Tag Archives: models

On the Loss of Large Mammals

The loss of large mammals and birds in the Pleistocene was highlighted many years ago (Martin and Wright 1967, Grayson 1977, Guthrie 1984 and many other papers). Hypotheses about why these extinctions occurred were flying left and right for many years with no clear consensus (e.g. Choquenot and Bowman 1998). The museums of the world are filled with mastodons, moas, sabre-tooth tigers and many other skeletons of large mammals and birds long extinct. The topic has come up again in a discussion of these extinctions and a prognosis of future losses (Smith et al. 2018). I do not want to question the analysis in Smith et al. (2018) but I want to concentrate on this one quotation that has captured the essence of this paper in the media:

“Because megafauna have a disproportionate influence on ecosystem structure and function, past and present body size downgrading is reshaping Earth’s biosphere.”
(pg. 310).

What is the evidence for this very strong statement? The first thought that comes to mind is from my botanical colleagues who keep reminding me that plants make of 99% of the biomass of the Earth’s ecosystems. So, if this statement is correct, it must mean that large mammals have a very strong effect on plant ecosystem structure and function. And it must also imply that large mammals are virtually immune to predators, so no trophic cascade can occur to prevent plant overgrazing.

I appreciate that it is very difficult to test such a statement since evolution has been going on for a long time before humans arrived, and so there must have been a lot of other factors causing ecosystem changes in those early years. Humans have a disproportionate love for biodiversity that is larger than us. So, we revel in elephants, tigers, bears, and whales, while at the same time we pay little attention to the insects, small mammals, most fish, and plankton. Because of this size bias, we are greatly concerned with the conservation of large animals, as we should be, but much less concerned about what is happening to the small chaps.

What is the evidence that large mammals and birds have a disproportionate influence on ecosystem structure and function? In my experience, I would say there is very little evidence for strong ecosystem effects from the collapse of the megafauna. DeMaster et al. (2006) evaluated a proposed explanation for ecosystem collapse caused by whaling in the North Pacific Ocean and concluded that the evidence was weak for a sequential megafauna collapse caused by commercial whaling. Trites et al. (2007) and Wade et al. (2007) supported this conclusion. Citing paleo-ecological data for Australia, Johnson (2010) and Rule et al. (2012) argued in another evaluation of ecosystem changes that the human-driven extinction of the megafauna in Australia resulted in large changes in plant communities, potentially confounded by climate change and increases in fire frequency about 40K years ago. If we accept these controversies, we are left with trying to decide if the current losses of large mammals are of similar strength to those assigned to the Pleistocene megafauna, as suggested by Smith et al. (2018).

If we define ecosystem function as primary productivity and ecosystem structure as species diversity, I cannot think of a single case in recent studies where this idea has been clearly tested and supported. Perhaps this simply reflects my biased career working in arctic and subarctic ecosystems in which the vast majority of the energy flow in the system rotates through the smaller species rather than the larger ones. Take the Great Plains of North America with and without the bison herds. What aspect of ecosystem function has changed because of their loss? It is impossible to say because of human intervention in the fire cycle and agricultural pre-emption of much of the landscape. It is certainly correct that overgrazing impacts can be severe in human-managed landscapes with overstocking of cattle and sheep, and that is a tragedy brought on by economics, predator elimination programs, and human land use decisions. All the changes we can describe with paleo-ecological methods have potential explanations that are highly confounded.

I think the challenge is this: to demonstrate that the loss of large mammals at the present time creates a large change in ecosystem structure and function with data on energy flow and species diversity. The only place I can see it possible to do this experimentally today would be in arctic Canada where, at least in some areas, caribou come and go in large numbers and with relatively little human impact. I doubt that you could detect any large effect in this hypothetical experiment. It is the little chaps that matter to ecosystem function, not the big chaps that we all love so much. And I would worry if you could do this experiment, the argument would be that it is a special case of extreme environments not relevant to Africa or Australia.

No one should want the large mammals and birds to disappear, but the question of how this might play out in the coming 200 years in relation to ecosystem function requires more analysis. And unlike the current political inactivity over the looming crisis in climate change, we conservation biologists should certainly try to prevent the loss of megafauna.

Choquenot, D., and Bowman, D.M.J.S. 1998. Marsupial megafauna, Aborigines and the overkill hypothesis: application of predator-prey models to the question of Pleistocene extinction in Australia. Global Ecology and Biogeography Letters 7: 167-180.

DeMaster, D.P., Trites, A.W., Clapham, P., Mizroch, S., Wade, P., Small, R.J., and Hoef, J.V. 2006. The sequential megafaunal collapse hypothesis: testing with existing data. Progress in Oceanography 68(2-4): 329-342. doi:10.1016/j.pocean.2006.02.007

Grayson, D.K. 1977. Pleistocene avifaunas and the Overkill Hypothesis. Science 195: 691-693.

Guthrie, R.D. 1984. Mosaics, allelochemics and nutrients: An ecological theory of late Pleistocene megafaunal extinctions. In: Quaternary Extinctions: A Prehistoric Revolution ed by P.S. Martin and R.G. Klein. University of Arizona Press Tucson.

Johnson, C.N. 2010. Ecological consequences of Late Quaternary extinctions of megafauna. Proceeding of the Royal Society of London, Series B 276(1667): 2509-2519. doi: 10.1098/rspb.2008.1921.

Martin, P.S., and Wright, H.E. (eds). 1967. Pleistocene Extinctions; The Search for a Cause. Yale University Press, New Haven, Connecticut. 453 pp.

