Drought tolerance: grasses of the world

One of the 500 species that are part of the Poa500 project to examine drought tolerance in grasses of the world.

I'm not sure this one is going to work. I believe that if you measure something interesting and have strong contrasts, you should learn something interesting. I've had pretty crazy schemes work out in the past. Hopped on a plane and measured roots on three continents. Had people send me soil from across the US to look at nutrient limitation. Measured foliar 15N for a couple hundred species at Konza. Even looked at spectroscopic assays of 20,000 cow poop samples to infer continental scale patterns of forage quality.

Each time, we learned something interesting by having strong contrasts and measuring something interesting. But to start to understand global patterns of drought tolerance by growing 500 species of grass in the growth chamber in relatively tiny tubes? I'm just not sure this one is going to work. 

Granted, what we're doing right now is just a pilot project and would be easier with the NSF Dimensions of Biodiversity grant funded. But, questions about the evolution and geographic distribution of drought tolerance are just too important not too try. At the heart of it, we just don't understand the traits that are associated with drought tolerance--what does a drought-tolerant plant consistently look like. In what climates are they most likely to be found. Are some lineages more likely to have evolved drought tolerance than others?

To begin to answer the question, USDA sent me seeds for 500 grass species from their seedbanks and I've serially germinated them over the past 2 months. After about a month, we measure a couple of gas exchange and morphological metrics on the leaves and then stop watering. When they stop conducting water (shut their stomata), we measure their water potential, which we call psi-crit.

A little over 100 species have hit their psi-crit so far. Here's probably the most interesting graph so far--drought tolerance (psi-crit) vs. the maximum width of the largest leaf on the plant. 

It seems like you can have narrow leaves on plants that aren't drought tolerant--species like Enneapogon oblongus. You can also have narrow leaves on plants that are drought tolerant--species like Bouteloua repens. You can also have wide leaves on plants that are not drought tolerant--species like Dichanthelium scoparium. But you can't have wide leaves on plants that are drought tolerant. Doesn't exist.

Of course, it doesn't take more than one species to prove something not impossible.

We still have a few species left to measure, of course.

What controls rooting depth?

What controls the maximum height of plants is relatively well understood. Plants can grow no taller than they can support themselves and than they can move water. Roughly every 10m of addition height requires xylem to resist an addition -1MPa of pressure. There are many other factors that could lead to selection for shorter plants, but tall plants need to be able to resist high negative pressures in their xylem.

Though we understand relatively well what controls the maximum height of trees, what controls maximum rooting depth is not well understood. If roots are moving water from their tips to the shoots, then similar constraints should apply to roots as stems. A root cannot go deeper than its xylem can resist the negative pressures of moving water that height. Even if it is belowground, the same physics apply. 

But is there another constraint besides this? Is maximum rooting depth determined largely by hydraulics? If moving water was not a limitation, e.g. in wet places, what would constrain rooting depth? Or lateral extent for that matter.

I suspect that phloem and sucrose transport could be just as big a constraint on rooting depth/extent as hydraulics are aboveground. I'm not sure I understand the details on this, though. If a typical plant tried to produce a root 10m long, could it move enough sugar through its phloem to sustain the growth of the root tip as well as intermediate tissue? What about 100m? Some plants can apparently go this deep, but could any plant? 

I have to admit, I'm not even sure what to begin measuring here. This summer, we're suppose to start doing root cross-sections on long grass roots, e.g. 2 m long, and look at the anatomical characteristics of xylem and phloem. Maybe this'll start to shed some light on what's hidden belowground.

Modification of Classic State Factors

Modified State Factors and Interactive Controls Diagram to account for external supplies of resources.

The general framework to understand ecosystem properties has been the State Factor framework. Created by Hans Jenny to understand differences in soils, it has been applied to ecosystem properties such as plant species composition and stand structure. Essentially, with Jenny's approach, soils were determined by the climate, the organisms present, the landscape relief or topography, parent material and time since a major disturbance.

