Blog post by Malcolm Burgess, Senior Conservation Scientist, RSPB Centre for Conservation Science

This week we published a paper with a number of partners exploring the 'trophic match-mismatch' hypothesis. But why have we been researching this? The reason is that it’s an important theory in the context of how birds, and other taxa, are affected by climatic change.

This is the first study of the extent of mismatch between birds, caterpillars and oak leafing at a large spatial scale, and provides new insights into the variation in the extent of mismatch across the UK.

What is the trophic match-mismatch hypothesis?

Many woodland birds, including our study species (blue tits, great tits and the declining pied flycatchers), need to time breeding to coincide with the availability of caterpillars. Caterpillars are the most important food for these breeding birds, and woodland caterpillars are only abundant for a short time during spring.

These birds nest only once a year, so timing is critical. They need to time when they lay their eggs to ensure their chicks are largest and hungriest at the same point in spring that caterpillars are most abundant. Too early, or too late, and they are mismatched, and there will not be enough food for adults to raise them.

Male pied flycatcher. Image by Ben Andrew (

How is it changing?

The onset of spring has been advanced by climatic change in recent decades, with warmer temperatures now occurring earlier in spring. In response to warmer temperatures, oak leaves and caterpillars also emerge earlier and consequently birds need to breed earlier to avoid being mismatched with their food supply.

Previous studies of match-mismatch have mostly been from single locations and study how it varies between years, but the extent to which mismatch varies across large spatial scales was previously unknown.

This is important to know, because if a population in one location suffers from being mismatched, it can be subsequently 'rescued' by other populations that were not mismatched through dispersal. If there is no spatial variation in the extent of mismatch, i.e. all populations are affected in the same way, this is unlikely.

Oak leaves. Image by Andy Hay (

Using an army of citizen scientists

This work has only been possible because of hundreds of citizen scientists who collected most of the data we used. Data collected by citizen scientists across the UK have been coordinated by researchers from the RSPB/University of Exeter and University of Edinburgh, and by Ken Smith.

The timing of oak leafing was determined through more than 10,000 contributions to the UK Phenology Network, coordinated by the Woodland Trust.

Each year recorders from the network note their observations of first events such as hearing the first cuckoo or seeing the first brimstone butterfly. We were able to summarise the timing of first oak leafing, which caterpillars are particularly associated with, for all 5x5km squares in the UK.

Using poo to count caterpillars

It is hard to measure the abundance of caterpillars, but since 2008 a small network of people across Britain have been doing so following a protocol devised by Ken Smith.

I have blogged about this before, in short we use a simple collecting tray placed under oak trees that is replaced twice per week throughout the spring. The trays contents are then sorted and dried, and because caterpillar poo (also called frass) is a distinct cylindrical shape it can be separated from everything else before being weighed.

Female pied flycatcher with caterpillars

There is usually a very clear peak in frass (i.e. caterpillar biomass) during the spring. When this peak occurs can vary by several weeks between years, and we already know the peak is related to temperature - in a warm spring the peak is earlier (and shorter, so caterpillars are available for less time) compared to a cold spring.

For the bird nesting data, we used data from nearly 100,000 nests monitored over 56 years for the BTO's Nest Record Scheme. Nests are monitored by citizen scientists across the UK following the methodology of the scheme, typically weekly visits to nests to count eggs and nestlings. Like the oak leafing data, the calculated first egg dates for each bird species were summarised annually for all 5x5m squares in the UK.

What we found

Our study shows that at more northerly latitudes the phenology of oak, caterpillars and birds is later, especially oak. Mismatch between caterpillar availability and bird breeding is more prevalent in earlier (warmer) springs, and is greatest for pied flycatchers. Most importantly we found little spatial variation in the magnitude of caterpillar-bird mismatch for any of the birds.

The greater population decline we have seen for many insectivorous birds in the south of the UK therefore does not appear to be caused by greater mismatch in the south than the north, and phenological mismatch alone is unlikely to explain spatial variation in population trends.

Our result also has the practical implication that insights into the degree of mismatch in one location can be generalised to trends at different latitudes. As springs continue to get warmer woodland birds will become increasingly mismatched with peak caterpillar abundance.

Read the full paper on the Nature Ecology & Evolution website.

Citations: Burgess, M. D., Smith, K. W., Evans, K. L., Leech, D., Pearce-Higgins, J., Branston, C. J., Briggs, K., Clark, J. R., du Feu, C. R., Lewthwaite, K., Nager, R. G., Sheldon, B. C., Smith, J. A., Wytock, R. C., Willis, S. G. and Phillimore, A. (2018). Tritrophic phenological match-mismatch in space and time. Nature Ecology & Evolution.