Scale It Down

When it’s cold outside and you snuggle up with a blanket and hot tea, are you more affected by the cold in the distance or by your immediate cozy surroundings? Plants experience similar variation in the conditions around them. For example, a plant that lives in the mountains will experience cold, harsh winter environments. However, plants may also be protected from the cold by nearby trees or a hillside. These objects are the plant equivalent of a blanket.

Climate conditions, like long-term patterns of temperature and rainfall, can impact plants at multiple different scales. Plants can be affected by the conditions immediately surrounding them (think on the scale of centimeters). These conditions are called the plant’s “microclimate”. For example, the amount of heat a single plant experiences over a growing season will affect how well the plant can grow. Climate conditions can also affect plants by changing their local neighborhoods (think a couple of meters). For example, warmer winters and longer growing seasons might also help the pests and invasive species that harm the plants. Additionally, climate at really broad spatial scales (think thousands of miles!) also matters. Species that are adapted to certain climate conditions might gradually be moving north or to elevations where conditions are cooler.

Finally, increases in temperature often speed up, or advance, life cycle events. In this case, a plant grows and flowers earlier in the year. Scientists refer to these changes in the timing of events as phenological shifts. One way to measure these phenological shifts is to measure the number of days that a life cycle event, like flowering, advances per one degree change in temperature. 

How do researchers measure how plants respond to climate change? 

Measuring phenological shifts in response to climate change involves a variety of approaches. Many studies examine how plants respond to the environment by revisiting the same population of plants year after year. Some studies manipulate the environment itself. They do this by warming the area around a plant using special heaters or heavy plastic. Finally, many studies use what are known as herbarium specimens. 

People have been collecting and drying plants for centuries. Each pressed plant is associated with a specific location and date in time (Figure 1). These plant pressings, or specimens, are stored in a plant library, or herbarium, that can be accessed for research. Because of this, herbarium specimens are important sources of historical information in plant science. We can use them to examine how plants have responded to temperature changes over the past couple centuries. 

But where do scientists get all this temperature data from? Nowadays, scientists use temperature loggers. Temperature loggers are portable instruments that scientists place next to a plant. They record the temperature near the plant over a long period of time (Figure 2). For herbarium datasets that potentially span centuries, scientists have access to longer-term records from nearby weather stations.

What did we want to find out in our research? 

Our team tried to figure out at what scale temperature might matter for the flowering time of a plant commonly found in the Rocky Mountains of North America. The Rockies are experiencing extremely rapid rates of climate change. We hypothesized that flowering time might be advancing with these quickly warming temperatures. Flowering time is the day of the year on which an individual plant produces its first flower. 

How did we conduct our research? 

We studied Silene acaulis, or moss campion (Figure 2), at Niwot Ridge, Colorado, USA. Moss campion is found in alpine habitats across the world. Alpine habitats do not contain trees due to their high elevation. They are experiencing rapidly warming temperatures. Moss campion’s habitat makes it an ideal plant species to look for effects of climate change. For six consecutive summers, we revisited the same four plant populations every other day. Each population has 150-200 individual plants. We recorded the day of the year on which each plant produced their first flower. This was our “observational dataset.”

Using our observational dataset, we first tested how warming temperatures affect flowering time. We measured temperature as growing degree days (GDD). This is a unique measurement used in plant ecology to estimate the amount of heat a plant needs before it can grow or produce flowers. Imagine how many cups of cocoa would warm you up after coming inside from a snowy day. One? Two? Maybe three! Just as more cups of cocoa warm you up slowly, each GDD a plant collects brings it closer to flowering. GDD increases as temperatures rise.

We then tested how the scale of our temperature data might affect our estimate of temperature’s effect on flowering time. We measured temperature in the field at two different scales. First, we placed temperature loggers at each of the four populations of moss campion to measure microclimate conditions. Second, we got temperature data from the local Niwot Ridge weather station. We expected that measuring temperature at the microclimate scale would allow us to detect a greater change in flowering time (more days advanced) in response to warming temperatures relative to using weather stations. We expected this because temperature at the microclimate scale, or right next to the plant, should have a more direct effect on plant reproduction.

We found that plants flowered about 0.10 days earlier per increase in GDD. This change in flowering time is important. As flowering occurs earlier and earlier in the spring, plants will experience conditions that they’ve never experienced before. They might experience frost events that kill flowers before the flowers can produce seeds. They might also flower too early for insects and other critters to collect and spread the plant’s pollen. This would mean that the plants cannot reproduce.  