Terry Chapin and others modified the state factors approach to understanding ecosystems to include interactive controls of ecosystem properties. Interactive factors are not independent of the ecosystem properties, but sit somewhere in between. For example, the macroclimate is independent of what species are present at a site, but the microclimate can be influenced by species composition. These interactive controls include disturbances, microclimate, resource availability, and species composition.

With these state factors and interactive controls, one can have a better framework for understanding how ecosystems are structured and function. For example, the amount of biomass in a forest is not just a function of the state factors, but is influenced by interactive controls such as disturbance and the species that are present in a stand, which is influenced but not determined solely by state factors.

The state factor/interactive control approach is a big improvement in our conceptual framework, but it's generally left out external supplies of resources. Atmospheric CO2 concentrations, dust inputs, and nitrogen deposition can have profound influences on ecosystem properties, but are not adequately included in the current conceptual framework. These factors are somewhat influenced by ecosystem properties. The amount of dry deposition is influenced by canopy structure, for example. Yet, it's probably better to consider these state factors. N deposition plumes are wide and more similar to climate than disturbance regimes.

It's a minor tweak in many respects, but a likely necessary improvement to a long-tenured conceptual framework.

Genetic future of bison

I've spent the last few days in Tulsa at a conference sponsored by the American Bison Society, which is part of the Wildlife Conservation Society. The conference was attended by a mix of scientists, government officials, ranchers, and tribal members. The conference centered around three panels, two of which focused on the genetics of bison. The third panel, which I helped put together, was on the ecology of the bison. We were largely intermission for genetic questions.

A quick bit of history. Bison once ranged in the millions in North America, but at the end of the 1800's had been reduced to about a thousand animals. Some of the remaining bison had been bred with cattle in an attempt to improve the performance of cattle and cattle DNA became part of collective genome of bison. It is not evenly distributed among modern bison--some lineages have more than others, some appear to have none. The presence of cattle DNA in bison has largely consumed discussion on bison for the past decade. How much is there? How is it distributed? Does it have functional significance? Can we get rid of it and repopulate our herds with "pure" bison?

The discussion on the topics were long, nuanced, and technological at times--lots of next generation sequencing and single nucleotide polymorphisms being discussed. ABS will draft official statements, but for me, I think the meeting will be known as a watershed in tolerance and understanding for the modern American Bison. In short, most bison in public herds have some cattle DNA. Quantitatively, less than 1% of the nuclear DNA might be from cattle, but it's there. It possibly could be culled out of the herd, but the bison DNA that we would lose would far outweigh the potential benefit of removing traces of cattle ancestry. Bison also aren't unique. Many of our remaining wild species have DNA from other "species" in them. Wolves, bactrian camels, Przewalski's horses all bear the genetic imprint of domesticated relatives. 

We'll see what the official statements say, but American bison will always be a symbol of America's past. Yet, our bison are also a modern symbol--a bit mixed up, bearing the traces of past pain and hope and ambition, but one that probably should not be atomized any more. In some small way, the conference reflected a modern and sophisticated sense of tolerance.

So, when people visit bison in our parks and preserves, they are likely to be seeing a little bit of Hereford. But, it's a small price to pay if it reminds us to learn more about the animal's history. Much of the bison community seems willing to accept the mark of history and focus anew on continuing to restore  them.

How to write a positive review

Title of a note in ESA's bulletin in 1988

Kendra passed this one on to me. It's a short note published in the Bulletin of the Ecological Society of America in 1988 on how to write an influential review. Rosenzweig, Davis, and Brown put together a short note on writing influential reviews. The structure of the note is important. There are sections entitled "Accentuate the positive" and "The Correlation Between Detail and Negativity in Reviews". The beauty of the note is the encapsulation of the generational shift in our science.

At the risk of sounding curmudgeonly, a lot of emphasis in the modern system has been on timeliness. In the past, reviews were more a chance to help colleagues than potential competitors.

The note is worth anyone reading. I'll abstract one part:

"...when scientists are under attack, they circle round, wagon-train style. The physicists aim outward at their opponents. Biologists, on the other hand, aim inward, at each other. Their weapons, of course, are disparaging reviews and negative comments.

