Phenology and sensitivity to climate II

Update on results.  Had to learn how to use R to the data tables. Somewhere on the order of a quarter million calculations were going into sorting through all the combinations of climate, years, species...I wasn't doing that in excel by hand.

After looking at ~50 species, species that flower earlier have lower minimum temperatures required for flowering. Not too surprising. It's colder in March than May.

More interesting is that species that flowered earlier were more sensitive to climate variability--required fewer days at threshold temperature to flower.

In short, it seems like climate variability (absent of changes in means) early in the season will have a greater consequence for phenology than later in the season.

In operational terms, a warm spell followed by a cold snap early in the season would disproportionately cause greater damage to plant reproduction than later in the season. 

Phenology and sensitivity to climate

One of the strongest separations of species in grasslands is their phenology. Most guide books separate grasses and forbs, forbs by flowering color, and then timing of flowering. From a global change perspective, phenologies become important in understanding how climate will alter ecosystem function. Early-flowering species appear to respond more to variation in climate than later-flowering species.

The analyses of these patterns are still pretty basic. One question that struck me is whether early-flowering species are more responsive to variation in climate, or just have a lower temperature threshold for responding.

The first step in partitioning this is to begin to quantify these patterns and compare. On a preliminary basis, I used the Gates' first flowering data that was collected in the 1930's-1950's. I then adapted the critical climate period approach to examine the climate correlates with phenology. In short the technique allows testing whether temperatures over different windows before the event each year are the best predictors across years of the timing of the event. For example, I could test whether the first flowering date is best predicted by temperatures 10 days preceding flowering, 15 days, 20 days...etc.

From a predictive standpoint, the approach sees to work pretty well. For example, Catalpa flowering was best predicted by a 40-d window of temperature that averaged 21.9°C. 10-days after this window, it flowered. Outside of one year, flowering for catalpa can be predicted pretty well. 

When I do this for a handful of species (grasses, forbs, and trees), a couple of patterns emerge.

First, species that flower later in the growing season (day of year = DOY) have higher temperature requirements that must be met. Early-season species need periods with daily maximum temperatures to average 13°C, while later species (like Catalpa) require periods over 20°C. [red dots are trees].

Second, species that flower earlier integrate over similar periods of time as late-flowering species. In general, about 45 days.

In general, I think the relative critical climate period technique holds potential for quantifying differences in climate sensitivity for phenological events. The interesting part of this work lies in relating these patterns back to the ecology of the species more than anything. For example, what is it about a species that lets it respond to climate so fast? I would guess species like dandelions (Taraxacum) that sit in rosettes have few developmental barriers to flowering. Primordia are there waiting. Other, more determinate species, need to produce a series of leaves before they initiate flowering.

Experimental warming and phenology

Phenology is the study of the timing of biological events. The phenology of organisms is a critical component of the functioning of our biosphere. Migratory animals time their movements to secure food resources. Plants time their leaf production to minimize exposure to harsh conditions.

No one would state that the everything in the natural world is perfectly optimized, but changes in climate are likely to alter then phenology of organisms in ways that the consequences are not understood. In response, the scientific community has monitored phenology for a long time--in some places hundreds of years--and conducted experiments to aid in forecasting future responses.

Published recently, Wolkovich et al. led a team that compared observed and experimental consequences of warming for phenology. Their conclusion was that experiments consistently underestimated the phenological consequences of warming.

Two commentaries were published along with the paper. Both sets of authors presented the case in defense of experiments that can be summarized as "nature is complicated".

Neither commentary undercuts the value of the work, but understanding the consequences of different types of warming and why phenology might not respond consistently to experimental warming underpins basic questions we have about the interactions between climate and the biotic world.

Wolkovich, E. M., B. I. Cook, J. M. Allen, T. M. Crimmins, J. L. Betancourt, S. E. Travers, S. Pau, J. Regetz, T. J. Davies, N. J. Kraft, et al. 2012. Warming experiments underpredict plant phenological responses to climate change. Nature 485:494-497.

this spring was warm

Not news to most at this point, but March was warm. At Konza, our first flowering phenologies were running 30 days ahead. Upper Midwest was even more extreme in temperatures.

