Using Experimentation to Understand the 10-Year Snowshoe Hare Cycle (PDF)

Received: 18 December 2016 | Accepted: 1 June 2017 DOI: 10.1111/1365-2656.12720 SYNTHESIS Using experimentation to understand the 10-­year snowshoe hare cycle in the boreal forest of North America Charles J. Krebs1 | Rudy Boonstra2 | Stan Boutin3 1 Department of Zoology...

Received: 18 December 2016 | Accepted: 1 June 2017 DOI: 10.1111/1365-2656.12720 SYNTHESIS Using experimentation to understand the 10-­year snowshoe hare cycle in the boreal forest of North America Charles J. Krebs1 | Rudy Boonstra2 | Stan Boutin3 1 Department of Zoology, University of British Columbia, Vancouver, BC, Canada Abstract 2 Department of Biological Sciences, University 1. Population cycles have long fascinated ecologists from the time of Charles Elton in of Toronto Scarborough, Toronto, ON, Canada the 1920s. The discovery of large population fluctuations in undisturbed ecosys- 3 Department of Biological Sciences, University tems challenged the idea that pristine nature was in a state of balance. The 10-year of Alberta, Edmonton, AB, Canada cycle of snowshoe hares (Lepus americanus Erxleben) across the boreal forests of Correspondence Canada and Alaska is a classic cycle, recognized by fur traders for more than Charles J. Krebs Email: [email protected] 300 years. 2. Since the 1930s, ecologists have investigated the mechanisms that might cause Funding information Natural Sciences and Engineering Research these cycles. Proposed causal mechanisms have varied from sunspots to food sup- Council of Canada plies, parasites, diseases, predation and social behaviour. Both the birth rate and Handling Editor: Ken Wilson the death rate change dramatically over the cycle. Social behaviour was eliminated as a possible cause because snowshoe hares are not territorial and do not commit infanticide. 3. Since the 1960s, large-scale manipulative experiments have been used to discover the major limiting factors. Food supply and predation quickly became recognized as potential key factors causing the cycle. Experiments adding food and restricting predator access to field populations have been decisive in pinpointing predation as the key mechanism causing these fluctuations. 4. The immediate cause of death of most snowshoe hares is predation by a variety of predators, including the Canada lynx (Lynx canadensis Kerr). The collapse in the re- productive rate is not due to food shortage as was originally thought, but is a result of chronic stress from predator chases. 5. Five major issues remain unresolved. First, what is the nature of the predator-in- duced memory that results in the prolonged low phase of the cycle? Second, why do hare cycles form a travelling wave, starting in the centre of the boreal forest in Saskatchewan and travelling across western Canada and Alaska? Third, why does the amplitude of the cycle vary greatly from one cycle to the next in the same area? Fourth, do the same mechanisms of population limitation apply to snowshoe hares in eastern North American or in similar ecosystems across Siberia? Finally, what effect will climatic warming have on all the above issues? The answers to these questions remain for future generations of biologists to determine. KEYWORDS boreal forest, food shortage, Kluane ecosystem, Lepus americanus, predation, sublethal stress, synchrony, travelling waves, Yukon J Anim Ecol. 2018;87:87–100. wileyonlinelibrary.com/journal/jane © 2017 The Authors. Journal of Animal Ecology | 87 © 2017 British Ecological Society 88 | Journal of Animal Ecology KREBS et al. 1 | INTRODUCTION 3. The hare cycle was caused by heavy predation and the time-lag in- herent in predator rates of population growth relative to that of the Population fluctuations have always captivated people partly because hares. they can affect livelihoods as pest species that attack crops or as game 4. The hare cycle was caused by both food and predators, with food animals that provide meat or furs. When Charles Elton discovered the shortage at the peak followed by heavy predation in the decline. detailed data compiled by the Hudson Bay Company on furs traded from different parts of Canada since 1673, he quickly realized that Our approach consisted of obtaining detailed information on de- these data were at variance with the common belief in the stability of mographic parameters combined with measures of quantity and quality natural systems not subject to human disturbance. Elton began what of hare food in the winter and estimates of kill rates of hare predators has become a tidal wave of literature on population cycles and their to test predictions following from each hypothesis. We went one step causes and consequences for ecosystems. The natural variation in de- further, however, by experimentally manipulating food availability and mography exhibited by cyclic populations has proven extremely useful predation rates. Our approach was not guided by a priori mathematical for understanding the complex interplay between plants, herbivores models because there was only limited empirical information to parame- and predators. terize such models and the manipulations were the obvious ones based The 10-­year cycle of snowshoe hares and their predators showed on the Lloyd Keith work. up clearly in fur returns from across Canada and Alaska, and presented itself as a test case for understanding one particular population cycle which had a 300-­year history written in fur returns on a continen- 2 | MATERIALS AND METHODS tal scale. The list of ecological factors that could produce a cycle was quickly identified since the mechanisms had to show a time-­lag in their All our studies on snowshoe hares were conducted by live trapping effects. Overgrazing and food shortage fits this profile, as does preda- and radiocollaring of individuals. The details of these methods and tion mortality, parasites and diseases. Other factors like sunspots were those for radiotelemetry have been described in detail in Hodges et al. brought into the picture but quickly dismissed as potential agents be- (2001). Population estimates of all the hare data presented here were cause their temporal fluctuations did not match the snowshoe hare calculated in DENSITY 4 and 5 (Efford, 2009). The major experiments cyclic time-­scale. are described in detail in Boutin et al. (2001), and much more detail Early work by Green and Larson (1938) and Green, Larson, and about our general methods is given in Krebs, Boutin, and Boonstra Bell (1939) postulated that a stress disease they called “shock disease” (2001). was the cause of cyclic declines. This early work was dismissed as an artefact of studies done on hares in crowded laboratory rooms. But 2.1 | Natural history of the snowshoe hare the general idea of Green and Larson (1938) that hares might suffer from some intrinsic ailment was dormant until 60 years later when it Critical to understanding the population dynamics of snowshoe hares was brought back in studies of chronic stress (Boonstra, 2013) as de- is knowledge of their life history and the constraints of their diet and scribed below. habitat. These hares are the major herbivores in the boreal forest In 1948, William Rowan, the Head of Zoology at the University of ecosystem, both in terms of biomass and of impact (Boonstra et al., Alberta, declared that the 10-­year cycle was the “outstanding prob- 2016). We summarize the key features of their natural history here lem of Canadian conservation.” He stimulated Lloyd Keith to carry out (see Hodges, 2000 for references). They are pure white in winter and a series of studies on the snowshoe hare cycle in Alberta, beginning brown in summer. All breeding takes place in summer, with a maxi- his career with an early book on cycles (Keith, 1963). Lloyd Keith was mum of four litters being produced. Females