Monthly Archives: January 2018

Three Approaches to Ecology

I ask the question here why ecology is not appreciated as a science at a time when it is critical to the survival of the existing world. So the first question we need to answer is if this premise is correct. I offer only one example. A university zoology department has recently produced a discussion paper on its plans for faculty recruitment over the next 15 years. This document does not include the word “ecology” in any of its forward planning. Now it is probably not unusual for biology or zoology departments in major universities to downplay ecology when there is so much excitement in molecular biology, but it is an indicator that ecology is not a good place to put your money and reputation as you await a Nobel Prize. So if we can accept the initial premise that ecology is not appreciated, we might ask why this situation exists, a point raised long ago by O’Connor (2000). Here are a few thoughts on the matter.

There are three broad approaches to the science of ecology – theoretical ecology, empirical ecology, and applied ecology. These three areas of ecology rarely talk to each other, although one might hope that they could in future evolve into a seamless thread of science.

Theoretical ecology deals with the mathematical world that has too often only a tangential concern with ecological problems. It has its own journals and a whole set of elegant discussions that have few connections to the real world. It is most useful for exploring what might be if we make certain mathematical assumptions. It is without question the most prestigious part of the broad science of ecology, partly because it involves elegant mathematics and partly because it does not get involved in all the complexities of real-world ecological systems. It is the physics of ecology. As such it carries on in its own world and tends to be ignored by most of those working in the other two broad areas of ecology.

Empirical ecology has set itself the task of understanding how the natural world works at the level of individuals, populations, communities and ecosystems. In its pure form it does not care about solving practical ecological or environmental problems, but its practitioners assume probably correctly that the information they provide will in fact be useful now or in the future. It seeks generality but rarely finds it because all individuals and species differ in how they play the ecological game of survival. If it has a mantra, it is “the devil is in the details”. The problem is the details of empirical ecology are boring to politicians, business people, and to much of the television generation now operating with a 7 second or 140 character limit on concentration.

Applied ecology is where the action is now, and if you wish to be relevant and topical you should be an applied ecologist, whether a conservation biologist, a forester, or an agricultural scientist. The mantra of applied ecologists is to do no harm to the environment while solving real world problems. Applied ecologists are forced to put the human imprint into empirical ecology, so they are very much concerned with declining populations and extinctions of plants and animals. The main but not the sole impact of humans is on climate change, so much of applied ecology traces back to the impacts of climate change on ecosystems, all added to by the increasing human population with its rising expectations. But applied ecologists are always behind the environmental problems of the day because the issues multiply faster than possible solutions can be evaluated. This ought to make for high employment for applied ecologists but in fact the opposite seems to be happening because governments too often avoid long-term problems beyond their 4-year mandate. If you do not agree, think climate change.

So, the consequence is that we have three independent worlds out there. Applied ecologists are too busy to apply the successful paradigms of empirical ecology to their problems because they are under strict time limits by their line managers who need to suggest immediate action on problems. They must therefore fire off solutions like golf balls in all directions, hoping that some might actually help solve problems. Empirical ecologists may not be helpful for applied ecologists if they are overwhelmed by the details of their particular system of study and are constrained by the ‘publish or perish’ mentality of the granting agencies.

Finally, we lay on top all this a lack of funding in the environmental sciences for investigating and solving both immediate and long-term ecological problems. And I am back to my favourite quote in the ecological literature:

“Humans, including ecologists, have a peculiar fascination with attempting to correct one ecological mistake with another, rather than removing the source of the problem.” (Schindler 1997).

What can we do about this? Three things. Pressure our politicians to increase funding on long-term environmental problems. This will provide the person-power to find and test solutions to our known problems. Vote with your ballot and your feet to improve sustainability. And whether you are young or old strive to do no harm to the Earth. And if all this is too difficult, take some practical advice not to buy a house in Miami Beach, or any house near the beach. Do something for the environment every day.

 

O’Connor, R.J. (2000) Why ecology lags behind biology. The Scientist 14(20):35. (October 16, 2000).

Schindler, D.W. (1997) Liming to restore acidified lakes and streams: a typical approach to restoring damaged ecosystems? Restoration Ecology 5:1-6

 

On Evolution and Ecology and Climate Change

If ecology can team up with evolution to become a predictive science, we can all profit greatly since it will make us more like physics and the hard sciences. It is highly desirable to have a grand vision of accomplishing this, but there could be a few roadblocks on the way. A recent paper by Bay et al. (2018) illustrates some of the difficulties we face.