Rule, S., Brook, B.W., Haberle, S.G., Turney, C.S.M., Kershaw, A.P., and Johnson, C.N. 2012. The aftermath of megafaunal extinction: ecosystem transformation in Pleistocene Australia. Science 335(6075): 1483-1486. doi: 10.1126/science.1214261.

Smith, F.A., Elliott Smith, R.E., Lyons, S.K., and Payne, J.L. 2018. Body size downgrading of mammals over the late Quaternary. Science 360(6386): 310-313. doi: 10.1126/science.aao5987.

Trites, A.W., Deecke, V.B., Gregr, E.J., Ford, J.K.B., and Olesiuk, P.F. 2007. Killer whales, whaling, and sequential megafaunal collapse in the North Pacific: a comparative analysis of the dynamics of marine mammals in Alaska and British Columbia following commercial whaling. Marine Mammal Science 23(4): 751-765. doi: 10.1111/j.1748-7692.2006.00076.x.

Wade, P.R., et al. 2007. Killer whales and marine mammal trends in the North Pacific – a re-examination of evidence for sequential megafaunal collapse and the prey-switching hypothesis. Marine Mammal Science 23(4): 766-802. doi: 10.1111/j.1748-7692.2006.00093.x.

On Mushrooms, Monitoring, and Prediction

Mushrooms probably run the world but we do not know this yet. My old friend Jim Trappe from Oregon State told me this long ago, and partly as a result of this interaction we began counting mushrooms at our boreal forest sites near Kluane, Yukon in 1993, long ago and even before the iPhone was invented. Being zoologists, we never perhaps appreciated mushrooms in the forest, but we began counting and measuring mushrooms appearing above ground on circular plots of 28m2. With the help of many students, we have counted about 12,000 plots over 24 years, even after being told by one Parks Canada staff member that they could not assist us because “real men do not count mushrooms”. At least we know our position in life.

At any rate the simple question we wanted to ask is whether we can predict mushroom crops one year ahead. We know that many species eat these mushrooms, from red squirrels (who dry mushrooms on spruce tree branches so they can be stored for later consumption), to moose (Alice Kenney has photographed them kneeling down to munch mushrooms), to caribou (Art Rodgers has videoed) to small rodents and insects, not to mention Yukon residents. We know from natural history observations that mushroom crops in the boreal forest are highly variable from year to year, ranging from 0.1 to 110 g/10m2 wet weight, for a CV of 138% (Krebs et al. 2008). The question is how best to predict what the crop will be next year.  Why do we want to know next year’s crop? Two reasons are that large crops provide food for many animals and thus affect overall ecosystem dynamics, and secondly that the essence of understanding in science is the ability to understand why changes occur and if possible to be able to predict them.

We assume it has to be driven by climate, so we can gather together climate data and it is here that the questions arise as to how to proceed. At one extreme we can gather annual temperatures and annual rainfall, and at the other extreme we can gather daily rainfall. We first make the assumption that it is only the weather during the summer from May to August that is relevant for our statistical model, so annual data are not useful. But then we are faced with a nearly infinite number of possible weather variables. We have chosen months as the relevant weather grouping and so we tally May temperature averages, May rainfall totals, growing degree days above 5°C, etc. for all the years involved. This leads us into a statistical nightmare of having far more independent variables than measurements of mushroom crops. If we have, for example, 15 possible measures of temperature and rainfall we can generate 32,768 models ignoring all the interactive models. There are several standard ways of dealing with this statistical dilemma, with stepwise regression being the old fashioned approach. But new methods and advice continue to appear (e.g. Elith et al. 2008, Ives 2015). The ability to compare different regression models with the AIC approach helps (Anderson 2008) as long as there is some biological basis to the models.

We adopted a natural history approach, given that many people believe that large mushroom crops are associated with above average rainfall. We are blessed in the Yukon with only one possible crop of mushrooms per year (at least for the present), so that also simplifies the kinds of models one might use. At any rate (as of 2016) the simplest regression model to predict mushroom biomass in a particular year turned out to involve only rainfall from May (early spring) of the previous year, with R2 = 0.55. But this success has just led us into more questions of why we cannot find a model that will explain the remaining 45% of the variance in annual crops. Should one just give up at this point and be happy that we can explain a large part of the annual variation, or should one press on doing more modelling and looking for other variables? Data dredging is more and more becoming an issue in the ecological literature, and in particular in ecological events likely to be at least partly associated with climate (Norman 2014).

Another ecological problem has been that we do not identify the species of mushrooms involved and deal only in biomass. It may be that species identification would help us to improve predictability. But there are perhaps 40 or more species of mushrooms in our part of the boreal forest, and so we now have to become mycologists. And then as Jim Trappe would tell me, all of this ignores the important questions of what is going on with these fungi underground, so we have only scratched the surface.

The next question is how long a predictive model based on weather will continue to hold in an area subject to rapid climate change. Climate change in the southern Yukon is relatively rapid but highly variable from year to year, and only continuing monitoring will keep us informed about how the physical measurements of temperature and rainfall translate into events in the biological world.

All of this is to say that counting and measuring mushrooms is enjoyable and keeps one connected to the real world. It is also a free type of good exercise, and part of citizen science. Continued monitoring is necessary to see how the boreal ecosystem responds to changing climate and to see if good years for mushroom crops become more frequent. And in good years, many kinds of mushrooms are good to eat if you can beat the squirrels to them.