breed synchronously convinced that the hare cycle was driven by two factors, food shortage and engage in post-­partum insemination. The gestation period is in winter, which depressed reproduction, and predation in the decline 35–37 days, with the first litter (born in late May in the Yukon) aver- phase. He and his students’ research in central Alberta was the first to aging three leverets, the second five, and the third and fourth (if they lay a firm quantitative foundation to the demography of the hare cycle have them) between three and five leverets. Young do not breed in (Keith, 1983; Keith & Windberg, 1978). their year of birth. Although the rare hare can live 7 years, most live From this earlier research, we launched our research programme very short lives (the average life span is 1 year), with almost all mor- in 1976 in the Kluane Region of the Yukon. Previous research had laid tality being attributed to predation (Boutin, Krebs, Sinclair, & Smith, out four clear hypotheses to test experimentally: 1986; Hodges et al., 2001; Keith & Windberg, 1978; Murray, Keith, & Cary, 1998), and 70% of breeding females being yearlings. Virtually 1. The hare cycle was caused by winter food shortage and the all avian and mammalian predators in the boreal forest eat hares (see delayed recovery of winter browse damaged by excessive brows- the food web, Figure 6), with even the granivore and herbivore (red ing by hares. squirrels and Arctic ground squirrels respectively) eating hare leverets 2. The hare cycle was caused by changes in food quality as a result of in summer. In winter, snowshoe hares are basically the only prey for heavy browsing by peak hare densities which triggered an increase predators to eat, since red squirrels are much less vulnerable due to in secondary compound production by the plants. reduced activity, ground squirrels hibernate, grouse are relatively rare KREBS et al. Journal of Animal Ecology | 89 and small mammals live in the subnivean space below the snow. Hare 10 Control diet in winter is largely restricted to the twigs of tall shrubs (dwarf 7 birch Betula glandulosa and willow Salix glauca) and the ends of spruce 4 1 Population density per ha branches available above the snowpack, but in summer includes forbs, 2 grasses, leaves and some woody browse. The availability of browse in 2 the winter is not only complex and dynamic as changing snow depth 3 allows hares to access twigs at greater heights but it also makes twigs 1 Food 4 population decline, 1981 to 1984. Two control trapping 0.7 inaccessible as hares do not burrow into snow to access covered twigs 5 (Keith, Cary, Rongstad, & Brittingham, 1984). Deaths during winter, 0.4 even during the peak and decline, are not driven by starvation caused 6 (red symbols) by absolute food shortage, but rather by increasing predation rates 0.2 7 symbols, Krebs, Boutin & Gilbert (Keith et al., 1984). 8 winters with no effect on the rate of population decline. 0.1 April Oct. April Oct. April Oct. April Oct. 1980 1981 1982 1983 2.2 | Our experimental results F I G U R E 1 Natural feeding experiment on snowshoe hares at 2.2.1 | Food addition experiments Kluane Lake, Yukon, during a population decline, 1981–1984. Two control trapping areas (red and blue symbols) were monitored until In 1976, we decided to test the simplest hypothesis for hare cycles, October 1981 when the feeding experiment began on one area that it was a result of winter food shortage. We set out three 10 by (symbols) with winter feeding of felled white spruce and aspen trees 10 grids (30-­m spacing of live trap stations, 7.3 ha) for winter feeding (dark green symbols, Krebs, Boutin, et al., 1986). Feeding continued during the following three winters with no effect on the rate of of commercial rabbit chow (Krebs, Gilbert, Boutin, Sinclair, & Smith, population decline 1986) for the period from 1977 to 1984. One of the two grids on islands was too difficult to provision in winter and we had to discard it. We had many problems with disturbances to feeding stations by that the carry-­over of severe browsing could influence the snowshoe bears and moose so these experiments were not perfect. They were, hare cycle via food quality. Fox and Bryant (1984) postulated that in- however, most consistent with the hypothesis that winter food short- creases in secondary plant chemicals after heavy browsing resulted in age was not necessary for snowshoe hare declines because although high levels for 2–3 years after browsing damage, and this time delay the food addition served to increase density during the increase and could be a delayed density-­dependent factor in generating hare cy- peak phases, it failed to stop the hare decline. We were criticized for cles via changes in food quality. Sinclair, Krebs, Smith, and Boutin these feeding experiments because we used artificial high-­quality rab- (1988) tested the secondary plant chemistry hypothesis by measuring bit chow as food and we pushed densities to above normal. To assess phenols and resins in winter food plants over one hare cycle. They the validity of these criticisms we carried out another experiment of found that, contrary to the predictions of this hypothesis, secondary feeding hares in winter on natural food. Hares completely debark and compounds decreased from the increase phase to the peak phase and consume the twigs of trees that blow down naturally. So we cut down into the decline of the hare cycle. While secondary chemicals had a large white spruce (Picea glauca (Moench) Voss) and aspen (Populus strong influence on food choice by hares (Rodgers & Sinclair, 1997), tremuloides Michx.) trees to feed hares on a 9-­ha plot in winter (Krebs, food quality did not appear to be a limiting factor for hare population Boutin, & Gilbert, 1986). Control and manipulated hare populations fluctuations. The complexities of plant secondary compound effects declined in unison with no measurable effect of the extra natural food on herbivores urges caution in reaching this conclusion (Torregrossa provided in the decline (Figure 1). We decided that, while winter food & Dearing, 2009) and more research is needed in other parts of the is very important to hares, it was not the limiting factor in the Kluane boreal forest. boreal forest in these years. We also found that the proximate cause of virtually all the mortalities of our sample of radiocollared hares 2.2.3 | Food addition and predator reduction could be attributed to predation (Boutin et al., 1986) which matched experiments the findings by Keith et al. (1984) studying a hare crash in Rochester, Alberta. Thus, we had two replicates of a hare crash pointing to preda- It is always possible that a combination of factors determines pop- tion as the causal agent. ulation growth rates, hence, during the next cycle (1986–1996) we decided to manipulate both food and predation. We fed three larger live trapping grids (20 × 20 trap points, 30-­m spacing, 32.5 ha) con- 2.2.2 | Food quality experiments tinuously year-­round with commercial rabbit chow spread by a ferti- The second hypothesis postulating food quality as a driver of the hare lizer spreader (Boutin et al., 2001). We surrounded one of these grids cycle has been more controversial and difficult to test. Bryant (1981) with a 2-­m-­high electrified fence around 1 km2 to keep mammalian showed that severe browsing in winter increased the level of toxic predators out. We could not keep avian predators out, so this was secondary compounds in four species of deciduous trees in Alaska and a predator reduction experiment, not a complete predator removal 90 | Journal of Animal Ecology KREBS et al. manipulation. We built a second electrified fence around another area to try to measure the impact of mammalian predator removal alone. 