I'm writing this just to post the maps...

https://www2.ucar.edu/atmosnews/opinion/6760/great-warm-wave-top-10-list

This NASA depiction of land-surface temperature anomalies between March 8 and 15 shows the effects of the Great Warm Wave kicking off across the Great Plains of the United States and central Canada. In some areas, the land surface was more than 18°F (10°C) warmer than average. (Image courtesy NASA Earth Observatory.)

Another map for March:

This is for March 2012, departure from normal (1980-2011).

Grazers in a warmer world

If the world gets warmer, what happens to grazers?

Not an easy question. There are many grassland climate change experiments, but these are of limited utility here. Grazed and ungrazed grasslands are starkly different such that the consequences of warming for ungrazed grasslands are unlikely to apply to grazed grasslands.

If experiments don't help, then we need to look at how grazers respond to short-term variability in climate and compare that with geographic patterns that might represent long-term patterns.

When I've looked at how bison respond to inter-annual variation in climate, hot years don't affect them.

Yet, when we look across temperature gradients, hot places have small bison. Across 22 herds and a quarter million weights of bison, it's clear that herds in hotter places have smaller animals. Sometimes up to 500 lbs lighter.

Why the difference between short- and long-term patterns?

This is where experiments come in handy. When exposed to elevated temperatures short-term, grasslands begin to lose nitrogen. Over the long-term, these losses accumulate which drives down the quality of grass for grazers to eat.

One hot year, no problem. Many hot years and grazers don't grow as big.

The differences among bison are dramatic.

Back of the envelope calculation shows that just a 1°C increase in temperature across the US could cost the cattle industry $1 billion. Considering projections are for multiple degree C increases, those costs would accumulate.

[Regarding details, I'm about to submit this paper. We'll see how it's met.]

How to be laconic

When Philip II of Macedon threatened to invade Sparta, he sent a message "You are advised to submit without further delay, for if I bring my army into your land, I will destroy your farms, slay your people, and raze your city."

Sparta's reply was just one word "If". 

Philip chose not invade Sparta.

Laconia was the region surrounding Sparta, and both names came to symbolize a bareness: laconic and spartan.

There are times in scientific arguments, be it a reviewing a paper, responding to criticism, or summarizing intellectual arguments, where it is critical to be laconic. The Spartans could have responded point-by-point. Instead they chose to reduce the argument to its essence, "If".

In science, long, negative reviews often mask vested interests or laziness of thought. The best antidote is brevity. 

Global Nitrogen Budgets

It's been said that the global nitrogen budget is currently at the stage that the carbon budget was 50 years ago. Some pools and fluxes are well constrained. Others not so well constrained.

As we continue to examine the global carbon and energy budgets, it becomes even more important to make sure the N cycle is nailed down. For example, N2O has the 300x the warming potential of CO2. And N availability is a key for understanding productivity and ecosystem C sequestration. It fuels productivity, can suppress decomposition, and influences water quality.

One of the least constrained parts of the global N budget is the terrestrial biosphere sink--how much N gets stored on land each year.

Schlesinger (2009) puts the biospheric increment of anthopogenically fixed N at 9 Tg N. His estimate is derived from the amount of N being deposited on land (46 Tg/y) * fraction deposited on forests (0.39) * fraction of deposited N stored in forested ecosystems (0.5) = 9 Tg. That's a pretty rough calculation. By his first-cut estimates, about the fate of 50 Tg y-1 is unknown.

Compare that number with a top-down estimate. Le Quéré estimate that the land sink for C averages 2.6 Pg C y-1 (1990-2000) and were as high as 4.7 Pg C y-1 in 2008.

The typical C:N of soil organic matter is 12, but can be as high as 20. If so, 4.7 Pg C would require 390 Tg of N. C:N of 150 (wood) would require 30 Tg N.

So how much N is currently accumulating in ecosystems today? 10 Tg N? 100? 400?

It's an incredibly unconstrained number, but probably one of the most pivotal for understanding how much C ecosystems. How it will become more constrained is not clear at this point. We aren't committed to the types of measurements required to narrow this number.

Until it does, a major part of the Earth's biotic metabolism goes unknown.