Interestingly, estimates of flowering time change were similar whether we used microclimate data (0.098 days earlier) from the temperature loggers or temperature data from the local Niwot Ridge weather station (0.102 days earlier). This similarity is promising for climate change and conservation researchers. Temperature loggers can be expensive. Our research suggests that free public data from nearby weather stations is suitable for measuring flowering time responses to warming temperatures.

But were the plants only responding to temperatures in that six-year period? Or did warming over a longer time frame matter for changes in flowering time? What if populations farther north in the Rockies were responding completely differently? To answer these questions, we next looked at more than a thousand herbarium specimens from across the Rockies. If the specimen had a flower on it, we recorded the date the specimen was collected as the plant’s flowering time. Our “herbarium dataset” allowed us to measure changes in flowering time over more than a century and across an entire mountain range.

We found that plants flowered 0.01 days earlier per increase in GDD in the herbarium dataset. This was a much smaller change than estimated from the observational dataset! This is important because the type of data we used to look at the effects of rising temperatures affected our results. The observational dataset predicted a 10x greater shift in flowering time. 

There are several reasons why we would detect a smaller advance in flowering time in the herbarium dataset than in our observational dataset. These reasons include greater variation and biases. First, there is much more variation in flowering time in the herbarium dataset than in the observational dataset. Herbarium specimens span a larger amount of space. There are specimens from across the Rockies in the United States and Canada. In contrast, the observational dataset was just from Niwot Ridge. Additionally, the herbarium dataset spans a larger time period. The specimens were collected from 1872-2021 (almost 150 years!). The observational dataset was collected from 2016-2021. Lastly, many different people collected the herbarium specimens in their own way. The observational dataset was collected by a trained team of researchers.

Second, herbarium specimens are often biased in four ways. Herbarium specimens are a snapshot. A collector pressed a flowering plant, so we know the flower was open on the date recorded. However, the flower could have been newly opened or very close to wilting. Herbarium specimens are also biased in space. This is because plants are collected mostly in accessible locations. Additionally, herbarium specimens are biased in time because plant collection efforts have declined since the mid-20th century. Finally, the fact that the specimens came from a wide variety of places and years also means that the temperature data was highly variable. Some places and years simply did not have data available. 

The mismatch in our results between the observational and herbarium datasets is important. It demonstrates that using different data sources can produce very different estimates of how climate change impacts flowering time. Scientists should bear this in mind when deciding what data to use and how to interpret it. 

What did we conclude? 

So at what scale does weather matter to a plant? We found that moss campion shifts its flowering time in response to warming temperatures across the Rockies. Moss campion also shifts its flowering time in response to conditions they experience immediately around them. These conditions include warm microsites that act as cozy blankets for the plants. However, plants demonstrated stronger changes in their flowering time in response to local rather than broad conditions. Future work needs to examine whether this pattern holds in other plant species. We are also interested in whether these patterns would hold if we collected observational data from populations across the entire range of the Rocky Mountains. 

The scale at which plants respond to the environment matters for how strong we estimate a plant’s response to be. This work will help improve scientists’ predictions about plant responses to ongoing climate change. It also highlights the importance of keeping good records! Both observational datasets and herbarium specimens become part of the long-term record that scientists can use to protect species in the future. Conserving plant species like moss campion is critical for maintaining biodiversity, especially in alpine habitats affected by climate change.

Glossary

Phenological shift: Changes in the timing of an event (for example, a flower starts blooming earlier)

Herbarium: Collection of plant samples that have been dried, pressed, and preserved for long-term study

Flowering: Production of flowers

Flowering time: Timing of when a plant produces flowers 

Growing degree days (GDD): A unique measurement in plant ecology that estimates the amount of heat a plant needs before it can grow or produce flowers. 

Written By: Meredith Zettlemoyer

Academic Editor: Biologist
Non-Academic Editor: A Grandfather

Original Paper

• Title: Estimating phenological sensitivity in contemporary vs. historical datasets: effects of climate resolution and spatial scale. 

• Authors: Meredith A. Zettlemoyer, Jill E. Wilson, Megan L. DeMarche

• Journal: American Journal of Botany 

• Date Published: December 2022

Please remember that research is done by humans and is always changing. A discovery one day could be proven incorrect the next day. It is important to continue to stay informed and keep up with the latest research. We do our best to present current work in an objective and accurate way, but we know that we might make mistakes. If you feel something has been presented incorrectly or inappropriately, please contact us through our website.