0.25 We fertilized two other large areas (1 km2) with commercial NPK ferti- lizer to test the bottom-­up model of regulation. All these experiments 0.20 Survival rate (per year) are described in detail in Krebs et al. (2001). The results surprised us (Krebs et al., 1995) and a synopsis of 0.15 these 10-­year experiments is illustrated in Figure 2. Feeding hares ap- proximately tripled population density but did not affect the decline 0.10 (as illustrated earlier in Figure 1). Density increase in feeding areas was largely by immigration rather than by increased reproductive suc- 0.05 cess. The largest effect occurred on the combined food + predator reduction area, where densities reached about 11 times the control 0.00 Control Fertilizer Food Predator Predator values at the cyclic peak. Statistically this shows an interaction be- exclosure exclosure tween food and predation. But this was a puzzle because we could + food not detect any indication that winter food supplies were insufficient F I G U R E 3 Annual survival rates (95% CL) for adult snowshoe on control areas (Hodges, Boonstra, & Krebs, 2006; Hodges, Stefan, hares with radiocollars during the decline phase of the cycle from & Gillis, 1999). Population declines in hares are the result of severely autumn 1990 to autumn 1992. Sample sizes ranged from 197 to 278 decreased survival rates during the crash and one of our objectives individuals for these estimates (After Krebs et al., 1995) with our experiments was to see if we could maintain survival rates to prevent the decline. Although we improved survival on all our treat- prevented us from replicating the important mammalian predator re- ments, the maximum effect was seen on the combined food + pred- moval treatment. We were faced with a dilemma; replicate the predator ator reduction area (Figure 3) where survival rates were high enough fence or add the interaction treatment of food supplementation and to maintain densities at peak control levels well after the other areas predator exclusion. We did have multiple contemporary control pop- had crashed. ulations plus the detailed demographic information collected in previ- As with our studies of the previous cycle we found that the proxi- ous cycles as context to compare our experimental results but a purist mate cause of virtually all adult hare mortality was predation. Thus, the would argue that the major differences we observed could have been evidence continued to build for predation rather than food shortage due to unknown inherent differences between control and experimen- as being a necessary driver of the hare cycle. Our experiments proved tal sites. Only further replication will resolve this issue. We also could conclusively that high-­quality food could not keep hares from the jaws not control hare or predator movements in and out of our treatment and talons of their predators nor did hares die of starvation when pro- areas which affected density and survival estimates (Turchin, 2003). tected from predators. We allowed hares to move freely in and out of the predator fences and Our experimental approach to testing the role of food and preda- food supplemented areas from fear of creating a “fence effect” but we tion in the hare cycle has been criticized on several fronts. Logistics fenced areas three times the size of our trapping grids to try to prevent dispersal out of the fence. Many of our radiocollared animals still moved 15 outside of the fence where they were killed by predators. It is also likely Exclosure that predators spatially aggregated where our treatments created high 13 Fertilizer hare densities, especially as the population decline continued. Ratio of treatment density Food 11 Exclosure We have continued to monitor three control grids since the major to control density + food project of 1986–1996 ended, so that we now have a 41-­year record of 9 snowshoe hare numbers in this part of the Yukon. Figure 4 illustrates 7 the sequence of hare densities for control areas at Kluane. There has 5 been an irregular but observable trend to lower and lower peak hare densities over the period from 1976 to 2016. Traditional knowledge 3 from Kluane First Nations people reported to us that the 1970–1971 1 peak was even higher than the 1980–1981 peak. The continued de- cline in hare peak numbers over this time has been accompanied by Increase Peak Decline Low Average a continual increase in birch and willow shrubs that are the basis of F I G U R E 2 Ratio of population densities for the four experimental the winter hare diet. Grabowski (2015) showed that standing biomass treatments to average control population densities at the same of dwarf birch approximately doubled between the 1987–1994 sam- phase of the hare cycle. If there is no treatment effect, we expect pling period and 2014, while grey willow (Salix glauca L.) increased a ratio of 1.0. During the peak and decline phases, the mammalian about 50% in biomass during that time. The cause of increased shrub predator exclosure doubled density, food addition tripled density, and the combined treatment of food addition and predator reduction growth was probably a mixture of reduced overwinter hare browsing, increased density 11-­fold increasing light levels caused by white spruce tree mortality from the KREBS et al. Journal of Animal Ecology | 91 5 4 Hare fall density/ha 3 F I G U R E 4 Snowshoe hare autumn 2 population density (per ha) on control trapping grids, 1977–2016. Upper 95% confidence limits are shown. There 1 has been a gradual but slightly uneven decline in peak densities since the 1981 peak [Colour figure can be viewed at 0 wileyonlinelibrary.com] 1980 1985 1990 1995 2000 2005 2010 2015 spruce bark beetle (Dendroctonus rufipennis (Kirby)) outbreak (Berg, by predation risk; (ii) increased maternal stress results in a decrease Henry, Fastie, De Volder, & Matsuoka, 2006) and a warming climate. in reproduction; (iii) maternal stress is inherited from mothers to off- From 1985 to 2016 in the Kluane area May average temperatures spring; and (iv) offspring from stressed mothers also produce fewer have been increased 1.5°C, and mean June to August temperatures offspring. The first three of these predictions have been confirmed. have increased 0.3°C (data from Haines Junction Meteorological Breeding females were most stressed during summers of peak preda- Station, Environment Canada). Early winter temperatures (October to tor numbers, stressed females produced fewer offspring and the off- December) have increased slightly (0.44°C) in these 31 years, while spring of stressed females also were also stressed (Figure 5), so that late winter average temperatures (January to March) have increased the continuation of reproductive failure was carried from generation 2.8°C. Thus, the climate in the Kluane area is warming but the warming to generation by maternal effects. We still need to show that high is uneven with slight summer warming and stronger winter warming, predator-­induced stress causes a failure to produce late summer litters all compounded by high variability from year to year. 3 and 4, and that offspring from stressed mothers actually have high stress levels when they themselves breed. The exact mechanism by which maternal stress programmes 2.2.4 | Alternative hypotheses for the decline in offspring is not yet known. It does not appear to be due to genetic reproductive rates changes (Sinclair, Chitty, Stefan, & Krebs, 2003). Changes in gene ex- Cary and Keith (1979) had shown in an elegant study that hare repro- pression have been found (Lavergne, McGowan, Krebs, & Boonstra, ductive output collapsed over the cycle but the collapse began 2 years 2014) and the most plausible hypothesis is that these changes are prior to hare peak density and continued through the decline phase. linked to epigenetic changes in expression of key regulatory genes, Stefan and Krebs (2001) repeated these observations for Kluane especially those affecting the stress axis (Ho & Burggren, 2010). One hares. The problem then became how to explain a collapse in repro- of the central enigmas of the hare cycle is the 2–5-­year low phase ductive rates in the absence of observable food limitation. Either we had insufficient information on access to food or food quality, or some other process reduced reproductive rates. Boonstra and Singleton 800 Offspring stress level (FCM ng/g) (1993) and Boonstra, Hik, Singleton, and Tinnikov (1998) found that 700 hares were severely stressed during the population decline and pos- 600 tulated that stress was both the proximate cause of the reproductive collapse and the long-­term cause of the low phase (acting through 500 maternal effects), and that a likely source of stress was the action of 400 predators searching for hares to consume. This suggestion arrived at a 300 critical time because new non-­invasive methods had been developed to measure stress in wild mammals without having to regularly collect 200 blood samples (Sheriff, Dantzer, Delehanty, Palme, & Boonstra, 2011). 