Anderson, D.R. (2008) Model Based Inference in the Life Sciences: A Primer on Evidence. Springer, New York. ISBN: 978-0-387-74073-7

Elith, J., Leathwick, J.R. & Hastie, T. (2008) A working guide to boosted regression trees. Journal of Animal Ecology, 77, 802-813. doi: 10.1111/j.1365-2656.2008.01390.x

Ives, A.R. (2015) For testing the significance of regression coefficients, go ahead and log-transform count data. Methods in Ecology and Evolution, 6, 828-835. doi: 10.1111/2041-210X.12386

Krebs, C.J., Carrier, P., Boutin, S., Boonstra, R. & Hofer, E.J. (2008) Mushroom crops in relation to weather in the southwestern Yukon. Botany, 86, 1497-1502. doi: 10.1139/B08-094

Norman, G.G. (2014) Data dredging, salami-slicing, and other successful strategies to ensure rejection: twelve tips on how to not get your paper published. Advances in Health Sciences Education, 19, 1-5. doi: 10.1007/s10459-014-9494-8

Fishery Models and Ecological Understanding

Anyone interested in population dynamics, fisheries management, or ecological understanding in general will be interested to read the exchanges in Science, 23 April 2016 on the problem of understanding stock changes in the northern cod (Gadus morhua) fishery in the Gulf of Maine. I think this exchange is important to read because it illustrates two general problems with ecological science – how to understand ecological changes with incomplete data, and how to extrapolate what is happening into taking some management action.

What we have here are sets of experts promoting a management view and others contradicting the suggested view. There is no question but that ecologists have made much progress in understanding both marine and freshwater fisheries. Probably the total number of person-years of research on marine fishes like the northern cod would dwarf that on all other ecological studies combined. Yet we are still arguing about fundamental processes in major marine fisheries. You will remember that the northern cod in particular was one of the largest fisheries in the world when it began to be exploited in the 16th century, and by the 1990s it was driven to about 1% of its prior abundance, almost to the status of a threatened species.

Pershing et al. (2015) suggested, based on data on a rise in sea surface temperature in the Gulf of Maine, that cod mortality had increased with temperature and this was causing the fishery management model to overestimate the allowable catch. Palmer et al. (2016) and Swain et al. (2016) disputed their conclusions, and Pershing et al. (2016) responded. The details are in these papers and I do not pretend to know whose views are closest to be correct.

But I’m interested in two facts. First, Science clearly thought this controversy was important and worth publishing, even in the face of a 99% rejection rate for all submissions to that journal. Second, it illustrates that ecology faces a lot of questions when it makes conclusions that natural resource managers should act upon. Perhaps it is akin to medicine in being controversial, even though it is all supposed to be evidence based. It is hard to imagine physical scientists or engineers arguing so publically over the design of a bridge or a hydroelectric dam. Why is it that ecologists so often spend time arguing with one another over this or that theory or research finding? If we admit that our conclusions about the world’s ecosystems are so meager and uncertain, does it mean we have a very long way to go before we can claim to be a hard science? We would hope not but what is the evidence?

One problem so well illustrated here in these papers is the difficulty of measuring the parameters of change in marine fish populations and then tying these estimates to models that are predictive of changes required for management actions. The combination of less than precise data and models that are overly precise in their assumptions could be a deadly combination in the ecological management of natural resources.

Palmer, M.C., Deroba, J.J., Legault, C.M., and Brooks, E.N. 2016. Comment on “Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery”. Science 352(6284): 423-423. doi:10.1126/science.aad9674.

Pershing, A.J., Alexander, M.A., Hernandez, C.M., Kerr, L.A., Le Bris, A., Mills, K.E., Nye, J.A., Record, N.R., Scannell, H.A., Scott, J.D., Sherwood, G.D., and Thomas, A.C. 2016. Response to Comments on “Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery”. Science 352(6284): 423-423. doi:10.1126/science.aae0463.

Pershing, A.J., Alexander, M.A., Hernandez, C.M., Kerr, L.A., Le Bris, A., Mills, K.E., Nye, J.A., Record, N.R., Scannell, H.A., Scott, J.D., Sherwood, G.D., and Thomas, A.C. 2015. Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery. Science 350(6262): 809-812. doi:10.1126/science.aac9819.

Swain, D.P., Benoît, H.P., Cox, S.P., and Cadigan, N.G. 2016. Comment on “Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery”. Science 352(6284): 423-423. doi:10.1126/science.aad9346.

Hypothesis testing using field data and experiments is definitely NOT a waste of time

At the ESA meeting in 2014 Greg Dwyer (University of Chicago) gave a talk titled “Trying to understand ecological data without mechanistic models is a waste of time.” This theme has recently been reiterated on Dynamic Ecology Jeremy Fox, Brian McGill and Megan Duffy’s blog (25 January 2016 https://dynamicecology.wordpress.com/2016/01/25/trying-to-understand-ecological-data-without-mechanistic-models-is-a-waste-of-time/).  Some immediate responses to this blog have been such things as “What is a mechanistic model?” “What about the use of inappropriate statistics to fit mechanistic models,” and “prediction vs. description from mechanistic models”.  All of these are relevant and interesting issues in interpreting the value of mechanistic models.

The biggest fallacy however in this blog post or at least the title of the blog post is the implication that field ecological data are collected in a vacuum.  Hypotheses are models, conceptual models, and it is only in the absence of hypotheses that trying to understand ecological data is a “waste of time”. Research proposals that fund field work demand testable hypotheses, and testing hypotheses advances science. Research using mechanistic models should also develop testable hypotheses, but mechanistic models are certainly are not the only route to hypothesis creation of testing.

Unfortunately, mechanistic models rarely identify how the robustness and generality of the model output could be tested from ecological data and often fail comprehensively to properly describe the many assumptions made in constructing the model. In fact, they are often presented as complete descriptions of the ecological relationships in question, and methods for model validation are not discussed. Sometimes modelling papers include blatantly unrealistic functions to simplify ecological processes, without exploring the sensitivity of results to the functions.

