Many measurements and few observations

Our University colleague Professor Sir David Spiegelhalter has written a brief opinion piece in the latest issue of Science on the future of probabilistic models, particularly for big datasets (think images or genomes).

Two points jumped out at me:

(1) Statistical problems have shifted from many observations (large n) and few parameters (small p) to small n and large p, creating pitfalls when testing large numbers of hypotheses.  This is because the standard “p-value”, which we’ve griped about in the past here, will declare 1 in 20 non-existent relationships “significant” simply by chance.  So procedures are needed to reduce false discoveries.  The bit that I didn’t really follow was why even bother minimizing false discoveries?  Wouldn’t an interpretation of effect sizes be more meaningful?

(2) Inferring causation from observational data will continue to be a challenge, especially when n gets cheap and p remains large.  Statistical theory to deal with causality will be needed more than ever, and thankfully, it is improving.  This is something we’re quite fond of having thought a fair bit about causality in the context of path analysis, structural equation modelling, and directed acyclic graphs (see our J Appl Ecol paper that just came out).  The problem, however, is that these approaches don’t come easy and I struggle to see how they can be used by non-statisticians (the models in our paper took years of faffing!).  Finding ways to make causal inference more accessible is going to be critical in the future.


NERC Responsive mode funding through time

In today’s group meeting, one of the topics we discussed was the impeding changes to NERC’s strategic research funding.  Peter Grubb wondered whether there had been an increased allocation of funding towards these schemes as opposed to the discovery science “responsive” mode stream since his time on the Terrestrial Ecology funding committee in the 1980s.

NERC does make funding data publically available here:  But, there doesn’t seem to be a general header for the strategic programmes – they’re all listed individually.  Instead, I’ve looked at changes in Responsive Mode funding.  In the past, I’ve seen people focus on how rejection rates associated with a particular scheme change over time rather than total expenditure (but see here for stats from New Zealand’s premiere funding stream).


The plot above shows the annual investment into Responsive Mode funding based on the starting dates of awards.  Both the annual investment and number of awards have been steadily increasing, aside from a few blips.

Perhaps, key however, is that investment has generally been keeping up with inflation.  The black line above the “Total spent” trend is the amount that should be invested to keep up with inflation, calculated as the mean of the consumer price index over the previous year.  Ideally, the red line should stay above or very close to the black line.

The take home message is that the UK government does a good job at making sure that funding for basic blue-skies science keeps up with the economy – at least based on my crude numbers.  Whether it represents a reasonable proportion of total government spending is obviously another question…

Which came first?

Modern-day angiosperms occupy just about every corner of our planet. But things were not always this way; early flowering plants are thought to have evolved as woody understorey species growing in warm, wet environments. So how did angiosperms come to be where they are today? More specifically, how did they adapt to living in areas which are prone to freezing?

This is the question that Amy Zanne and colleagues set out to answer in their latest paper:

Zanne et al. (2013) Three keys to the radiation of angiosperms into freezing environments. Nature, doi: 10.1038/nature12872.

Before telling you what I thought of the paper, I’m first going to briefly summarize it (hopefully by then I will have come up with something clever to say…).

Freezing poses a problem to vascular plants because during freeze-thaw cycles air bubbles are formed inside the xylem and these can jam hydraulic pathways. This process, referred to as freezing induced embolism, occurs much more frequently in species that have large conduits. The authors suggest that there are three main ways through which plants can solve the freezing problem: they can (1) develop smaller diameter vessels, (2) interrupt hydraulic transport during freezing periods by becoming deciduous, and (3) become herbaceous. The question is, did plants develop these adaptations in response to the new environmental conditions into which they were venturing, or had these traits evolved previously? Which came first, the trait or the evolutionary pressure?

To answer this question, Zanne and colleagues put together a massive dataset in which they characterize thousands of flowering plants species according to whether or not they are exposed to freezing in their range, deciduous vs. evergreen, large vs. small conduits and on the basis of their growth form (woody vs. herbaceous). They combine this with the newest, most complete molecular phylogeny of flowering plants (you have to check it out!), and set out to test how plants have shifted among these various trait states as they expose themselves to the cold.

As you might expect, angiosperms that are exposed to freezing in their range are either herbaceous, or have remained woody by developing small conduits and/or dropping their leaves. Plants lineages shift relatively frequently between freezing and non-freezing environments. When they do so, they tend to shift from evergreen to deciduous (i.e., deciduousness evolves as a response to temperatures dipping below zero). In contrast, although plants that have moved into colder regions tend to be either herbaceous or have small vessels, both of these traits seem to be pre-requisites to successful colonization (i.e., plants had already evolved as herbaceous or to have small conduits before encountering freezing temperatures). All of this is summarized beautifully in figures 2 and 3 of the paper.

I’ve been waiting to read this paper for the last couple of months (since I saw Amy present the work at INTECOL in London). It’s a really neat study; a smart way of taking full advantage of a global dataset to answer fundamental questions that bridge the gap between ecology and evolution. Most importantly, it makes you think. Here a few questions to which I don’t know the answer:

1)      If small conduits and herbaceous life style did not evolve as a way of coping with freezing, what evolutionary pressure is behind these traits? Drought seems a likely candidate to me (see this paper by Choat et al. which Howard brought up in one of previous our meetings). Many of the adaptations needed for plants to resist freezing (small vessels, herbaceous and even deciduous) are the same that allow plants to cope with high soil water deficits. In order for angiosperms to colonize higher latitudes, did they first have to contend with arid environments?