100 The hypothesis that predator-­induced stress caused the repro- 100 200 300 400 500 600 700 800 900 ductive collapse was tested experimentally and observationally by Maternal stress level (FCM ng/g) Sheriff, Krebs, and Boonstra (2009, 2010, 2011), who measured stress F I G U R E 5 Maternal inheritance of stress levels from females levels by means of both plasma cortisol levels and their faecal me- to offspring. Each point is from the litter of a single female (n = 8) tabolites. Four criteria had to be confirmed before this hypothesis measured over the increase to decline phase of the cycle, ± 1 SE could be tentatively accepted: (i) hares are sensitive to stress caused (After Sheriff et al., 2010; Figure 2) 92 | Journal of Animal Ecology KREBS et al. following the decline (Figure 4). If there is stress-­induced maternal Keith, Cary, Yuill, and Keith (1985), Keith, Keith, and Cary (1986) car- programming of offspring that persists into adulthood, this could ex- ried out an extensive study of helminth parasites of snowshoe hares plain the low phase. But why does it last a variable length of time and in central Alberta. Five parasite species were prominent in the hare what causes the females eventually to return to their highly fecund population but there was no indication that prevalence affected any state? We suspect, but do not know, that the epigenetic changes may reproductive parameters of the hare population over the population take time to dilute out of the population. Sheriff, McMahon, Krebs, cycle. Murray, Cary, and Keith (1997) reduced nematode prevalence and Boonstra (2015) found that the length of the low phase was a experimentally in hares during 2 years to determine if parasitism in- function of the severity of the decline phase, suggesting that the more creased vulnerability to predation. Virtually all hares in their study died severe the predation risk, the greater the epigenetic signature, and by from predation, and they found a significant increase in predation on extension, the more generations required to remove that signature. untreated hares relative to treated ones, which suggested that there This working hypothesis remains to be tested. An alternative hypoth- might be an interaction between parasitism and predation. esis by Tyson, Haines, and Hodges (2010) suggests that the prolonged In the Kluane area there has been no evidence found of high mor- low phase of the cycle may be due to the special role that great horned tality in snowshoe hares directly from diseases or parasitism, so the owls may play as predators during this phase. The cycle low and the only potential link may be through increasing vulnerability to preda- factors that trigger the return to increase remain the least well-­studied tors. The role of pathogens in the system is as yet unexplored. More components of the hare cycle. research is needed on the role of parasites and pathogens in hare cycles, although our Kluane studies would suggest that these effects would be small. In the same manner, we and others like Murray et al. 2.2.5 | Alternative hypotheses for the decline in (1997) have found almost no deaths in hares that could be ascribed to survival rates starvation (Hodges et al., 2006). Hare populations decline in the pres- There are many predators that feed on snowshoe hares from “herbi- ence of superabundant food, as shown in Figure 1. vores” like red squirrels (Tamiasciurus hudsonicus (Erxleben)) and griz- zly bears (Ursus arctos (L.)) to more specialized carnivores like Canada 2.2.6 | Synchrony in snowshoe hare cycles lynx (Figure 6). The result of this food web is that few hares die of old age, and for about 95% of hares the immediate cause of death is pre- There remains a widespread belief that snowshoe hare cycles occur in dation (Boutin et al., 1986; Hodges et al., 2001). But the cycle is often synchrony across all North America. This is not correct, as was pointed mislabelled and modelled as a hare–lynx cycle, and there is a need to out long ago by Chitty (1950). The reality is much more interesting. consider models that consider multiple predators as well as multiple Smith (1983) analysed the questionnaire data of the Snowshoe Rabbit prey species (Tyson et al., 2010). Enquiry of the 1930s and 1940s and suggested that hare peaks fol- Two alternative explanations for declines in hare survival are that lowed a travelling wave (Figure 7) with delays up to 4 years in peak parasites or diseases reduce the condition of hares, thus allowing pred- numbers across Canada. ators to deliver the coup de grâce. There is as yet no good evidence To determine if a travelling wave is currently occurring, Krebs et al. that this explanation operates in the Kluane system (Sovell, 1993). (2013) gathered the existing survey data of snowshoe hares from Golden Great-horned Hawk owl eagle Lynx Coyote Owl Red fox Goshawk Marten Kestrel Wolverine Red-tailed hawk Passerine Wolf birds Northern harrier Humans Weasel Grizzly black bear Spruce Small Red Ground Snowshoe grouse and Moose Insects dall sheep rodents squirrel squirrel hare ptarmigan bark beetles F I G U R E 6 Herbivore and carnivore food web for major species in the Kluane Yukon terrestrial ecosystem. Species shaded in yellow are the main species for which we have quantitative data. Occasional diet items are not shown in this diagram [Colour figure can be viewed at wileyonlinelibrary.com] KREBS et al. Journal of Animal Ecology | 93 F I G U R E 7 Synchrony in snowshoe hare cycles across Canada from 1931–1948, as measured by questionnaires (Chitty, 1950). The average peak phase across Canada was scaled as 0.0, and the contour lines indicate peaks occurring earlier than average (red, negative contours) or later than average (green, positive contours). During this period hare peaks were reached earliest in the central boreal region of northern Saskatchewan and Manitoba (After Smith, 1983) central British Columbia, Yukon, Northwest Territories and Alaska 2.2.7 | Variable amplitude in snowshoe hare cycles from 1970 to 2012. No extensive data were available from central and A final general observation about 10-­year cycles has been that they eastern Canada. The western part of the travelling wave described by are highly variable in amplitude (Krebs et al., 2014). Some hare peaks Smith (1983) from trapper questionnaire data in the 1930s and 1940s are very high (super-­peaks) and show obvious signs of extensive still exists (Figure 8). The reason for this travelling wave is not yet clear. browsing on winter shrubs, and other peaks are nearly invisible to the The best suggestion is that it results from the movements of surplus casual observer (Figure 4). Understanding the problem of amplitude predators in search