I can refer to my own area of research expertise, population cycles for an example here.  It is not enough for example to have a pattern of ups and downs with a 10-year periodicity to claim that the model is an acceptable representation of cyclic population dynamics of for example a forest lepidopteran or snowshoe hares. There are many ways to get cyclic dynamics in modeled systems. Scientific progress and understanding can only be made if the outcome of conceptual, mechanistic or statistical models define the hypotheses that could be tested and the experiments that could be conducted to support the acceptance, rejection or modification of the model and thus to inform understanding of natural systems.

How helpful are mechanistic models – the gypsy moth story

Given the implication of Dwyer’s blog post (or at least blog post title) that mechanistic models are the only way to ecological understanding, it is useful to look at models of gypsy moth dynamics, one of Greg’s areas of modeling expertise, with the view toward evaluating whether model assumptions are compatible with real-world data Dwyer et al.  2004  (http://www.nature.com/nature/journal/v430/n6997/abs/nature02569.html)

Although there has been considerable excellent work on gypsy moth over the years, long-term population data are lacking.  Population dynamics therefore are estimated by annual estimates of defoliation carried out by the US Forest Service in New England starting in 1924. These data show periods of non-cyclicity, two ten-year cycles (peaks in 1981 and 1991 that are used by Dwyer for comparison to modeled dynamics for a number of his mechanistic models) and harmonic 4-5 year cycles between 1943 and1979 and since the 1991 outbreak. Based on these data 10-year cycles are the exception not the rule for introduced populations of gypsy moth. Point 1. Many of the Dwyer mechanistic models were tested using the two outbreak periods and ignored over 20 years of subsequent defoliation data lacking 10-year cycles. Thus his results are limited in their generality.

As a further example a recent paper, Elderd et al. (2013)  (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3773759/) explored the relationship between alternating long and short cycles of gypsy moth in oak dominated forests by speculating that inducible tannins in oaks modifies the interactions between gypsy moth larvae and viral infection. Although previous field experiments (D’Amico et al. 1998) http://onlinelibrary.wiley.com/doi/10.1890/0012-9658(1998)079%5b1104:FDDNAW%5d2.0.CO%3b2/abstract concluded that gypsy moth defoliation does not affect tannin levels sufficiently to influence viral infection, Elderd et al. (2013) proposed that induced tannins in red oak foliage reduces variation in viral infection levels and promotes shorter cycles. In this study, an experiment was conducted using jasmonic acid sprays to induce oak foliage. Point 2 This mechanistic model is based on experiments using artificially induced tannins as a mimic of insect damage inducing plant defenses. However, earlier fieldwork showed that foliage damage does not influence virus transmission and thus does not support the relevance of this mechanism.

In this model Elderd et al. (2013) use a linear relationship for viral transmission (transmission of infection on baculovirus density) based on two data points and the 0 intercept. In past mechanistic models and in a number of other systems the relationship between viral transmission and host density is nonlinear (D’Amico et al. 2005, http://onlinelibrary.wiley.com/doi/10.1111/j.0307-6946.2005.00697.x/abstract;jsessionid=D93D281ACD3F94AA86185EFF95AC5119.f02t02?userIsAuthenticated=false&deniedAccessCustomisedMessage= Fenton et al. 2002, http://onlinelibrary.wiley.com/doi/10.1046/j.1365-2656.2002.00656.x/full). Point 3. Data are insufficient to accurately describe the viral transmission relationship used in the model.

Finally the Elderd et al. (2013) model considers two types of gypsy moth habitat – one composed of 43% oaks that are inducible and the other of 15% oaks and the remainder of the forest composition is in adjacent blocks of non-inducible pines. Data show that gypsy moth outbreaks are limited to areas with high frequencies of oaks. In mixed forests, pines are only fed on by later instars of moth larvae when oaks are defoliated. The pines would be interspersed amongst the oaks not in separate blocks as in the modeled population. Point 4.  Patterns of forest composition in the models that are crucial to the result are unrealistic and this makes the interpretation of the results impossible.

Point 5 and conclusion. Because it can be very difficult to critically review someone else’s mechanistic model as model assumptions are often hidden in supplementary material and hard to interpret, and because relationships used in models are often arbitrarily chosen and not based on available data, it could be easy to conclude that “mechanistic models are misleading and a waste of time”. But of course that wouldn’t be productive. So my final point is that closer collaboration between modelers and data collectors would be the best way to ensure that the models are reasonable and accurate representations of the data.  In this way understanding and realistic predictions would be advanced. Unfortunately the great push to publish high profile papers works against this collaboration and manuscripts of mechanistic models rarely include data savvy referees.

D’Amico, V., J. S. Elkinton, G. Dwyer, R. B. Willis, and M. E. Montgomery. 1998. Foliage damage does not affect within-season transmission of an insect virus. Ecology 79:1104-1110.

D’Amico, V. D., J. S. Elkinton, P. J.D., J. P. Buonaccorsi, and G. Dwyer. 2005. Pathogen clumping: an explanation for non-linear transmission of an insect virus. Ecological Entomology 30:383-390.

Dwyer, G., F. Dushoff, and S. H. Yee. 2004. The combined effects of pathogens and predators on insect outbreaks. Nature 430:341-345.

Elderd, B. D., B. J. Rehill, K. J. Haynes, and G. Dwyer. 2013. Induced plant defenses, host–pathogen interactions, and forest insect outbreaks. Proceedings of the National Academy of Sciences 110:14978-14983.

Fenton, A., J. P. Fairbairn, R. Norman, and P. J. Hudson. 2002. Parasite transmission: reconciling theory and reality. Journal of Animal Ecology 71:893-905.

