The return of the European beaver (Castor fiber) in Flanders, Belgium: ecology and risk assessment

Datum: 14 december 2015

Locatie: UAntwerp, Campus Middelheim, A.143 - Middelheimlaan 1 - 2020 Antwerpen

Tijdstip: 16 uur

Promovendus: Kristijn Swinnen

Promotor: Herwig Leirs & Erik Matthysen

Korte beschrijving: PhD defence of Mr. Kristijn Swinnen - Faculty of Science, Department of Biology


Beavers are large semi-aquatic rodents and well known for their ability to cut down trees and create dams. Both actions change the environment and modify the present habitat, which is why they are called ecosystem engineers. Both the Eurasian beaver (Castor fiber) and the North-American beaver (Castor canadensis) were heavily persecuted by humans, mainly for their fur. The Eurasian beaver (hereafter 'beaver') was almost completely exterminated by the beginning of the 20th century. Only eight small populations, totalling 1200 individuals, remained. Since then, increased protection and reintroductions combined with natural spread were responsible for a growth in population size and range. Currently, a minimum of 1 million beaver populate Eurasia.

Also in Belgium, beavers were once common in wooded river valleys. However, around 1900 all beavers were exterminated. Reintroductions in neighbouring countries were responsible for the first returning beavers in 1990 and between 1998 and 2001 (Wallonia) and in 2003 (Flanders) unofficial reintroductions took place. Because of the unofficial nature of the reintroductions, no scientific research was executed although beavers have the potential to influence landscapes which is not necessarily desirable in one of the densest human populated areas of Europe.

The aim of this thesis is increase ecological knowledge of beavers in a human dominated (predator free) landscape and consequentially use this ecological knowledge to mitigate possible human-beaver conflicts.

We collected data to examine if based on ecology, beavers have a future within Flanders. We examined beaver activity patterns by using camera traps and found they are mainly crepuscular and nocturnal. Furthermore, beavers increase activity during nights with bright moonlight. We hypothesise that beavers continue to be nocturnal, because of historic persecution by humans, although they are currently protected and no natural predators are present. We also developed a method to automatically distinguish between beaver and other camera trap recordings. We examined the genetic background of the Flemish beaver population(s). Hairs were collected non-invasively, and we analysed mitochondrial DNA from the hair roots. We found that beavers originating from 4 relictpopulations are currently present within Flanders and we have no indications for presence of C. canadensis within Flanders. Furthermore, our results indicate that the genetic exchange between both major watersheds is at best difficult.

We examined the potential habitat suitability of the currently uncolonised parts of Flanders. We analyse the habitat characteristics which are present at the locations where beavers are currently settled and examined the rest of Flanders for similar environmental conditions. This was done by using species distribution models. Not surprisingly, distance to water, willow stands, wetland vegetation and poplars were the main explanatory variables. We estimate that within Flanders there is sufficient habitat to support 924 beaver territories.

However, a location can comply with the minimal ecological requirements but beaver presence may create conflicts with other land-users. Main concern regarding the return of beavers is the construction of beaver dams. Beaver dams are built in some territories, but it is no requirement for beaver settlements. We examined which environmental parameters can explain the decision to build a dam or not. Water depth, water width, stream velocity, distance from woody vegetation and the bank height all differed significantly between control and dam territories. The best discrimination between territories was made using only water depth. When the water level was less than 68 cm, building of a beaver dam was highly likely. In deeper water, dams were not built. Whenever a confluence is present in dam territories, beaver dams are situated downstream of this confluence. In our study area, beaver dams increase water levels 15-87 cm (average 47 cm), bank overtopping occurred at two locations (with 21 and 29 cm of water above the lowest point of the bank upstream of the dam) and in other territories water level could rise 1 to 71 cm before bank overtopping would occur.

Our results can be used as a management tool in order to evaluate possible risks linked with the return of beavers in a human dominated landscape. Management measures can be used to prevent or mitigate beaver-human conflicts. Increased monitoring at locations which are highly likely to be colonized, and the assessment of critical points within these areas (poplar plantations, weak river banks, locations sensitive to flooding) can take place. Also structural measures such as bank reinforcements or measures to prevent the blockage of narrow water passages can be taken in areas which are predicted to have a high potential for beaver settlement. By preventing or quickly solving human wildlife conflicts, long-term coexistence between humans and beavers can be achieved.

In conclusion: we expect beaver populations to continue to grow and easily reach the target population of 167 territories, aimed for by the species protection act. We discuss how our findings can contribute to improve the species protection act and conclude that long term coexistence between a beaver and human population is possible, as long as there is sufficient support from society. However, this does not imply a problem free cohabitation. Conflicts with beavers will occur, but an improved ecological understanding combined with a good beaver management can minimise the future conflicts.