of higher prey abundance as the snowshoe hare variation is a landscape issue with all the problems of studying events population collapses (Sherratt, 2001). We know from collared animals that are spatially extensive (Lewis, Hodges, Koehler, & Mills, 2011). that lynx move in all directions up to 1100 km from their point of origi- On a small scale, Ginzburg and Krebs (2015) explored the possibility nal capture (Mowat, Poole, & O’Donoghue, 2000) (Figure 9). Row et al. that snowshoe hare cycle amplitude was defined by the minimal abun- (2012) showed that Canada lynx were essentially panmictic across all dance of hare predators in the low phase of the cycle. Because all the 6000 km of the Canadian mainland, suggesting widespread dispersal predators of hares increase more slowly than hares, hares will escape on a continental scale. heavy predation losses for a longer phase of increase, thus reaching If patterns of synchrony are indeed driven by mobile predators, higher densities until the predators catch up. Hence, the lower the we are still left with needing an explanation of why the cycle seems density of predators during the low phase, the greater the peak hare to “start” in central Saskatchewan and radiate outwards (Figure 8). We density and the more browsing damage. This hypothesis works for have no idea why this might be. 94 | Journal of Animal Ecology KREBS et al. F I G U R E 8 A travelling wave of peak snowshoe hare abundance in northwestern North America during the period from 1970 to 2012. Red dots indicate sites with quantitative data on hare abundance. Blue arrows indicate approximate travelling wave with each additional arrow from south to north indicating a 1–2 year time-­lag in the arrival of peak numbers (Data from Krebs et al., 2013) [Colour figure can be viewed at wileyonlinelibrary.com] Kluane hare populations but needs to be tested in other parts of the (Vaccinium spp.) is adapted to the mild temperatures and deep snows hare’s geographic range. and dominates in northwestern Eurasia. Ultimately the occurrence of The effect of habitat patchiness on hare cyclic amplitude offers the 10-­year cycle in the boreal forest of North America is driven bot- another explanation of super-­peaks. Extensive forest fires, for exam- tom-­up by severe winter climate. That being the case, we predict that ple, may produce optimal hare habitat over large areas. If what is opti- severity of the winter climate will maintain the tall shrub vegetation mal for hares is much less optimal for predators, this could affect the as one progresses eastward across the continent (hence the declining overall predation pressure during the population increase phase. In a tree line as one progresses eastward) and that 10-­year cycles are ex- similar manner if parasites vary in abundance in areas fragmented by pected to occur throughout these boreal forest regions. In Eurasia, the fires, a reduction in parasite prevalence might allow better reproduc- western regions are heavily influenced by warm air masses and mari- tive success and survival (Murray et al., 1998). time climate from the Atlantic. However, east of the Ural Mountains, we expect a severe continental climate (analogous to that in northern Canada) across Siberia and with it a change in the vegetation to favour 2.2.8 | Hare cycles in eastern North America and tall shrubs in the understorey and 10-­year hare cycles. The evidence, eastern Eurasia based on fur returns over relatively short time periods of ~20 years, All our activities have been focused in western North America and we is suggestive, but consistent with this prediction. Bulmer (1984) con- do not know the degree to which they apply to the boreal forest of cluded that the mountain hare appeared to have a cycle length of eastern North America, but we expect that they do. The 10-­year cycle 8 years in the Komi region just west of the Urals, and of 11 years in the is dominant in the boreal forest of Canada and across Siberia but in the far east of Siberia in the Yakutia region. In both cases, the peak in the boreal forest of northern Europe 3–4-­year cycles of voles dominate European lynx returns lagged 1–2 years behind the hare peak. community processes. Boonstra et al. (2016) explained these strik- ing differences as being climatically driven (Figure 10). Temperatures 2.3 | Models of the hare cycle and in the former are 15–20°C colder in winter. This directly affects the alternative approaches vegetation that can occur on the two areas—tall shrubs (birch and wil- low) are adapted to severe cold and shallow snows of winter and domi- Turchin (2003) formulated several criticisms of our experimental work, nate in western North America, whereas a luxurious dwarf shrub layer the most notable being that our predictions and analyses were not KREBS et al. Journal of Animal Ecology | 95 F I G U R E 9 Dispersal movements recorded from radiocollared Canada lynx from the point of capture in the Yukon and Northwest Territories to the point of death due to fur trapping. A maximum straight line movement of 1,100 km has been recorded (After Mowat et al., 2000) [Colour figure can be viewed at wileyonlinelibrary.com] F I G U R E 1 0 A proposed explanation for how winter climate—acting mainly through temperature and snow quantity—give rise to the different vegetation and food web dynamics found in the boreal forests of western North America and northwestern Europe (From Boonstra et al., 2016) [Colour figure can be viewed at wileyonlinelibrary. com] theoretically informed by any sort of mathematical model; something King and Schaffer (2001) when they state that “qualitative differences in he considered to be essential when dealing with systems driven by com- behaviour can result from quantitative differences in parameter values” plex nonlinear dynamics. The essence of the argument is captured by such that “the articulation of alternative verbal hypotheses and their 96 | Journal of Animal Ecology KREBS et al. evaluation by strong inference—the conventional biological approach— suggested about what the cause of these cycles could be. The list may be doomed from the outset”. We interpret this to mean that de- of possible mechanisms for population changes is very large, start- pending on circumstances, our experimental treatments might point to ing with climate, food supplies, predation, parasitism, disease and the importance of one factor, whereas in another cycle they could point an array of social factors like territoriality and infanticide. Each of to an alternative factor even though the underlying mechanisms for the these general mechanisms can then be broken down into a series of cycle remain consistent. The only way to evaluate this argument is with alternatives. For example, food shortage as a general limiting factor replicated studies on different hare cycles. This has been done now for could operate in summer or winter, involve juveniles or adults, could 45 years at Kluane Lake and our work was based on pioneering experi- involve food quality as well as quantity, and be involved with sec- ments and studies by Lloyd Keith and his colleagues in Alberta (Keith ondary chemicals in the food plants. If we take only the five broad et al., 1984). The result of these replicate studies is that the demography categories above, we can calculate that there are 31 combinations and proximate causes of mortality of the cycles studied are consistent, of these, each one of which is a distinct hypothesis. Faced with this so that a single set of mechanisms should be sought as an explanation. impossible agenda, ecologists must use natural history insights to re- Attempts to model the hare cycle have been surprisingly few duce the hypotheses to a manageable level. Given this reduction in (summarized in Korpimäki & Krebs, 1996). Early models published by possible mechanisms, we