First signs of spring – not when they used to be

4 April 2014 by Malcolm Campbell, posted in Biology, Environment, Science

I am on the alert for the first signs of spring, to hear the chance note of some arriving bird, or the striped squirrel's chirp, for his stores must be now nearly exhausted, or see the woodchuck venture out of his winter quarters.” from Walden by Henry David Thoreau (1817-1862)

After a long winter, signs of spring are seen as welcome harbingers of easier times ahead. The emergence of early spring blossoms, like crocuses and snowdrops, proclaim the end of winter’s stasis. They usher in the coming seasons of growth and fertility.

For those that have been held in winter’s icy grasp for the past months, the signs of spring bring about great relief, and a change in demeanour. It’s no small surprise that those who endure the travails of the frosty season are very much on the lookout for any signs that it may finally have come to a close. spring2

One such person was Henry David Thoreau, the nineteenth century American environmental writer and philosopher. Between 1852 and 1860, Thoreau kept a keen lookout for signs of spring, and carefully documented them. Near his home in Concord, Massachusetts, Thoreau tracked the arrival of birds, the chirping of squirrels, and the emergence of woodchucks from their burrows.

Notably, Thoreau watched plants awaken from their winter slumber.

Thoreau documented when buds burst and the first flush of leaves appeared – the springtime “leafing out” of trees and shrubs. He was particularly studious in 1854, 1855 and 1860, making detailed records of the first flush of leaves to appear. These records have been retained to this day. Thanks to Thoreau, we have a remarkable record of the timing of spring in Concord more than a century-and-a-half ago.

Using Thoreau’s data we can actually make comparisons between the plant-derived signs of spring in Thoreau’s time compared with those same signs in Concord today. We can ask specific questions about the arrival of spring in Concord.

Are there differences in the timing of the first flush?

Are any differences attributable to changes in the environment between Thoreau’s time and today?

It turns out that the answers to both questions are “yes”.

In wonderful research done by Caroline Polgar, Amanda Gallinat, and Richard Primack, at Boston University, they compared Thoreau’s leafing out data with data they obtained between 2009 and 2013. Caroline Polgar and colleagues focused on 43 species of woody plants that lose their leaves in the autumn and regrow them in the spring – deciduous, perennial trees and shrubs. These species were selected because they are still present in Concord, and because there were more than 2 years of leafing-out data from the 1800s. Each spring, from 2009 until 2013, Caroline Polgar and colleagues recorded the leafing out dates for these species. Then they compared them with Thoreau’s data. spring4

Remarkably, Caroline Polgar and her colleagues found that 23 of Concord’s tree and shrub species are now leafing out an average of 18 days earlier than they did in the time of Thoreau. Why might this be?

An obvious answer is temperature. Since Thoreau’s time, springtime temperatures, over the months of March and April, have increased. In the 1850s, over those months in Concord, the average springtime temperature was 2.9oC. This is in contrast to the average temperature over the same time period between 2009 and 2013, which was 5.1oC.

This increase in springtime temperature in Concord is attributable to two factors. The first is global climate change. The second is a “heat island” effect that occurs in the environs of large cities – cities like Boston. Concord is nestled right next to Boston. In fact, of the two factors contributing to the increase in springtime temperature in Concord over the last 150 years, the urban heat island effect is likely to be the larger contributor.

Caroline Polgar and her colleagues addressed the possible role that the increased temperature around Concord might be playing in bringing about the earlier leafing out. They brought stem segments from the trees and shrubs into the laboratory and examined their response to cold and warmth relative to leafing out. They examined the way the plants responded to temperature in terms of the timing of their annual bud burst.

Woody plants are perennial plants – they live for years and years. Woody perennial plants that live in temperate zones have to contend with annual fluctuations in temperature. These changes can be substantial – from sub zero in the winter, to mid 30oC temperatures in the summer.

To survive the cold winter temperatures, woody perennials enter a period of dormancy in the autumn. They lose their leaves, and set hard buds that protect the cells that are responsible for the growth of the stem. In the autumn, shorter day lengths and cold temperatures can inform the plant to enter the dormant state. In the springtime, increased day length and temperatures inform the plant to exit dormancy and start growing again. spring1

It’s crucial that plants get the cues right in the springtime. If the days are too short, and the temperatures not warm enough, the plant runs the risk of growing again when temperatures could dip too low. Sub zero temperatures could be lethal to the growing tissues, and the plant could have wasted its growth efforts at best, or end up dying, at worst. For this reason, woody perennials are generally rather conservative when it comes to interpreting environmental cues.

Normally, woody perennials have a chilling requirement before they will burst bud. That is, they must experience low enough temperatures for a requisite amount of time before they will respond to warmth and day length to exit dormancy and grow again.

In their laboratory experiments, Caroline Polgar and her colleagues found that many of the 23 species that had earlier leafing-out times in Concord also had a relatively low chilling requirement for exiting dormancy. That is, these plants did not require as low temperatures for as long a time in order to respond to the warmer springtime temperatures. Importantly, Caroline Polgar and her colleagues noted that day length was not a crucial cue for leafing out, but temperature was.

Taken together, the results support the hypothesis that the earlier leafing-out times in today’s Concord are related to warmer temperatures. The 18 day overall changes in leafing out between the 1850s and today translates into new leaves emerging 5 days earlier for each 1°C warming in the springtime temperatures that has occurred over that time period.

On the face of it, an early spring seems wonderful thing. Eighteen more days of green and warmth in place of ice and cold seem like a reasonable trade. After a long winter, chilled by the frigid polar vortex, an earlier spring sounds delightful.

But is an earlier spring really a good thing? spring7

The short answer is probably not.

