Improving the effectiveness of mitigation and monitoring for bats on roads and railways in the UK Anna Berthinussen & John Altringham
Two year government funded project Prompted by PhD research showing that: Roads can have large scale negative impacts on bat abundance and diversity Current mitigation practice is failing (untested and/or ineffective) Berthinussen, A. & Altringham, J. (2012) The effect of a major road on bat activity and diversity. Journal of Applied Ecology, 49, 82-89 Berthinussen, A. & Altringham, J. (2012) Do Bat Gantries and Underpasses Help Bats Cross Roads Safely? PLoS ONE 7(6): e38775.
Aims: Improve current standards for road/rail developments and bat mitigation Develop bat survey techniques for pre, during and post road and rail construction, to assess impacts and the effectiveness of mitigation Provide robust standardised protocols for practitioners & practical guidelines for successful future mitigation
Two survey methods developed: 1) Assess the impacts of roads and railways on bats Landscape scale method using transects to assess impacts on a population level 2) Assess the effectiveness of crossing structures over/under roads for bats Local scale observational methods
1) Assessing impact : Landscape scale transect study 10 x 1 km walked transects perpendicular to the road at each site Stationary spot checks at 100 m intervals Record bat activity for 10 min at each spot check (and weather/habitat) Transects commence 30 min after sunset and are walked in different directions
1) Assessing impact : Landscape scale transect study 10 x 1 km walked transects perpendicular to the road at each site Stationary spot checks at 100 m intervals Record bat activity for 10 min at each spot check (and weather/habitat) Transects commence 30 min after sunset and are walked in different directions Tested at 7 road and 2 railway sites
1) Assessing impact : Landscape scale transect study Analysis Automated call identification software e.g. BatClassify (www.bitbucket.org/chrisscott/batclassify) Statistical modelling: Generalised Estimating Equations (GEE) A multiple regression model that accounts for auto-correlation Interpretation Precautionary principle Statistical significance is not the only factor large declines (e.g. over 20%) that are nonsignificant should also be considered due to the intrinsic variability in bat activity data and the difficulty of detecting change.
Results Negative effects at all 3 motorway sites established, under construction & recently completed Onset is rapid, construction is a concern Activity down Diversity down
Results Similar effects found on some A roads (3 of 4 studied), including two-lane, single carriageway: Activity down Diversity down
Results Some railways also show the effect (1 of 2 studied) Activity down Diversity down
2) Assessing the effectiveness of crossing structures Visual observations of crossing bats paired with echolocation recordings 6 x 60 min surveys at dusk or dawn per site Count crossing bats and record flight height and distance from crossing structures - LED markers for distance - Night vision and infra-red lights in underpasses
2) Assessing the effectiveness of crossing structures Visual observations of crossing bats paired with echolocation recordings 6 x 60 min surveys at dusk or dawn per site Count crossing bats and record flight height and distance from crossing structures - LED markers for distance - Night vision and infra-red lights in underpasses Tested at 9 mitigation structures
2) Assessing the effectiveness of crossing structures Analysis Set definitions for: Use of the structure: bats flying within 5 m of it / through an underpass Unsafe crossing height: < 5 m above road Use boxplots / percentages to compare proportions of bats using structure to those not using it or at risk of collisions with traffic Wilcoxon Signed Rank test to compare numbers crossing before and after construction Interpretation To be effective: A similar number of bats must be using the commuting route before and after construction, and at least 90% of bats must be using the structure to cross the road safely
Underpasses One underpass used by 95% of crossing bats: Large, wide and bridges an existing commuting route Two other underpasses less effective a third of bats crossed unsafely: A487
Wire bat gantries None were found to be effective
Wire bat gantries Two designs - none were found to be effective Few bats were recorded at any gantry Has construction already driven the bats away? Were there many bats present before construction?
Overpass Only 3 bats observed in the vicinity, none flew over the bridge
Environmental overbridge Some potential used by 62% of bats, but 19% crossed at unsafe heights
Landscape/green bridge Effective - used by 97% of crossing bats
Conclusions Roads and railways can have a negative impact on bats Mitigation is important both during and after construction Wire bat gantries don t work Green bridges and underpasses have most potential, but design and location are important Standardised and objective methods should be used routinely and results reported Similar issues for other wildlife
References Defra report: Berthinussen, A. and Altringham, J. (2015) Development of a cost-effective method for monitoring the effectiveness of mitigation for bats crossing linear transport infrastructure. Defra contract report WC1060 Available at: http://tinyurl.com/bats-roads Previous studies: Berthinussen, A. & Altringham, J. (2012) The effect of a major road on bat activity and diversity. Journal of Applied Ecology, 49, 82-89 Berthinussen, A. & Altringham, J. (2012) Do Bat Gantries and Underpasses Help Bats Cross Roads Safely? PLoS ONE 7(6): e38775