further, there is a need to develop methods that can allow us to predict the spatial and temporal changes in beaver populations across modern landscapes to underpin the recovery of the species in such a way that their benefits can be maximised whilst minimising the risk of potential conflict. This thesis presents four papers to advance our scientific understanding in this regard, as follows:
The hydrological mechanisms that cause storm event peak flow attenuation in beaver wetlands were explored at a beaver dam complex on a third order stream. Data from 612 discrete flow events were measured at a flow gauge, downstream of a beaver dam complex, seven years before and three years after it was constructed; 634 events from a neighbouring control catchment, over the same time period, were also extracted from the time series. A selection of general linear models were fitted between event peak flow and total event rainfall. The differences in the slope of the regression, before and after beavers, indicate that flow attenuation, due to beaver activity, increases with greater rainfall. This increasing attenuation effect is attributed to floodplain flow diversion and transient storage because the observed attenuation volumes greatly exceed the available storage capacity of the beaver ponds alone.
Drone-derived structure from motion photogrammetry surveys were carried out, providing a high-resolution understanding of changes in woodland canopy structure, over a one-year period. Riparian woodland has a complex structure and uncertainty in estimated point elevations can be spatially patchy and locally high. The adoption of robust error propagation methods to accurately estimate canopy height change was found to be very important. Beaver foraging slightly reduced mean canopy height but significantly increased the variability in canopy height change. Quantile regression was used to quantify the difference in canopy elevation change across two regions of riparian woodland: with and without evidence of beaver foraging. The rates of canopy growth and height decline were greater in regions where beavers were actively foraging, indicating that beaver foraging may increase canopy height variability which could have varying implications for riparian/aquatic species and woodland management.
In order to better predict the landscape scale impact of beavers on ecosystem structure and function, it is necessary to develop methods to accurately predict their potential habitat distribution and where dams, which have the largest environmental impact, are likely to occur. To address this, we developed a modelling approach using high resolution, nationally-available datasets to create a Beaver Forage Index (BFI) model – a raster dataset describing the suitability of landcover for beaver forage and; a Beaver Dam Capacity (BDC) model which describes the density of dams that could be supported within a given reach. Beaver preferentially foraged in regions with higher BFI values and are more likely to dam (and build more dams) in reaches with higher BDC. Using these models, it is possible to estimate the number of dams that might occur at the catchment scale at beaver population capacity.
Though beavers have only been living in the wild in GB for a short period of time, their populations are growing rapidly. It is essential to build a stronger understanding of how beaver populations expand, at what rates and how management interventions, such as translocations or lethal control, might impact population dynamics. To gain this insight, we conducted annual beaver feeding sign surveys to map the distribution of beaver impacts throughout the River Otter catchment, SW England. Using a semi-automated approach, that combines kernel density estimates and expert knowledge, the number of territories in the River Otter catchment was estimated over a 5-year period. A spatially explicit method for predicting the catchment population carrying capacity was developed which uses BDC and BFI models in combination with empirical understanding on territory sizes from across Europe. Adopting the assumption of logistic growth in beaver populations, we use the observed rates of population increase, constrained by the estimated carrying capacity range to model the expansion rate of the beaver population. A range of theoretical management scenario simulations were carried out revealing that, even low-moderate management interventions may have very uncertain outcomes for population viability and therefore any management plan, involving translocation or culling of animals, should be carefully designed.
The findings presented in this thesis advance our understanding on the impacts of beaver on hydrological function, riparian woodland structure and provide methods for understanding the spatio-temporal distribution of beavers and their impacts. This understanding has already been used to inform management policies within national agencies and non-governmental organisations across GB and has the potential to inform the management of beavers across Europe. |
|