In Concord, the cost of the earlier winter has been a shift in vegetation. Shrub species that were once under-represented in the mix of flora are increasing in prominence. Those shrub species that have a lower chilling requirement for their spring leafing-out have, unsurprisingly, fared the best. With their lower chilling requirement, the arrival of warmer temperatures earlier in the springtime has been a boon. They are able to flush earlier, to grow more, and thereby to outcompete other neighbouring plants. These species are considered “invasive” in that they are gaining ground in a given ecosystem at the expense of other native resident species.

Aside from the shrub species that are able to outcompete their neighbours, even those that are able to compete and accommodate the new temperature regime are not the same as they were. They will undoubtedly flower and produce seeds and fruit at a different time of the year. The implications of this shift can be profound.

Plants do not reside within ecosystems in isolation. They are primary producers in a complex interwoven web that involves a multitude of other species – microbes, plants and animals alike. Animals, particularly pollinators and herbivores, which reside in the ecosystem are dependent on the timing of plant growth and development to align with their own life cycles, to support their needs and those of their offspring. The misalignment of plant annual cycles with animals’ annual cycles can have dire consequences for the animals concerned.

Many herbivores have fertility and gestation cycles that are timed so that birth of offspring coincides at times when the plants on which they feed are plentiful. For example, both caribou and roe deer have their calves at a time of the year so that herds have the best grazing opportunity – aligned with the greening up of the tundra in the case of caribou, and with leafing out in the forests where roe deer reside. The fertility cycles of both caribou and roe deer are tied to day length. In the case of caribou, migrations to the calving and grazing grounds are also cued by day length. For both roe deer and caribou, the time is synchronous across the herd – all females pretty well calve at the same time. This is important, as it means that any impact felt by one calve is going to be experienced by all calves across the herd. spring8

In a 27-year study of Greenland caribou, Eric Post and Mads Forchhammer found very little change in the time of year when the herds gave birth to calves. The average springtime temperature rose 4oC where the herds roamed. As a consequence, the plants that the caribou ate emerged earlier and were offset from when the herds needed them the most. As a consequence, calve mortality rose and calve production dropped fourfold over the study period.

A similar outcome was observed for roe deer in Europe by Florian Plard and colleagues. As with the caribou range, over a 27-year time period, springtime temperatures rose in the range of the roe deer. Mean springtime temperatures rose by 0.07oC per year.

The higher springtime temperature in the roe deer’s range resulted in earlier green-up in the forests where the deer fed. The deer prefer to feed on shoots and leafy parts of plants before the onset of flowering. This preferred food now emerged earlier in the spring.

Over the same time period, the birth date of the roe deer, which is determined by day length, remained unchanged. As was the case with the caribou, this led to a mismatch in plant availability and herd need. This, in turn, resulted in decreased survival of young.

For both caribou and roe deer the consequences of an early spring is potentially dire. While birth date is under strong selection by the earlier spring, it is not a trait that is strongly inherited.  Consequently, any evolutionary response of birth date to climate change might be slow in coming.

This is decidedly not good for either species. And this is but two examples of many. spring3

Signs of spring are profoundly important indicators of change. Over the shorter time scale, they indicate that there will soon be a dramatic shift from the long, sometimes arduous, winter. In any given year, the signs of spring may be a welcomed indicator of a turn from tough conditions to more hospitable times. Over a longer term, the signs of spring can be, paradoxically, an indicator of a turn for the worse – where some organisms will struggle where they did not have to before. As we embrace the spring after a tough winter, we need to pause and think about the longer term, about what such signs may mean for life in the future, and what we can do to ensure that the signs remain welcome ones, and not omens of dire consequences.

Images: All photographs by Malcolm M. Campbell.


Gaillard JM, Hewison AJM, Klein F, Plard F, Douhard M, Davison R, & Bonenfant C (2013) How does climate change influence demographic processes of widespread species? Lessons from the comparative analysis of contrasted populations of roe deer. Ecology Letters 16: 48-57

Miller-Rushing AJ, & Primack RB (2008) Global warming and flowering times in Thoreau's Concord: a community perspective. Ecology 89: 332-341

Plard F, Gaillard J-M, Coulson T, Hewison AJM, Delorme D, et al. (2014) Mismatch between birth date and vegetation phenology slows the demography of roe deer. PLoS Biol 12(4): e1001828. doi: 10.1371/journal.pbio.1001828

Polgar C, Gallinat A, & Primack RB (2014) Drivers of leaf‐out phenology and their implications for species invasions: insights from Thoreau's Concord. New Phytologist 202: 106–115 doi: 10.1111/nph.12647

Post E, Forchhammer MC (2008) Climate change reduces reproductive success of an Arctic herbivore through trophic mismatch. Philosophical Transactions of the Royal Society of London B Biol Sci 363: 2369–2375. doi: 10.1098/rstb.2007.2207


2 Responses to “First signs of spring – not when they used to be”

  1. Philip Strange Reply | Permalink

    Another example comes from the work of the naturalist and author Richard Fitter and his son Alastair. Richard Fitter kept notebooks recording the flowering times of plants around his home in rural Oxfordshire in the UK for more than 50 years. When his son analysed this information, he found that the flowering patterns of plants in the four decades between 1950 and 1990 were very similar but that in the 1990s flowering advanced by an average of 4.5 days. The findings were published in Science in 2002 with both father and son as authors and the changes in flowering time were taken to be a strong biological signal of climate change.

    • Malcolm Campbell Reply | Permalink

      What a fantastic comment, Philip! The research you describe by Richard and Alastair Fitter is terrific! Your comment was an excellent reminder that I needed to revisit their paper. It's a wonderfully comprehensive piece of research, perfectly described in their very readable paper. It is deserving of a blog post on its own! Something to follow up on...

Leave a Reply

5 − three =