must test each hypothesis experimentally Fox and Bryant (1984) and Trostel, Sinclair, Walters, and Krebs (1987) in the field. We have focused on food and predation as the two most served as background for our experimental manipulations, but large likely main processes driving hare population dynamics. Since in this gaps in the empirical information needed to parameterize the sorts case each experimental test must be carried out for 10 years, pro- of models suggested by Turchin (2003) limited their utility. King and gress cannot be rapid. Schaffer (2001) constructed a standard trophic interaction model link- There are two shortcuts to do this difficult, bottom-­up exper- ing vegetation, hares and predators (Turchin, 2003) and parameter- imental work. First, one could build a mathematical model of the ized it using the empirical findings from Rochester and Kluane. The cycle and compute the consequences of the assumptions made in model could produce cycles that were similar in period and amplitude the model. This is an attractive top-­down methodology but, in our to empirical observations and it was also able to reproduce results of case, despite extensive modelling by many ecologists, little light was the Kluane experiments. King and Schaffer (2001) analyses led them shed on the actual mechanisms causing the cycle until we did our to the following conclusion: “Regardless of the relative importance manipulations at Kluane Lake. The reason for this is twofold. Many of predation and starvation in a given locale, the model predicts that models contain parameters that cannot possibly be measured in nat- although resource limitation is responsible for arresting the increase ural systems. Secondly, a whole host of simplifying assumptions must phase of the cycle, it is increasing predation mortality that brings be made to avoid an infinite regress in model building. The result about the crash.” The King and Schaffer model has been criticized as has been that a host of models exist for cyclic population dynamics, being “overfitted” because of the large number of parameters involved virtually none of which are useful in guiding experimental studies in (Ginzburg & Jensen, 2004). In addition, the key link to vegetation in natural systems. the model was created by a standard density dependence relationship A second shortcut in ecological studies has been to utilize lab- between hare condition and reproduction or mortality. There was no oratory populations in microcosms to mimic nature. While this is a consideration of the non-­consumptive effects of predation on hare re- useful strategy for many systems, it does not work for mammal or production hypothesized by Boonstra et al. (1998) as an alternative to bird populations both because of scale and of lack of complexity and resource limitation as an explanation of reproductive changes. realism. The dispersal of individuals is a critical process for many All in all, empirical studies and mathematical models are important populations, and in every moderate size microcosm it is difficult tools for testing factors important in the hare cycle. We do not agree to permit dispersal dynamics and habitat selection as would occur with Turchin (2003) that systems should be modelled before proceed- in natural ecosystems. Home ranges of snowshoe hares vary from ing with empirical studies and experiments. Both methods need to about 2 ha to 7 ha depending on sex and density (Boutin, 1984), so inform the other. We have been frustrated by the lack of testable pre- that holding hares in small enclosures risks potential artefacts. The dictions stemming from modelling papers (but see King and Shaffer for most misleading early episode for snowshoe hares was the conclu- an exception) and we have pointed out some of the frustrations mod- sion of Green et al. (1939) that shock disease caused hare declines ellers have with our empirical analyses. In the end, the conclusions of (Chitty, 1959). the modelling work were not vastly different from our empirical work Both the food and the predator hypotheses are difficult to test for or the statistical modelling of Stenseth, Falck, Bjørnstad, and Krebs field populations unless clear hypotheses are stated with explicit pre- (1997) but more linkage between various approaches is needed. dictions and unless experimental manipulations are carried out. This is perhaps the major advance of our work over the last 40 years—that experimental design is critical for population studies and manipula- 3 | DISCUSSION tions can be done at relevant spatial scales. But the other crucial thing here is that we had to look inside the black box that is the animal to When the 10-­year cycle of snowshoe hares was first described about look for mechanisms that might explain the reproductive changes in 100 years ago by biologists, there were a multitude of hypotheses the hare cycle from studies of stress physiology. KREBS et al. Journal of Animal Ecology | 97 3.1 | Next steps to the magnitude and mechanism of how maternal effects could act and disappear during the low. We now have information in the literature on three to five complete 4. New technologies like proximity radiocollars are available to follow snowshoe hare cycles from only two study areas in the boreal forest. the spatial location and activity of hares and their predators, and To date, the basic patterns seem consistent; predation is a prime driver should provide better insight into both the consumptive and non- of the cycle, but there are four important areas for future research. consumptive effects of predators on hare demography. 1. The Kluane experiments need to be repeated and improved upon. There is a clear need for geographic expansion of studies to other regions of the boreal forest and further replication. 4 | CONCLUSIONS 2. The mechanism behind the decrease in hare reproduction in the late increase, peak and crash needs to be precisely nailed down. The early knowledge of snowshoe hare cycles helped to move the The current hypothesis of an indirect, non-consumptive effect of scientific consensus away from the idea that nature was in a state of predation via the stress axis needs further testing. balance, which was disturbed only by humans or perhaps angry dei- 3. The cyclic low and early recovery remains as the most poorly un- ties. It took ecologists until about 1950 to begin to take seriously the derstood part of the cycle. At least two hypotheses now exist to question of why natural populations fluctuated, and what we could explain the low phase and models could be useful in testing ideas as do about them if they damaged our livelihoods. Snowshoe hares Does winter food shortage cause the hare cycle? Manipulate winter food Do predation and food shortage together cause the hare cycle? Manipulate summer and winter food and reduce predation and combine these treatments Interactive effects of added food and predator reduction Does food shortage cause reduced reproductive output? Could predators cause reduced reproductive output? Measure stress in Measure stress in Is stress level reproductive females offspring of stressed inherited? over the cycle females Hypothesis Confirmation F I G U R E 1 1 Flow chart of the 40 years of research that have allowed us to develop a comprehensive view of the ecological Predation causes the hare cycle both via direct mortality mechanisms behind the 10-­year cycle of and reduced reproduction. Predator chases stress females, snowshoe hares in the Yukon. The negative resulting in a reduction in reproductive output. Stress symbols indicate no effect of the proposed effects are inherited via a maternal effect, which maintains a factor [Colour figure can be viewed at low phase for 2–4 years. wileyonlinelibrary.com] 98 | Journal of Animal Ecology KREBS et al. were an early bellwether of ecosystem fluctuations that