2)      I was surprised to see how few extant woody species have large conduits, regardless of whether or not they are exposed to freezing (see figure 2 in the paper). The authors define large vessels as anything above 0.044mm in diameter, as this is the threshold “above which freezing-induced embolisms are believed to become frequent at modest tensions”. But of the 860 species for which vessel diameter data was available, only 2% fall within the large conduit category. Plants are much more likely to transition from large to small vessels than vice versa (regardless of freezing/non-freezing). Why is this? Would changing the vessel diameter threshold affect these results?

What do you think?


Standardizing radial growth

As I’ve been enjoying the tropical weather here in Sudbury over the past week

Everyone’s gone home for winter

I’ve been thinking about how ecologists report radial tree growth.  And I’m not sure we’re doing it right…

During my PhD, one of the papers I read often was:

Bee, J. N., Kunstler G. & Coomes, D.A. (2007) Resistance and resilience of New Zealand tree species to browsing. Journal of Ecology 95, 1014–1026.

Table 5 specifically is worth drawing attention to:


It looks like trees in north temperate regions grow really fast – more so even than in the tropics (i.e. Panama).  But what happens in winter?  In Sudbury, temperatures have been awfully cold in December and they’re likely to hover around similar levels until March.

Sudbury minimum air temperatures (brrrrr)

Minimum air temperatures in Sudbury, ON — brrrrr……

Surely, there isn’t much growth in trees during this period.  So does it make any sense to report diameter growth from January to December?  Wouldn’t it be better to report growth relative to some standardized measure of growing season, such as the number of days when air temperatures are >6°C?

In the case of global comparisons, such as in the Bee et al. table, this might be a moot point.  North American trees still win, just by more.  But standardization is likely to be a real issue for studies that use latitudinal  gradients as a space-for-time substitution to test the potential effects of climate warming.  A few examples are:

Silva LCR, Anand M, Leithead MD (2010) Recent widespread tree growth decline despite increasing atmospheric CO2. PLoS ONE 5(7): e11543.
Huang, J., J. C. Tardif, Y. Bergeron, B. Denneler, F. Berninger, and M. P. Girardin. (2010) Radial growth response of four dominant boreal tree species to climate along a latitudinal gradient in the eastern Canadian boreal forest. Global Change Biology 16:711–731.
Lloyd AH, Bunn AG and Berner L (2011) A latitudinal gradient in tree growth response to climate warming in the Siberian taiga Global Change Biology 17: 1935–45.

Focusing on Fig. 1 in Silva et al. 2010, red maple (Acer rubrum) at 47°N seems to be growing slower than at 52°N.  But temperature, and hence growing season, differs between these two sites.  I wonder whether trees would grow at similar rates if annual basal increments were reported relative to the length of the growing season?…  Ultimately, what ecologists are regularly reporting is an “absolute” outcome – the product of growth and length of the growing season – rather than the “true” rate of growth.

Leave a comment and let me know what you think!


Climate Change: the Buried Agenda

Currently, the developed world seems content to accept that financial stringency is a price worth paying for bankers’ excesses. Meanwhile, anyone advocating strategies to meet the cost of mitigation and adaptation climate change are close to being vilified as scaremongering extremists. Yet, it seems that multinational corporations by and large accept the prognoses of the Stern report (, and are examining strategies to protect their resource inputs and supply chains for the future (

Of course, the problem is one of timing: hang around for 100 million years, and the current shift in CO2 and GHG emissions will appear as a mere blip in the geological record; but hanging in for 5 years, over the standard political renewal period, fuels current expectations that living standards, economic growth, and availability of mass transportation will progressively increase.

Another problem is one of resource availability, and ease of access. Fossil fuel resources have powered the astonishing technological progression in the past two centuries, using buried plant reserves which have allowed us to trade on past sunlight. It had been initially simple to mine or pump these reserves, but then increasingly difficult to exploit resources in the North Sea, Arctic and Gulf of Mexico. Now we have the prospect, or perhaps spectre, that extraordinary rendition of shale gas threatens to turn excessive US energy consumption from net importer to exporter (Hughes JD 2013 Nature 494, 307–308 doi:10.1038/494307a).

Yet using current solar capture systems, we could meet the annual global demand for energy via sunlight, perhaps, from an area 1,000 x 1000 kilometers of the Sahara  (equivalent to x4 the area of the UK: David Mackay, 2009 Sustainable Energy, UIT Cambridge, or Where is the investment, the vision, the technology to exploit this “free” source of energy? Recent governments have been seemingly indifferent to development of industry technologies to exploit renewables and improve energy efficiency in our housing and business stock. The UK should take a lead and set an example of best practice and embrace the mixed energy generation strategies proposed by Mackay. Say no to nimbyism and do it in your own backyard!

Can we afford to be indifferent, risk the compelling IPCC evidence that man-made climate change is a 95% certainty, and still look our grandchildren in the eye? What will it take to break us from our complacent acceptance of the climatic status quo, the political oxymoron of “sustainable growth” and the short-termist mentality embodied by the “out for nowt and here’s me barrow”- cheap flights, cheap clothes, cheap food is taken for granted by all. And what will it take to convince the general public that we need to act, and quickly, to limit fossil fuel consumption and curb GHG emissions?  Will we need to wait for several more “100 year” extreme climatic events in short order, leading to more human suffering, death and destruction, only this time in an area where real estate is valued considerably more than a few square metres of tin shack and their unfortunate human occupants in the Philippines?

So rather than allowing this agenda to be buried, let us resuscitate, rejoin and reinvigorate the climate change debate.

Howard Griffiths, denizen of a ginkgo-enveloped room, honoured to provide the first post on the Plantsci Ecologists blog