Beyond the mean: using drones to understand spatial drivers of phenological responses
– Can time series of remotely-sensed multispectral reflectance data be used to (a) identify the species of tree crowns and (b) reliably measure the spring green-up phenology of individual trees?
– Does spatiotemporal variability in the green-up of deciduous trees lead to spatial and temporal variation in optima for higher trophic levels?
– Does spatial variation in (a) local tree species diversity or (b) local tree phenology act to reduce selection on phenological traits in consumer populations, hence buffering them from temporal variation in climatic variability?
As climates have warmed over recent decades, populations of many species, across a wide range of biomes, have shifted their phenology in response at variable rates. A growing body of research suggests that phenotypic plasticity is the key mechanism underlying these responses. Hence, knowledge of the evolutionary forces acting on plasticity is vital to understand the scope for populations to cope with rapidly changing environments. However, studies of these ecological responses tend to summarise them in terms of changes in population means in response to changes in mean environmental factors. This focus on population-level responses collapses spatially variable processes operating over multiple scales down to single points. For example, spatial variability in spring green-up of deciduous trees can vary markedly across years and across space, likely leading to spatial and temporal variation in optima for other trophic levels for which these are key drivers. A major gap in our current knowledge concerns how this spatial variability selects for plasticity within populations. Examination of this question has been hampered by the challenging nature of carrying out empirical work, which requires highly detailed data on the phenological landscape experienced by organisms (i.e. measuring vegetation phenology at the scale of individual plants). One way to overcome this challenge is to use drones to collect detailed remotely-sensed data on vegetation phenology.
Aims of the Project
The key aim of this project is to use the classic tri-trophic model system of passerine songbirds, caterpillars and deciduous trees to explore how spatial variability in vegetation phenology selects for plasticity in timing of breeding within wild populations of great tits (Parus major) and blue tits (Cyanistes caeruleus).
The project will involve using a drone to collect time series of remotely-sensed data on the spring phenology of a c. 100 ha plot of woodland. These data will be used to develop a method for reliably measuring spring green-up timing for individual trees and therefore describe the phenological landscape experienced by tits. Tree phenology data will then be used together with information on insect phenology and individual blue and great tit breeding attempts in fixed-location nest boxes, to test hypotheses linking spatial and temporal variation in tree and insect phenology to the strength of selection on phenology of birds. Finally, this project will explore how spatial variability in phenology of trees can buffer consumer populations, across multiple trophic levels, against temporal variation in climatic variability.
This research will thus generate unique insight into how individuals respond to complex phenological environments, significantly deepening our understanding of the processes governing how organisms cope with climate change. The approaches that will be tested as part of this project will, potentially, be generalisable to many different seasonal environments where remote-sensing can inform about small-scale spatial and temporal drivers.
Methods to be used
Innovative drone image capture and machine-learning technology will be used to characterize the spring green-up for the entire canopy of a 100-ha study plot in Wytham Woods near Oxford, at the scale of individual trees. This will be combined with breeding data from long-term nest box populations of great tits and blue tits. Data will also be collected on insect timing across the study plot.
Specialised skills required
No specialist skills are required, but experience with spatial statistics and handling and analysing large datasets will be highly beneficial.
If you are interested in applying for this project, please contact Ben Sheldon firstname.lastname@example.org, Ella Cole email@example.com and Yadvinder Malhi firstname.lastname@example.org in the first instance.
- CASE Projects available for students starting in 2020
- Mechanisms for variability of the Quasi-Biennial Oscillation in laboratory and numerical models
- Improving environmental, social and economic outcomes of nationally significant infrastructure projects
- Integrating socio-economic impacts in marine resource management
- Mechanisms and impacts and of host-parasite relationships
- Quantifying the role of AMOC decline in climate change impacts
- Drivers of life history variation in flatworms
- Movement analysis of zooplankton through filming
- Can humanity have its cake and eat it too? Extending the Mitigation Hierarchy to support implementation of international biodiversity targets in the context of the Sustainable Development Goals
- Near-source effects on global wavefields: A forensic seismology study applied to nuclear monitoring
- The fate of mercury (Hg) during thermal maturation of sediments and its implications for interpreting the geological record
- Tracking ocean circulation in the Cretaceous ‘Chalk Sea’
- Native and invasive ladybirds in a changing U.K. climate
- A unified approach to multi-scale modelling of wind farms and complex terrain
- Terrestrialisation in vertebrates using evidence from synchrotron tomography
- The dynamics of flowing, subsurface salt sheets: observational constraints and theoretical models
- Beyond the mean: using drones to understand spatial drivers of phenological responses
- Global measurements of Sea Surface Temperature using the Infrared Atmospheric Sounding Interferometer
- Using marine UAVs to integrate seabird census with phenology
- Seabird Watch: testing out of the box image analysis methods and citizen science data for large-scale monitoring of seabirds.
- Bayesian models for the analysis of botanical citizen science data: Understanding bias and improving inference