The yellow warbler (Setophaga petechia) has a broad breeding range across the United States and Canada, and could therefore be a good species to survey because it inhabits widely different climatic zones. Bay et al. (2018) identified genomic variation associated with climate across the breeding range of this migratory songbird, and concluded that populations requiring the greatest shifts in allele frequencies to keep pace with future climate change have experienced the largest population declines, suggesting that failure to adapt may have already negatively affected population abundance. This study by Bay et al. (2018) sampled 229 yellow warblers from 21 locations across North America, with an average of 10 birds per sample area (range n = 6 to 21). They examined 104,711 single-nucleotide polymorphisms. They correlated genetic structure to 19 climate variables and 3 vegetation indices, a measure of surface moisture, and average elevation. This is an important study claiming to support an important conclusion, and consequently it is also important to break it down into the three major assumptions on which it rests.

First, this study is a space for time analysis, a subject of much discussion already in plant ecology (e.g. Pickett 1989, Blois et al. 2013). It is an untested assumption that you can substitute space for time in analyzing for future evolutionary changes.

Second, the conclusions of the Bay et al. paper rest on an assumption that you have adequate data on the genetics involved in change and on the demography of the species. A clear understanding of the ecology of the species and what limits its distribution and abundance would seem to be prerequisites for understanding the mechanisms of how evolutionary changes might occur.

The third assumption is that, if there is a correlation between the genetic measures and the climate or vegetation indices, one can identify the precise ‘genomic vulnerability’ of the local population. Genomic variation was most closely related to precipitation variables at each site. The geographic area with one of the highest scores in genomic vulnerability was in the desert area of the intermountain west (USA). As far as I can determine from their Figure 1, there was only one sampling site in this whole area of the intermountain west. Finally Bay et al. (2018) compared the genomic vulnerability data to the population changes reported for each site. Population changes for each sampled site were obtained from the North American Breeding Bird Survey data from 1996 to 2012.

The genetic data and its analysis are more impressive, and since I am not a genetics expert I will simply give it a A grade for genetics. It is the ecology that worries me. I doubt that the North American Breeding Bird Survey is a very precise measure of population changes in any particular area. But following the Bay et al. paper, assume that it is a good measure of changing abundance for the yellow warbler. From the Bay et al. paper abstract we see this prediction:

“Populations requiring the greatest shifts in allele frequencies to keep pace with future climate change have experienced the largest population declines, suggesting that failure to adapt may have already negatively affected populations.”

The prediction is illustrated in Figure 1 below from the Bay et al. paper.

Figure 1. From Bay et al. (2018) Figure 2C. (Red dot explained below).

Consider a single case, the Great Basin, area S09 of the Sauer et al. (2017) breeding bird surveys. From the map in Bay et al. (2018) Figure 2 we get the prediction of a very high genomic vulnerability (above 0.06, approximate red dot in Figure 1 above) for the Great Basin, and thus a strongly declining population trend. But if we go to the Sauer et al. (2017) database, we get this result for the Great Basin (Figure 2 here), a completely stable yellow warbler population for the last 45 years.

Figure 2. Data for the Great Basin populations of the Yellow Warbler from the North American Breeding Bird Survey, 1967 to 2015 (area S09). (From Sauer et al. 2017)

One clue to this discrepancy is shown in Figure 1 above where R2 = 0.10, which is to say the predictability of this genomic model is near zero.

So where does this leave us? We have what appears to be an A grade genetic analysis coupled with a D- grade ecological model in which explanations are not based on any mechanism of population dynamics, so that the model presented is useless for any predictions that can be tested in the next 10-20 years. I am far from convinced that this is a useful exercise. It would be a good paper for a graduate seminar discussion. Marvelous genetics, very poor ecology.

And as a footnote I note that mammalian ecologists have already taken a different but more insightful approach to this whole problem of climate-driven adaptation (Boutin and Lane 2014).

Bay, R.A., Harrigan, R.J., Underwood, V.L., Gibbs, H.L., Smith, T.B., and Ruegg, K. 2018. Genomic signals of selection predict climate-driven population declines in a migratory bird. Science 359(6371): 83-86. doi: 10.1126/science.aan4380.

Blois, J.L., Williams, J.W., Fitzpatrick, M.C., Jackson, S.T., and Ferrier, S. 2013. Space can substitute for time in predicting climate-change effects on biodiversity. Proceedings of the National Academy of Sciences 110(23): 9374-9379. doi: 10.1073/pnas.1220228110.

Boutin, S., and Lane, J.E. 2014. Climate change and mammals: evolutionary versus plastic responses. Evolutionary Applications 7(1): 29-41. doi: 10.1111/eva.12121.

Pickett, S.T.A. 1989. Space-for-Time substitution as an alternative to long-term studies. In Long-Term Studies in Ecology: Approaches and Alternatives. Edited by G.E. Likens. Springer New York, New York, NY. pp. 110-135.

Sauer, J.R., Niven, D.K., Hines, J.E., D. J. Ziolkowski, J., Pardieck, K.L., and Fallon, J.E. 2017. The North American Breeding Bird Survey, Results and Analysis 1966 – 2015. USGS Patuxent Wildlife Research Center, Laurel, MD. https://www.mbr-pwrc.usgs.gov/bbs/