Seminar schedule

Ceballos and Ehrlich (2009) Discoveries of new mammal species .... PNAS 106:3841–3846

Hank's "Harper's Index" of Mammal Discoveries

• Number of categories of discovering new species: 3 (completely new finds (morphologically distinct), discovery that a well-known organism was actually > 1 species, and third, the elevation of subspecies to species. These last two are very similar, but the authors do not even address the > 600 cases of the third.)
• Number of new mammals found since 1993: 408
• Number of missing spellings of limestone forms: 1 (don't put the karst before the horse).
• Percentage of the land surface exploited, for crops, rangeland, building, and other: 70%
• Magnitude of the underestimate of unnoticed extinctions: gross (could we use range size to model this and actually quanitfy it?).
  
• Number of actual lemur species once thought to be only two species: 13
• Average range of previously known land mammals: 400,000 sq.km
• Average range of newly discovered land mammals: 84,000 sq.km
• Percentage of cells (cell=10,000sq.km) with rare species with low human population densities: 46%
• Percentage of cells (cell=10,000sq.km) with rare species with "relatively high" human population densities: >20%
• Number of commentators suggesting that the discovery of new species is a problem for conservation: 3
• Number of authors asserting that the discovery of new species is a not problem for conservation: 4

Sinclair, T. R. (2009) Taking the measure of biofuel limits. American Scientist 97:400-407.

I am enjoying greatly Sinclair's concise treatment of basic plant physiology, biochemistry, and the physical environment in which C3 and C4 crops are grown. It is the height of back-of-the-envelope artistry and clear thinking, which are hallmarks of strong quantitative, empirical biologists.

Sinclair starts with the loaming problem: the US Energy Independence and Security Act (currently) mandates that by 2022, the US should be producing 144 billion barrels of ethanol, roughly 25% percent or one barrel of ethanol for every three barrels of gasoline/diesel. This is the daunting task - it is a shit-load (my word, not his) of ethanol. Sinclair then asks whether the physical limits to plant growth will allow this mandate to be met by growing plants.

Having set up the problem, he goes about describing the elements of the puzzle:

Total annual ethanol production =
g Sugar / MJ of light intercepted by the canopy per day (C3 vs. C4) X
MJ incident light / sq. m. (max vs. average) X
days in the growing season X
grain vs. whole plant harvest X
gal ethanol / tonnes feedstock (corn vs. stalk) X
water use efficiency (C3 vs. C4 in dry vs. humid env.) X
Leaf area / land area (LAI) X
LAI / g nitrogen in tissue (C3 vs. C4) X
g N available in soil

I don't think that the above is a perfect rendering of Sinclair's elucidation, but it is close enough for now.

Sinclair next goes on to describes the sustainability of biomass harvest, in terms of N flux and the pool in the soil. He points out that the

rate of annual change in available soil N (g) =
annual application - harvest - runoff - leaching
+ production(cyanobacteria, thunderstorms)
+ mineralization(dead biomass)
+ N sequestration (perennials only)

Last, he considers land available, pointing out that most fertile land in humid regions is already in production. He concludes cautiously with

"... realistic assessments of the production challenges and costs ahead impose major limits."

This approach is a great companion to the work of Searchinger et al., Fargione et al., and Tilman et al. that have focused on land use and biodiversity issues. Much work lies ahead, and Sinclair has been of great help to me.

PNAS - "Science for managing ecosystem services..."

A pile of folks (Carpenter et al.) provided a blueprint, or rather an eight page precis of a blueprint, for what ecologists and their collaborators should be doing now to help humankind (Carpenter et al., 2009, PNAS 106:1305-1312). I find the whole thing rather overwhelming, but I must be strong, and take heart.

First, some of the overwhelming bits. It seems as though we are supposed to know and understand everything about the current state of the natural environment and its processes, in every location, over time, current social (cultural, political, and economic) institutions, policies and practices, AND how they all interact, so that we can predict unpredictable future events. "Oh," I said. "Is that all?"

Every question seems to spin out of control with a huge number of factors and feedback loops that need to be included.

It sounds as though a "School of Sustainability" would include the business school, college of arts and sciences, the school of architecture, the med school ... . This will be like the IPCC committees, on steroids.

OK, so what can I do? How do we move forward? Carpenter et al. were kind enough to suggest a few research areas, including (i) the analysis of biodiversity in a social-ecological context, (ii) match quantitative models to conceptual goals, and (iii) figure out how to predict the unpredictable ... oops, scratch that last one -- I mean "address nonlinear and abrupt changes," and (iv) expand the quantification, understanding and communication of uncertainty. I must admit, these feel helpful because they have the appearance of being tractable, and happen to interest me.

Another aspect of this report that I cling to is "place-based" research. This seems to suggest the assumption (or hypothesis) that spatial variation in social and environmental drivers will require local assessment, testing, evaluation, etc., of any science or policy. Thus, we should be able to argue, for instance that a set of feedback loops operate in Ghana, Peru, and Sweden, but Ohio is different for reasons A, B, and C, and so we need to test whether these feedback loops operate here in Ohio, USA. Perhaps I am scared or lazy, but I hope that experiments replicated in place and time are valued by funding agencies. They should be, but sometimes novelty seems more important than utility.

It seems essential, and a great opportunity, to "Learn from existing management programs." I think that current efforts of monitoring and evaluation of past and current practices are probably woefully inadequate. It may be quite productive to simply ask agencies and programs how we can help. How can we bring our expertise to bear on doing jobs that are already identified as important.

Tscharntke et al. 2005

Tscharntke and colleagues provide a nice framework for understanding the upsides and downsides of agricultural intensification for biodiversity, ecosystem services and their interaction. The framework combines the functioning of the local ecosystems, their spatial arrangement, and the consequences of the arrangement. They cover a lot of ground, and should raise lots of questions. Post your questions (comments) here.

Testosterone Levels in Dominant Sociable Males Are Lower than in Solitary Roamers

As a pet owner, and scientist, I often shake my head at how much credit we give our species. "Oh, we're just SOOO complicated and special --- not like those 'animals'!"

Our program recently had the great fortune of a visit by Carsten Schradin, who gave a nice seminar on the sociobiology of the social striped mouse (Rhabdomys
pumilio). Many of the findings he discussed reminded me nothing so much as stereotypes of our own species. As an outsider looking in, I find the parallels between non-human and human behavior are wonderfully ironic.

I don't anthropomorphize non-human behavior. Rather, I prefer to think that I re-animate human behavior.

[the title of this blog comes from Schradin et al. 2009, Am Nat, v. 173).]

Quantitative training in EEEB and R

To train our EEEB students (grad students in Ecology, Evolution and Environmental Biology) in quantitative methods, I have been putting a lot of effort into teaching the R language to willing and sometimes unwilling students.

This coming Spring (2010), I will lead a 1 credit seminar here at Miami using Ben Bolker's recent book, "Ecological Models and Data in R" (Bolker, 2008). This is a fabulous tome, by a gentle and insightful teacher. It fits into the general mood in the field of academic ecology, that measuring real quantities (estimation) is very important (as opposed to only hypothesis testing). Ben's book captures this perfectly, and further, shows how those real quantities are often the parameters in simple models of populations and ecosystems. As Bolker states, "The idea behind realistic static models is that they link together simple deterministic and stochastic models of each process in a chain of ecological processes..." [italics mine].

I anticipate that the book will be very well received in the seminar, because it is practical and clearly written, and sufficiently comprehensive to, as Bolker states, "...pose, and answer, ecological questions in a quantitative way." Thus, I anticipate that his book will be helping us to design and analyze experiments, and ultimately publish papers and finish dissertations.