The Volkswagen Syndrome and Ecological Science

We have all been hearing the reports that Volkswagen fixed diesel cars by some engineering trick to show low levels of pollution, while the actual pollution produced on the road is 10-100 times higher than the laboratory predicted pollution levels. I wonder if this is an analogous situation to what we have in ecology when we compare laboratory studies and conclusions to real-world situations.

The push in ecology has always been to simplify the system first by creating models full of assumptions, and then by laboratory experiments that are greatly oversimplified compared with the real world. There are very good reasons to try to do this, since the real world is rather complicated, but I wonder if we should call a partial moratorium on such research by conducting a review of how far we have been led astray by both simple models and simple laboratory population, community and ecosystem studies in microcosms and mesocosms. I can almost hear the screams coming up that of course this is not possible since graduate students must complete a degree in 2 or 3 years, and postdocs must do something in 2 years. If this is our main justification for models and microcosms, that is fair enough but we ought to be explicit about stating that and then evaluate how much we have been misled by such oversimplification.

Let me try to be clear about this problem. It is an empirical question of whether or not studies in laboratory or field microcosms can give us reliable generalizations for much more extensive communities and ecosystems that are not in some sense space limited or time limited. I have a personal view on this question, heavily influenced by studies of small mammal populations in microcosms. But my experience may be atypical of the rest of natural systems, and this is an empirical question, not one on which we can simply state our opinions.

If the world is much more complex than our current understanding of it, we must conclude that an extensive list of climate change papers should be moved to the fiction section of our libraries. If we assume equilibrial dynamics in our communities and ecosystems, we fly in violation of almost all long term studies of populations, communities, and ecosystems. The problem lies in the space and time vision of our science. Our studies are too short to show even a good representation of dynamics over a 100 year time scale, and the problems of landscape ecology highlight that what we see in patch A may be greatly influenced by whether patches B and C are close by or not. We see this darkly in a few small studies but are compelled to believe that such landscape effects are unusual or atypical. This may in fact be the case, but we need much more work to see if it is rare or common. And the broader issue is what use do we as ecologists have for ecological predictions that cannot be tested without data for the next 100 years?

Are all our grand generalizations of ecology falling by the wayside without us noticing it? Prins and Gordon (2014) in their overview seem to feel that the real world is poorly reflected in many of our beloved theories. I think this is a reflection of the Volkswagen Syndrome, of the failure to appreciate that the laboratory in its simplicity is so far removed from real world community and ecosystem dynamics that we ought to start over to build an ecological edifice of generalizations or rules with a strong appreciation of the limited validity of most generalizations until much more research has been done. The complications of the real world can be ignored in the search for simplicity, but one has to do this with the realization that predictions that flow from faulty generalizations can harm our science. We ecologists have very much research yet to do to establish secure generalizations that lead to reliable predictions.

Prins, H.H.T. & Gordon, I.J. (2014) Invasion Biology and Ecological Theory: Insights from a Continent in Transformation. Cambridge University Press, Cambridge. 540 pp. ISBN 9781107035812.

A Survey of Strong Inference in Ecology Papers: Platt’s Test and Medawar’s Fraud Model

In 1897 Chamberlin wrote an article in the Journal of Geology on the method of multiple working hypotheses as a way of experimentally testing scientific ideas (Chamberlin 1897 reprinted in Science). Ecology was scarcely invented at that time and this has stimulated my quest here to see if current ecology journals subscribe to Chamberlin’s approach to science. Platt (1964) formalized this approach as “strong inference” and argued that it was the best way for science to progress rapidly. If this is the case (and some do not agree that this approach is suitable for ecology) then we might use this model to check now and then on the state of ecology via published papers.

I did a very small survey in the Journal of Animal Ecology for 2015. Most ecologists I hope would classify this as one of our leading journals. I asked the simple question of whether in the Introduction to each paper there were explicit hypotheses stated and explicit alternative hypotheses, and categorized each paper as ‘yes’ or ‘no’. There is certainly a problem here in that many papers stated a hypothesis or idea they wanted to investigate but never discussed what the alternative was, or indeed if there was an alternative hypothesis. As a potential set of covariates, I tallied how many times the word ‘hypothesis’ or ‘hypotheses’ occurred in each paper, as well as the word ‘test’, ‘prediction’, and ‘model’. Most ‘model’ and ‘test’ words were used in the context of statistical models or statistical tests of significance. Singular and plural forms of these words were all counted.

This is not a publication and I did not want to spend the rest of my life looking at all the other ecology journals and many issues, so I concentrated on the Journal of Animal Ecology, volume 84, issues 1 and 2 in 2015. I obtained these results for the 51 articles in these two issues: (number of times the word appeared per article, averaged over all articles)

Explicit hypothesis and alternative hypotheses



















No. articles







There are lots of problems with a simple analysis like this and perhaps its utility may lie in stimulating a more sophisticated analysis of a wider variety of journals. It is certainly not a random sample of the ecology literature. But maybe it gives us a few insights into ecology 2015.

I found the results quite surprising in that many papers failed Platt’s Test for strong inference. Many papers stated hypotheses but failed to state alternative hypotheses. In some cases the implied alternative hypothesis is the now-discredited null hypothesis (Johnson 2002). One possible reason for the failure to state hypotheses clearly was discussed by Medawar many years ago (Howitt and Wilson 2014; Medawar 1963). He pointed out that most scientific papers were written backwards, analysing the data, finding out what it concluded, and then writing the introduction to the paper knowing the results to follow. A significant number of papers in these issues I have looked at here seem to have been written following Medawar’s “fraud model”.