The earth pulses with fascinating ecological and evolutionary questions, and threatens with environmental concerns. The questions are as intellectually challenging as those facing any other scientific discipline. The answers are essential to deal with the environmental prob lems that beset the world. But we cannot con vince other scientists (or the public, or government officials) of the importance of our work if we seem to be calling each other incompetents."

For me, the older generation of ecologists were exceedingly civil to one another. It's good to remember that our research stands on the shoulders of our predecessors, not the backs of our competitors.


You can find the paper here: http://www.jstor.org/stable/20167054

Independence of leaf and stem traits

Multivariate relationships between leaf and stem traits for 600+ Neotropical tree species. 

Before reading Baraloto et al. in Ecology Letters, I think the best assumption for how leaves and stems correspond is that plants with high-activity leaves (low tissue density, high N concentrations, low leaf longevity) would be associated with low density wood. Pioneer species are typically thought of this way--think Cecropia. Essentially cheap leaves come with cheap stems. Late-successional species typically have low-activity leaves and high wood density.


Yet, if you think about it a bit more, pines and firs have low wood density, yet their leaves live a long time. So what is the pattern? Do leaf and stem traits correlate or are they independent.


Baraloto et al. compared key leaf and stem functional traits for over 600 tropical tree species. The authors convincingly show that leaf and stem economic axes are orthogonal. Species with low leaf tissue density and high foliar nitrogen concentrations are equally as likely to be associated with high stem density as low. In short, cheap leaves can be born on expensive stems. 


The factorial ecology of leaf and stem economics has still to be worked out and the obvious next question is to reexamine patterns with roots, but the paper is an excellent example of the power of sampling large numbers of species and distilling data to a clear, simple message.



Baraloto, C., C. E. Timothy Paine, L. Poorter, J. Beauchene, D. Bonal, A. M. Domenach, B. Herault, S. Patino, J. C. Roggy, and J. Chave. 2010. Decoupled leaf and stem economics in rain forest trees. Ecology Letters 13:1338-1347.

Comparing nutrient availability with traitscapes

An overlay of foliar N concentrations and nitrogen isotop ratios from the Konza flora (black) to a global dataset (gray).

One of the key questions for understanding plant community assembly is to understand the environments that species inhabit--not just the dominant species, but the hundreds of species that are only occasionally or rarely seen to the casual observer. Do the rare species mirror the more abundant species in their traits? Or are they rare because they are built for different environments?

At Konza Prairie, there are over 500 herbaceous species. Over last 2 years we measured the leaves of over 400 species at Konza. One of the interesting patterns was examining the relationships between the leaf N concentrations and the foliar N isotopes. Together these two best reflect the N availability of the environment the plant inhabits. High N concentrations and high del15N generally mean that the plant is growing in an environment with high N availability.

Konza is considered a strongly N-limited ecosystem. The responses of aboveground productivity to N addition are some of the highest in North American grasslands--ANPP triples with N addition. Given this, at Konza, one thing that was surprising was how many species had really high N concentrations in their leaves. A fair number of species that didn't fix nitrogen had N concentrations over 40 mg g-1, or 4%. That's really high.

When you look at the flora as a whole, there were a lot of species that were found in high N availability sites. Edges of roads. Bison wallows. Places with high dung inputs. A lot of the diversity of Konza is likely maintained because of these high N availability sites.

When you look at Konza species by species the picture changes from one dominated by severe N limitation to one with a broad spectrum of N availability.  In fact, we could compare Konza to the rest of the world with a global dataset on foliar N and N isotopes. Not only do many of the species occupy high N availability sites at Konza, but the typical species at Konza actually occupies areas of higher N availability than the "rest of the world".

The analysis of traits across a broad portion of a flora--the community's traitscape--is not novel, but definitely an undersubscribed approach. As we build more global datasets and measure more and more species, a lot more insight to how communities are constructed and florae assembled will come into new light.