needed to and Engineering Research Council of Canada. The authors declare be understood in northern landscapes. As such they became a classi- no conflict of interest in this publication. The facilities of the Kluane cal Canadian ecological icon that appears in the beginning of almost Lake Research Station of the Arctic Institute of North America were every biology textbook. essential to this long-­term research programme, and we thank Andy We have achieved a tentative explanation of what causes snow- and Carole Williams and Sian Williams and Lance Goodwin for their shoe hare cycles (Figure 11). That is progress in our view. But we assistance. We dedicate this review to the memory of Charles Elton recognize that all scientific conclusions are tentative no matter how and Dennis and Helen Chitty who pioneered all this research. extensive the study. In our case, we draw the sweeping conclusion that the cause of snowshoe hare population cycles across all of the boreal forests of Canada, Alaska and Siberia are caused by predation AU T HO R S ’ CO NT R I B U T I O NS acting directly on mortality and indirectly on a landscape of fear pro- All authors contributed to the field research and to the writing of this ducing chronic stress in breeding females. We have no idea if this synthesis paper. sweeping generalization is correct, and can only wait for additional studies of hares across this vast region to test our conclusions. We fully subscribe to the concept of science as “conjecture and refuta- DATA ACC ES S I B I L I T Y tion,” and there is still much to do to test our understanding of boreal forest ecology. The snowshoe hare population data presented here are available from We have gradually expanded our view from population ecology the Dryad Digital Repository https://doi.org/10.5061/dryad.684s1 to community and ecosystem ecology of the Yukon boreal forest. The (Krebs, Boonstra, & Boutin, 2017). These data include the metadata, interactions between the components of the boreal forest community phases of the population cycle, capture–recapture data for all the con- need to be described and understood before we can hope to predict trol grids live trapped, fence+food mark–recapture data, population how climate change will alter these interactions. We would suggest estimates for control grids and for the natural feeding experiment. that detailed studies of movement patterns of the major predators— coyotes (Canis latrans Say), lynx, great horned owls (Bubo virginianus REFERENCES Gmelin) and goshawks (Acciper gentilis L.)—in this ecosystem will help Berg, E. E., Henry, J. D., Fastie, C. L., De Volder, A. D., & Matsuoka, S. M. us to tie what happens locally to the extensive forests of northern (2006). Spruce beetle outbreaks on the Kenai Peninsula, Alaska, and North America. There is a major gap in the few current studies of plant Kluane National Park and Reserve, Yukon Territory: Relationship to dynamics within this ecosystem, both from the point of view of plant– summer temperatures and regional differences in disturbance regimes. herbivore interactions and that of plant succession during a time of Forest Ecology and Management, 227, 219–232. Boonstra, R. (2013). Reality as the leading cause of stress: Rethinking the rapid climate change. These issues are long-­term problems with which impact of chronic stress in nature. Functional Ecology, 27, 11–23. science deals poorly. If understanding the hare cycle was a 40-­year Boonstra, R., Andreassen, H. P., Boutin, S., Hušek, J., Ims, R. A., Krebs, C. J., problem, with many questions yet unanswered, the community and … Wabakken, P. (2016). Why do the boreal forest ecosystems of ecosystem issues in the boreal forest are 100-­ and 200-­year problems northwestern Europe differ from those of western North America? BioScience, 66, 722–734. at a time when the science funding time frame is 3–5 years. We need Boonstra, R., Hik, D., Singleton, G. R., & Tinnikov, A. (1998). The impact to ask ourselves what the ecologists of the year 2100 will wish we of predator-­induced stress on the snowshoe hare cycle. Ecological had done now to advance the understanding of natural systems. A fair Monographs, 68, 371–394. question with no simple answer. Boonstra, R., & Singleton, G. R. (1993). Population declines in the snow- shoe hare and the role of stress. General and Comparative Endocrinology, 91, 126–143. ACKNOWLE DG E MEN TS Boutin, S. (1984). Home range size and methods of estimating snowshoe hare densities. Acta Zoologica Fennica, 171, 275–278. This research would not have been possible without the collabora- Boutin, S., Krebs, C. J., Nams, V. O., Sinclair, A. R. E., Boonstra, R., O’Donoghue, M., & Doyle, C. (2001). Experimental design and practical tion of many university professors, graduate students, postdoctoral problems of implementation. In C. J. Krebs, S. Boutin, & R. Boonstra fellows and technicians. In particular, we thank Tony Sinclair, the late (Eds.), Ecosystem dynamics of the boreal forest: The Kluane Project (pp. Jamie Smith, Roy Turkington, Kathy Martin, Susan Hannon and Mark 49–66). New York, NY: Oxford University Press. Dale for their collaboration on the major project. Alice Kenney is the Boutin, S., Krebs, C. J., Sinclair, A. R. E., & Smith, J. M. N. (1986). Proximate causes of losses in a snowshoe hare population. Canadian Journal of longest serving research associate, followed closely by Irene Wingate, Zoology, 64, 606–610. Scott Gilbert, Jean Carey, Elizabeth Hofer, Peter Upton and Mark Bryant, J. P. (1981). Phytochemical deterrence of snowshoe hare browsing O’Donoghue. Many of our research students and assistants have gone by adventitious shoots of four Alaskan trees. Science, 213, 889–890. on to further research positions, including David Hik, Karen Hodges, Bulmer, M. G. (1984). A statistical analysis of the 10-­year cycle in Canada. Dennis Murray, Vilis Nams, Michael Sheriff, Jean Carey and Richard Journal of Animal Ecology, 43, 701–718. Cary, J. R., & Keith, L. B. (1979). Reproductive change in the 10-­year cycle Ward. For all our summer assistants and volunteers, we are most of snowshoe hares. Canadian Journal of Zoology, 57, 375–390. grateful for the help that allowed us to achieve our research goals. Chitty, H. (1950). The snowshoe rabbit enquiry, 1946-­48. Journal of Animal Funding for this research project was provided by the Natural Sciences Ecology, 19, 15–20. KREBS et al. Journal of Animal Ecology | 99 Chitty, D. (1959). A note on shock disease. Ecology, 40, 728–731. Krebs, C. J., Boutin, S., & Gilbert, B. S. (1986). A natural feeding experiment Efford, M. G. (2009). DENSITY 4.4: Software for spatially explicit capture– on a declining snowshoe hare population. Oecologia, 70, 194–197. recapture. Dunedin, New Zealand: Department of Zoology, University Krebs, C. J., Bryant, J., O’Donoghue, M., Kielland, K., Doyle, F., McIntyre, of Otago. C., … Golden, H. (2014). What factors determine cyclic ampli- Fox, J. F., & Bryant, J. P. (1984). Instability of the snowshoe hare and woody tude in the snowshoe hare cycle? Canadian Journal of Zoology, 92, plant interaction. Oecologia, 63, 128–135. 