But make of such data as you will, and I appreciate that many people write papers in a less formal style than Medawar or Platt would prefer. And many have alternative hypotheses in mind but do not write them down clearly. And perhaps many referees do not think we should be restricted to using the hypothetical deductive approach to science. All of these points of view should be discussed rather than ignored. I note that some ecological journals now turn back papers that have no clear statement of a hypothesis in the introduction to the submitted paper.

The word ‘model’ is the most common word to appear in this analysis, typically in the case of a statistical model evaluated by AIC kinds of statistics. And the word ‘test’ was most commonly used in statistical tests (‘t-test’) in a paper. Indeed virtually all of these paper overflow with statistical estimates of various kinds. Few however come back in the conclusions to state exactly what progress has been made by their paper and even less make statements about what should be done next. From this small survey there is considerable room for improvement in ecological publications.

Chamberlin, T.C. 1897. The method of multiple working hypotheses. Journal of Geology 5: 837-848 (reprinted in Science 148: 754-759 in 1965). doi:10.1126/science.148.3671.754

Howitt, S.M., and Wilson, A.N. 2014. Revisiting “Is the scientific paper a fraud?”. EMBO reports 15(5): 481-484. doi:10.1002/embr.201338302

Johnson, D.H. (2002) The role of hypothesis testing in wildlife science. Journal of Wildlife Management 66(2): 272-276. doi: 10.2307/3803159

Medawar, P.B. 1963. Is the scientific paper a fraud? In “The Threat and the Glory”. Edited by P.B. Medawar. Harper Collins, New York. pp. 228-233. (Reprinted by Harper Collins in 1990. ISBN: 9780060391126.)

Platt, J.R. 1964. Strong inference. Science 146: 347-353. doi:10.1126/science.146.3642.347

On Adaptive Management

I was fortunate to be on the sidelines at UBC in the 1970s when Carl Walters, Ray Hilborn, and Buzz Holling developed and refined the ideas of adaptive management. Working mostly in a fisheries context in which management is both possible and essential, they developed a new paradigm of how to proceed in the management of natural resources to reduce or avoid the mistakes of the past (Walters & Hilborn 1978). Somehow it was one of those times in science where everything worked because these three ecologists were a near perfect fit to one another, full of new ideas and inspired guesses about how to put their ideas into action. Many other scientists joined in, and Holling (1978) put this collaboration together in a book that can still be downloaded from the website of the International Institute for Applied Systems Analysis (IASA) in Vienna:

Adaptive management became the new paradigm, now taken up with gusto by many natural resources and conservation agencies (Westgate, Likens & Lindenmayer 2013). Adaptive management can be carried out in two different ways. Passive adaptive management involves having a model of the system being managed and manipulating it in a series of ways that improve the model fit over time. Active adaptive management takes several different models and uses different management manipulations to decide which model best describes how the system operates. Both approaches intend to reduce the uncertainty about how the system works so as to define the limits of management options.

The message was (as they argued) nothing more than common sense, to learn by doing. But common sense is uncommonly used, as we see too often even in the 21st century. Adaptive management became very popular in the 1990s, but while many took up the banner of adaptive management, relatively few cases have been successfully completed (Walters 2007; Westgate, Likens & Lindenmayer 2013). There are many different reasons for this (discussed well in these two papers), not the least of which is the communication gap between research scientists and resource managers. Research scientists typically wish to test an ecological hypothesis by a management manipulation, but the resource manager may not be able to use this particular management manipulation in practice because it costs too much. To be useful in the real world any management experiment needs to have careful, long-term monitoring to map its outcome, and management agencies do not often have the opportunity to carry out extensive monitoring. The underlying cause then is mainly financial, and resource agencies rarely have an adequate budget to cover the important wildlife and fisheries issues they are supposed to manage.

If anything, reading this ‘old’ literature should remind ecologists that the problems discussed are inherent in management and will not go away as we move into the era of climate change. Let me stop with a few of the guideposts from Holling’s book:

Treat assessment as an ongoing process…
Remember that uncertainties are inherent…
Involve decision makers early in the analysis…
Establish a degree of belief for each of your alternative models…
Avoid facile and narcotic compression of indicators such as cost/benefit ratios that are generally inappropriate for environmental problems….

And probably remind yourself that there can be wisdom in the elders….

The take-home message for me in re-reading these older papers on adaptive management is that it is similar to the problem we have with models in ecology. We can produce simple models or in this case solutions to management problems on paper, but getting them to work properly in the real world where social viewpoints, political power, and scientific information collide is extremely difficult. This is no reason to stop doing the best science and to try to weld it into management agencies. But it is easier said than done.

Holling, C.S. (1978) Adaptive Environmental Assessment and Management. John Wiley and Sons, Chichester, UK.

Walters, C.J. (2007) Is adaptive management helping to solve fisheries problems? Ambio, 36, 304-307.

Walters, C.J. & Hilborn, R. (1978) Ecological optimization and adaptive management. Annual Review of Ecology and Systematics, 9, 157-188.

Westgate, M.J., Likens, G.E. & Lindenmayer, D.B. (2013) Adaptive management of biological systems: A review. Biological Conservation, 158, 128-139.

Is Ecology like Economics?

One statement in Thomas Piketty’s book on economics struck me as a possible description of ecology’s development. On page 32 he states:

“To put it bluntly, the discipline of economics has yet to get over its childish passion for mathematics and for purely theoretical and often highly ideological speculation at the expense of historical research and collaboration with the other social sciences. Economists are all too often preoccupied with petty mathematical problems of interest only to themselves. This obsession with mathematics is an easy way of acquiring the appearance of scientificity without having to answer the far more complex questions posed by the world we live in.”

