A thank you and some results!

Christine Tansey's picture

You may have noticed that Track a Tree has been particularly quiet over the last few months, and as I mentioned in a previous post, this has been the result of some intense thesis-writing.

I am now delighted to say that I recently submitted my thesis. Two chapters were based on the records you collected through Track a Tree, with a further two chapters based on observations submitted to our big sister project, Nature’s Calendar. It is hard to say how grateful I am to everyone who has participated in or supported Track a Tree during the last few years. While I hope that the project will have a life beyond my PhD, the impact it has had on my own research, and my enjoyment of spring in the woods, cannot be underestimated. Thank you all.


I’m now waiting for my viva (the oral examination/defence of the thesis) to be scheduled. In the meantime, I’d like to share a few of my key findings, which cover some analyses using Nature’s Calendar, as well as my work on Track a Tree. There will be further updates as I prepare the chapters for submission to scientific journals, where they will undergo peer review, so please forgive the brief nature of this summary.

As all of you Track a Tree-ers will be familiar with, the seasonal timing of growth and development in plants and animals is called phenology, and includes events such as the flowering of plants during spring. Changes in the timing of phenological events are well recognised as one of the key biological impacts of climate change, and citizen science schemes such as Nature’s Calendar have played a key role in identifying such changes. To understand and predict how these changes will affect species in future, we need to know how phenology responds to aspects of the environment such as temperature.

In my PhD work, I explored the role of temperature on the flowering and leafing times of a range of UK plants, as well as examining how changes to the phenology of one species could affect other species that they interact with. Below, I have described the key findings from each chapter of my thesis, split into sections for Nature’s Calendar and Track a Tree results.

Nature’s Calendar

The first part of my research looked at the phenological optimum, which is the timing that will lead to the most successful growth and reproduction of a plant. This optimum is likely to depend on environmental conditions such as temperature. If long-term temperatures change, this might alter the optimum time for plants to produce leaves and flowers.

I used over 200,000 phenological observations that had been submitted to Nature’s Calendar, to test whether 22 plant species would be able to keep up with changes in their optimum phenological timing, that might arise as a consequence of climate change. I identified the most important cues for leafing or flowering spring events in these species. All species were sensitive to the warming temperatures of spring, and I found that their spring events would get earlier by between 3-8 days for each 1°C increase in temperature. For some species, cooler temperatures in autumn or winter, or day-length, were also important cues. Those with phenology in the earlier part of the year tended to be influenced by cool temperatures, and those with later phenology tended to respond to day-length.

In this group of 22 species, I found that seven species are likely to be able to keep up with, or track, future changes in their optimum timing. However, I also found that four species may not be able to track the rate of change. The remaining eleven species tested may be able to partially keep up with change in the optimum conditions from year to year. I wasn’t able to draw a firmer conclusion for these eleven, as my results for these plant species showed less clear patterns.

The next stage of my research used the same 22 species, and I tested whether they had any characteristics that could help predict how their phenology responds to temperature. I found that the ability to keep up with different temperatures from year to year, also known as plasticity, depends on whether the event type is flowering or leafing, and whether the plant is a woody or herbaceous perennial. These traits may help make predictions about how the phenology of different species with similar characteristics could respond to future changes in temperature. However, this finding needs to be further investigated in future work to gain a better idea of what traits are important.

Track a Tree

The last two main sections of my thesis were based on Track a Tree, which as you know, was set up to examine the effect of shading on the competition for light between canopy trees and flowering plants in woodlands. All of our of wonderful participants collected unique phenology observations of plant species that interact with one another. I then used your records to examine the relative timing of canopy tree leafing and the flowering of ground flora species in UK woodlands, much as I described in my blog post on the initial findings from Track a Tree in 2014.

I found that the first leafing and peak flowering of several trees and flowering plants was consistent across the UK. For example, wood anemone peak flowering was around 22 days earlier than the first leafing of oak trees wherever they occurred together, whether this was in the North or South of the UK.

In the figure below, you can see the timing of a) lesser celandine, and, b) bluebell peak flowering, plotted against the timing of oak leafing dates in different woodlands across the UK. Each of the dots represents the mean timing at a single wood in a single year (2013 = orange, 2014 = grey, and 2015 = black). The dashed line represents flowering and leafing of the two species happening at the same time. The solid black line is the average relationship between the two species, found by the statistical model (based on your records) that I ran in these analyses.

The solid black line falls below the dashed line for lesser celandine (see a), which means that its peak flowering happens earlier than oak leafing across the UK. For bluebell (see b), the solid black line falls very close to the dashed line, which means that bluebell peak flowering happens at approximately the same time as oak leafing across the UK. I found that the relative phenology for these species is consistent across the UK, which is shown by the solid black lines being close to parallet to the dashed lines.


The timing of oaks versus birch, and oaks versus ash, was also consistent across the UK, but it was different in different places for oak versus the non-native species, sycamore. In colder parts of the UK, oak leafing happens later, but sycamore leafing takes place at a similar time regardless of where in the UK sycamore is.

The results from this part of my work suggest that the flowering of species on the woodland floor may be able to track canopy leafing times of native tree species in future conditions, which we can predict because they already occur in different conditions across the UK. This could help maintain patterns of shading and competition for light in woodlands. The result for sycamore is especially interesting however, as it is a widespread non-native species in our woodlands, yet we may be less able to predict how it will compete for light with native trees in the future.

Finally, I used the Track a Tree data to test how variable tree leafing phenology is, both for different tree species and for the same species in different woodlands. I found that tree leafing varies by different extents for different species. For example, this means that first leafing in a number of individual trees could happen over a period of two weeks in one species, but over a month in another species. However, I did find that there was not very much variation in leafing date observed among individual trees of the same species in the same woodland. This means that in any particular wood, species that depend on the leafing of woodland trees, such as caterpillars, may only have a narrow window of opportunity to make use of them (e.g. by feeding on new leaves) during the spring. This finding may help us understand how the leafing of different trees will affect the species they interact with in future.

Overall, the research I have done as part of my PhD has helped broaden the range of approaches that can be used to understand plant phenology in a changing climate. With your contribution as citizen scientists, and new analytical techniques (yes, there was a lot of statistics in my thesis…), hopefully we’ll be able to continue to improve predictions of how phenological changes in plants will affect the other species that interact with, or rely upon, them.

For my thesis, I analysed the records collected as part of Track a Tree between 2013- 2015, but I was unable to include 2016's observations, as spring in the woods was happening while I wrote. I will be including all records from 2016 in my next analysis, which will be part of preparing a journal paper. Please do send in any of your records you have not yet submitted so that they can contribute to the next stage.

Once again, I’d like to thank everyone who has had any involvement in Track a Tree, as well as all the contributors to Nature’s Calendar. Without citizen science we would know a great deal less about phenology!




Hi Christine. I'd be interested to hear what statistical analyses and tests you used.  I presume from the charts that you used regression, but did you find any other techniques to be relevant?

Christine Tansey's picture

Hi Steve, thanks for your question, I'll send you an email tomorrow with a little more information about the stats. Hope that is ok!