1039–1048. Ginzburg, L. R., & Jensen, C. X. J. (2004). Rules of thumb for judging ecolog- Krebs, C. J., Gilbert, B. S., Boutin, S., Sinclair, A. R. E., & Smith, J. N. M. ical theories. Trends in Ecology and Evolution, 19, 121–126. (1986). Population biology of snowshoe hares. I. Demography of food-­ Ginzburg, L. R., & Krebs, C. J. (2015). Mammalian cycles: Internally defined supplemented populations in the southern Yukon, 1976-­84. Journal of periods and interaction-­driven amplitudes. PeerJ, 3, e1180. Animal Ecology, 55, 963–982. Grabowski, M. (2015). Interspecific boreal shrub growth response to cli- Krebs, C. J., Kielland, K., Bryant, J., O’Donoghue, M., Doyle, F., McIntyre, mate, fertilization and herbivory. MSc, University of British Columbia, C., … Burke, T. (2013). Synchrony in the snowshoe hare cycle in north- Vancouver. western North America, 1970-­2012. Canadian Journal of Zoology, 91, Green, R. G., & Larson, C. L. (1938). Shock disease and the snowshoe hare 562–572. cycle. Science, 87, 298–299. Lavergne, S., McGowan, P. O., Krebs, C. J., & Boonstra, R. (2014). Impact of Green, R. G., Larson, C. L., & Bell, J. F. (1939). Shock disease as the cause high predation risk on genome-­wide hippocampal gene expression in of the periodic decimation of the snowshoe hare. American Journal of snowshoe hares. Oecologia, 176, 613–624. Hygiene, Section B, 30, 83–102. Lewis, C. W., Hodges, K. E., Koehler, G. M., & Mills, L. S. (2011). Influence Ho, D. H., & Burggren, W. W. (2010). Epigenetics and transgenerational of stand and landscape features on snowshoe hare abundance in frag- transfer: A physiological perspective. Journal of Experimental Biology, mented forests. Journal of Mammalogy, 92, 561–567. 213, 3–16. Mowat, G., Poole, K. G., & O’Donoghue, M. (2000). Ecology of lynx in north- Hodges, K. E. (2000). Ecology of snowshoe hares in southern boreal and ern Canada and Alaska. In L. F. Ruggiero, K. B. Aubry, S. W. Buskirk, G. montane forests. In L. F. Ruggiero, K. B. Aubry, S. W. Buskirk, G. M. M. Koehler, C. J. Krebs, K. S. McKelvey, & J. R. Squires (Eds.), Ecology Koehler, C. J. Krebs, K. S. McKelvey, & J. R. Squires (Eds.), Ecology and and conservation of lynx in the United States (pp. 265–306). Denver, CO: conservation of lynx in the United States (pp. 163–206). Denver, CO: University Press of Colorado. University Press of Colorado. Murray, D. L., Cary, J. R., & Keith, L. B. (1997). Interactive effects of sub- Hodges, K. E., Boonstra, R., & Krebs, C. J. (2006). Overwinter mass loss of lethal nematodes and nutritional status on snowshoe hare vulnerability snowshoe hares in the Yukon: Starvation, stress, adaptation or artifact? to predation. Journal of Animal Ecology, 66, 250–264. Journal of Animal Ecology, 75, 1–13. Murray, D. L., Keith, L. B., & Cary, J. R. (1998). Do parasitism and nutritional Hodges, K. E., Krebs, C. J., Hik, D. S., Stefan, C. I., Gillis, E. A., & Doyle, status interact to affect production in snowshoe hares? Ecology, 79, C. E. (2001). Snowshoe hare demography. In C. J. Krebs, S. Boutin, & 1209–1222. R. Boonstra (Eds.), Ecosystem dynamics of the boreal forest: The Kluane Rodgers, A. R., & Sinclair, A. R. E. (1997). Diet choice and nutrition of cap- Project (pp. 141–178). New York, NY: Oxford University Press. tive snowshoe hares (Lepus americanus): Interactions of energy, protein, Hodges, K. E., Stefan, C. I., & Gillis, E. A. (1999). Does body condition affect and plant secondary compounds. Ecoscience, 4, 163–169. fecundity in a cyclic population of snowshoe hares? Canadian Journal Row, J. R., Gomez, C., Koen, E. L., Bowman, J., Murray, D. L., & Wilson, of Zoology, 77, 1–6. P. J. (2012). Dispersal promotes high gene flow among Canada lynx Keith, L. B. (1963). Wildlife’s ten-year cycle. Madison, WI: University of populations across mainland North America. Conservation Genetics, 13, Wisconsin Press. 1259–1268. Keith, L. B. (1983). Role of food in hare population cycles. Oikos, 40, Sheriff, M. J., Dantzer, B., Delehanty, B., Palme, R., & Boonstra, R. (2011). 385–395. Measuring stress in wildlife: Techniques for quantifying glucocorti- Keith, L. B., Cary, J. R., Rongstad, O. J., & Brittingham, M. C. (1984). coids. Oecologia, 166, 869–887. Demography and ecology of a declining snowshoe hare population. Sheriff, M. J., Krebs, C. J., & Boonstra, R. (2009). The sensitive hare: Wildlife Monographs, 90, 1–43. Sublethal effects of predator stress on reproduction in snowshoe Keith, L. B., Cary, J. R., Yuill, T. M., & Keith, I. M. (1985). Prevalence of hares. Journal of Animal Ecology, 78, 1249–1258. helminths in a cyclic snowshoe hare population. Journal of Wildlife Sheriff, M. J., Krebs, C. J., & Boonstra, R. (2010). The ghosts of predators Diseases, 21, 233–253. past: Population cycles and the role of maternal programming under Keith, I. M., Keith, L. B., & Cary, J. R. (1986). Parasitism in a declining pop- fluctuating predation risk. Ecology, 91, 2983–2994. ulation of snowshoe hares. Journal of Wildlife Diseases, 22, 349–363. Sheriff, M. J., Krebs, C. J., & Boonstra, R. (2011). From pattern to process: Keith, L. B., & Windberg, L. A. (1978). A demographic analysis of the snow- How fluctuating predation risk impacts the stress axis of snowshoe shoe hare cycle. Wildlife Monographs, 58, 1–70. hares during the 10-­year cycle. Oecologia, 166, 593–605. King, A. A., & Schaffer, W. M. (2001). The geometry of a population cycle: Sheriff, M. J., McMahon, E., Krebs, C. J., & Boonstra, R. (2015). Predator-­ A mechanistic model of snowshoe hare demography. Ecology, 82, induced maternal stress and population demography in snowshoe 814–830. hares: The more severe the risk, the longer the generational effect. Korpimäki, E., & Krebs, C. J. (1996). Predation and population cycles of Journal of Zoology, 296, 305–310. small mammals. BioScience, 46, 754–764. Sherratt, J. A. (2001). Periodic travelling waves in cyclic predator–prey sys- Krebs, C. J., Boonstra, R., & Boutin, S. (2017). Data from: Using exper- tems. Ecology Letters, 4, 30–37. imentation to understand the 10-­year snowshoe hare cycle in the Sinclair, A. R. E., Chitty, D., Stefan, C. I., & Krebs, C. J. (2003). Mammal pop- boreal forest of North America. Dryad Digital Repository, https://doi. ulation cycles: Evidence for intrinsic differences during snowshoe hare org/10.5061/dryad.684s1 cycles. Canadian Journal of Zoology, 81, 216–220. Krebs, C. J., Boutin, S., & Boonstra, R. (2001). Ecosystem dynamics of the Sinclair, A. R. E., Krebs, C. J., Smith, J. N. M., & Boutin, S. (1988). Population boreal forest: The Kluane Project. New York, NY: Oxford University Press, biology of snowshoe hares III. Nutrition, plant secondary compounds 511 pp. and food limitation. Journal of Animal Ecology, 57, 787–806. Krebs, C. J., Boutin, S., Boonstra, R., Sinclair, A. R. E., Smith, J. N. M., Dale, Smith, C. H. (1983). Spatial trends in Canadian snowshoe hare, Lepus M. R. T., … Turkington, R. (1995). Impact of food and predation on the americanus, population cycles. Canadian Field-­Naturalist, 97, snowshoe hare cycle. Science, 269, 1112–1115. 151–160. 100 | Journal of Animal Ecology KREBS et al. Sovell, J. R. (1993). Attempt to determine the influence of parasitism on a Turchin, P. (2003). Complex population dynamics: A theoretical/empirical syn- snowshoe hare population during the peak and initial decline phases of the thesis. Princeton, NJ: Princeton University Press. hare cycle. MSc. University of Alberta, Edmonton. Tyson, R., Haines, S., & Hodges, K. E. (2010). Modelling the Canada lynx Stefan, C. I., & Krebs, C. J. (2001). Reproductive changes in a cyclic popu- and snowshoe hare population cycle: The role of specialist predators. lation of snowshoe hares. Canadian Journal of Zoology, 79, 2101–2108. Theoretical Ecology, 3, 97–111. Stenseth, N. C., Falck, W., Bjørnstad, O. N., & Krebs, C. J. (1997). Population regulation in snowshoe hare and Canadian lynx: Asymmetric food web configurations between hare and lynx. Proceedings of the National How to cite this article: Krebs CJ, Boonstra R, Boutin S. Using Academy of Sciences of the USA, 94, 5147–5152. experimentation to understand the 10-­year snowshoe hare Torregrossa, A.-M., & Dearing, M. D. (2009). Nutritional toxicology of mammals: Regulated intake of plant secondary compounds. Functional cycle in the boreal forest of North America. J Anim Ecol. Ecology, 23, 48–56. 2018;87:87–100. https://doi.org/10.1111/1365-2656.12720 Trostel, K., Sinclair, A. R. E., Walters, C. J., & Krebs, C. J. (1987). Can preda- tion cause the 10-­year cycle? Oecologia, 74, 185–192.