If this is at least a partially correct summary of ecology’s history, we could argue that finally in the last 20 years ecology has begun to analyze the far more complex questions posed by the ecological world. But it does so with a background of oversimplified models, whether verbal or mathematical, that we are continually trying to fit our data into. Square pegs into round holes.

Part of this problem arises from the hierarchy of science in which physics and in particular mathematics are ranked as the ideals of science to which we should all strive. It is another verbal model of the science world constructed after the fact with little attention to the details of how physics and the other hard sciences have actually progressed over the past three centuries.

Sciences also rank high in the public mind when they provide humans with more gadgets and better cars and airplanes, so that technology and science are always confused. Physics led to engineering which led to all our modern gadgets and progress. Biology has assisted medicine in continually improving human health, and natural history has enriched our lives by raising our appreciation of biodiversity. But ecology has provided a less clearly articulated vision for humans with a new list of commandments that seem to inhibit economic ‘progress’. Much of what we find in conservation biology and wildlife management simply states the obvious that humans have made a terrible mess of life on Earth – extinctions, overharvesting, pollution of lakes and the ocean, and invasive weeds among other things. In some sense ecologists are like the priests of old, warning us that God or some spiritual force will punish us if we violate some commandments or regulations. In our case it is the Earth that suffers from poorly thought out human alterations, and, in a nutshell, CO2 is the new god that will indeed guarantee that the end is near. No one really wants to hear or believe this, if we accept the polls taken in North America.

So the bottom line for ecologists should be to concentrate on the complex questions posed by the biological world, and try first to understand the problems and second to suggest some way to solve them. Much easier said than done, as we can see from the current economic mess in what might be a sister science.

Piketty, T. 2014. Capital in the Twenty-First Century. Belknap Press, Harvard University, Boston. 696 pp. ISBN 9780674430006

The Snowshoe Hare 10-year Cycle – A Cautionary Tale

We have been working on the ten-year cycle of snowshoe hares (Lepus americanus) in the southwest Yukon since 1975 trying to answer the simple question of what causes these cyclic fluctuations. I think that we now understand the causes of the cyclic dynamics, which is not to say all things are known but the broad picture is complete. But some misunderstanding persists, hence this one page summary. Some biology first.

The snowshoe hare cycle has been known from Canada lynx fur return data for more than 100 years, and of course known to First Nations people much before that. Hares are herbivores of small trees and shrubs, they reproduce at age 1 and rarely live more than 1-2 years. They have 2-4 litters in a summer, with litter size around 4-6. Juvenile losses are high and at best populations increase about three-to-four-fold per year. Almost everything eats them – lynx, coyotes, great-horned owls, goshawks, a long list of predators on the young. Reproduction collapses with rising density and females reduce their output from 4 litters to 2 in the peak and decline phase.

The obvious driving factors when Lloyd Keith and his students began working on the hare cycle in Alberta in the 1960s were winter food shortage and predation. When there is a high hare peak, damage to shrubs and small trees is obvious. But it was quite clear in Keith’s studies that the decline phase continued well after the vegetation recovered, and so he postulated a two-factor explanation, winter food shortage followed by high predation losses. He looked for disease and parasite problems in hares but found nothing.

Testing the winter food limitation would appear to be simple but is fraught with problems. Everyone believes that food is an ultimate limiting factor, so that it must be involved in the cyclic dynamics. We began testing food limitation in the mid-1970s and found that one could add natural food or artificial food (rabbit chow) and apparently have no effect on cyclic dynamics. Hares came to the food grids so the density increased by immigration, but the decline started at the same time and at the same rate as on control grids. So what is the role of food?

Our next attempt was to do a factorial experiment adding food, reducing predation, and doing both together. The details are important, replication was never enough for the manipulated treatments, we did it only for 10 years rather than 20 or 30. What we found was that there was an interaction between food addition and mammal predator exclusion so that the combined treatment increased to a much higher density than any single treatment. But this result came with a puzzle. What is the role of food? Hares showed no evidence of malnutrition in the peak or decline, fed hares did not increase their reproductive output. What produced the strong interaction between food addition and predator reduction?

The next breakthrough came when Rudy Boonstra suggested that predator-caused stress might underlie these strange dynamics. Because we could now measure stress with faecal cortisol measures we could test for stress directly in free-ranging hares. The surprise was that this idea worked and Michael Sheriff capped off the stress hypothesis by showing that not only does predator-induced stress reduce reproductive rates, but the stress effect is inherited maternally in the next generation.

The bottom line: the whole dynamics of the snowshoe hare cycle are predator-induced. All the changes in mortality and reproduction are direct and indirect effects of predators chasing and eating hares. The experimental food/predator interaction was mechanistically wrong in targeting food as a major limiting factor.

This of course does not mean that food is irrelevant as an important factor to study in hare cycles. In particular very high peak populations damage shrubs and small trees and we do not yet have the details of how this works out in time. Secondary chemicals are certainly involved here.

Why does all this matter? Two points. First, the hare cycle is often trumpeted as an example of a tri-trophic interaction of food – hares – predators, when in fact it seems to be a simple predator-prey system, as Lotka suggested in 1925. Models of the hare cycle have proliferated over time, and there are far more models of the cycle in existence than there are long-term field studies or field experiments. It is possible to model the hare cycle as a predator-prey oscillation, as a food plant-hare oscillation, as a parasite-hare interaction, as a cosmic particle – hare oscillation, as an intrinsic social – maternal effects interaction, and I have probably missed some other combinations of delayed-density dependent factors that have been discussed. That one can produce a formal mathematical model of the hare cycle does not mean that the chosen factor is the correct one.

The other point I would leave you with is the large amount of field work needed to sort out the mechanisms driving the population dynamics of hares. Ecology is not simple. This enigma of the ten-year cycle has always been a classic example in ecology and perhaps it is now solved. Or perhaps not?

Boonstra, R., D. Hik, G. R. Singleton, and A. Tinnikov. 1998. The impact of predator-induced stress on the snowshoe hare cycle. Ecological Monographs 68:371-394.

Boutin, S., C. J. Krebs, R. Boonstra, M. R. T. Dale, S. J. Hannon, K. Martin, A. R. E. Sinclair, J. N. M. Smith, R. Turkington, M. Blower, A. Byrom, F. I. Doyle, C. Doyle, D. Hik, L. Hofer, A. Hubbs, T. Karels, D. L. Murray, V. Nams, M. O’Donoghue, C. Rohner, and S. Schweiger. 1995. Population changes of the vertebrate community during a snowshoe hare cycle in Canada’s boreal forest. Oikos 74:69-80.

Keith, L. B., and L. A. Windberg. 1978. A demographic analysis of the snowshoe hare cycle. Wildlife Monographs 58:1-70.

Keith, L. B. 1990. Dynamics of snowshoe hare populations. Current Mammalogy 4:119-195.

Krebs, C. J., S. Boutin, R. Boonstra, A. R. E. Sinclair, J. N. M. Smith, M. R. T. Dale, K. Martin, and R. Turkington. 1995. Impact of food and predation on the snowshoe hare cycle. Science 269:1112-1115.

Krebs, C. J., S. Boutin, and R. Boonstra, editors. 2001. Ecosystem Dynamics of the Boreal Forest: the Kluane Project. Oxford University Press, New York.

Sheriff, M. J., C. J. Krebs, and R. Boonstra. 2009. The sensitive hare: sublethal effects of predator stress on reproduction in snowshoe hares. Journal of Animal Ecology 78:1249-1258.

Yan, C., N. C. Stenseth, C. J. Krebs, and Z. Zhang. 2013. Linking climate change to population cycles of hares and lynx. Global Change Biology 19:3263-3271.

Some Reflections on Evo-Eco

Some ecologists study evolutionary processes and we call them evolutionary ecologists. They have their own journals and are a thriving field of science. Other ecologists study populations, communities, and ecosystems in ecological time and do not in general concern themselves with evolutionary changes.The question is should they? Evo-Eco is a search for evolutionary changes that have a decisive impact on observable ecological changes like that of a collapsing bird population.

There are two schools of thought. The first is that evo-eco is very important and the changes that ecologists are trying to understand are partly caused by ecological mechanisms like predation and competition but are also associated with genetic changes that affect survival and reproduction. Consequently an ecologist studying the declining bird population should study both genetics and ecology. The second school of thought is that evo-eco is rarely of any importance in causing ecological changes, so that we can more or less ignore genetics if we wish to understand why this bird population is disappearing.

A practical problem immediately rears its head. To be safe we should all follow evo-eco in case genetics is involved in dynamics. But given the number of problems that ecologists face, the number of scientists available to analyse them, and the research dollars available it is rare to have the time, energy or money to take the comprehensive route. Conservation ecologists are perhaps the most tightly squeezed of all ecologists because they have no time to spare. Environmental managers request answers about what to do, and the immediate causes of conservation problems are (as everyone knows) habitat loss, introduced pests and diseases, and pollution.

The consequence of all this is that the two schools of thought drift apart. I cannot foresee any easy way to solve this issue. Progress in evolutionary ecology is often very slow and knowing the past rarely gives us much insight into predicting the human-affected future. Progress in conventional ecology is faster but our understanding is based on short-term studies of unknown generality for future events. Both schools of thought race along with mathematical models that may or may not tell us anything about the real world, but are conceptually elegant and in a pinch might be called progress if we had time to test them adequately.

The most useful evo-eco approach has been to look at human-caused selection via fishing for large sized fish or hunting for Dall sheep with the largest horns. The overuse of antibiotics for human sickness and as prophylactics for our farm animals is another classic case in which to understand the ecological dynamics we need to know the evolutionary changes that we humans have caused. These are clear cases in which genetic insights can teach us very much.

I end with a story from my past. In the 1950s, nearly 70 years ago now, Dennis Chitty working at Oxford on population fluctuations in small grassland rodents considered that he could reject most of the conventional explanations for animal population changes, and he suggested that individuals might change in quality with population density. This change he thought might involve genetic selection for traits that were favourable only in high density populations that reappeared every 3-4 years. So in some strange sense he was one of the earliest evo-eco ecologists. The result was that he was nearly laughed out of Oxford by the geneticists in control. The great evolutionary geneticist E.B. Ford told Chitty he was completely mad to think that short term selection was possible on a scale to impact population dynamics. Genetic changes took dozens to hundreds of years at the best of time. There were of course in the 1950s only the most primitive of genetic methods available for mammals that all look the same in their coat colour, and the idea that changes in animal behaviour involving territoriality might cause genetic shifts on a short-term period gradually lost favour. Few now think that Chitty was right in being evo-eco, but in some sense he was ahead of his time in thinking that natural selection might operate quickly in field populations. Given the many physiological and behavioural changes that can occur phenotypically in mammals, most subsequent work on grassland rodents has become buried in mechanisms that do not change because of genetic selection.

When we try to sort out whether to be concerned about evo-eco, we must strike a compromise between what the exact question is that we are trying to investigate, and how we can best construct a decision tree that can operate in real time with results that are useful for the research question. Not every ecological problem can be solved by sequencing the study organism.

Chitty, D. 1960. Population processes in the vole and their relevance to general theory. Canadian Journal of Zoology 38:99-113.