Part 4

Mountain hare harvest opportunities and carrying capacity

The harvest footprints of the mountain hare suggest another way to think about sustainable culling, based on encounter-based harvest rates. With this approach, bag sizes simply reflect how many mountain hares are actually encountered during hunting, rather than being set in advance.

The main benefit of encounter-based harvest is that it is simple and robust. It does not require population surveys, age-structure information, or life-table estimates. Harvest instead becomes a largely random outcome that reflects the current population in the field. Management is based on following long-term trends in harvest statistics. Repeated declines in bag sizes can signal that the carrying capacity of the forest habitat is deteriorating.

In contrast to adaptive management used for moose, where the aim is to actively regulate population size, encounter-based harvest mainly serves as an indicator of actual carrying capacity. It tells us how well the ecosystem is functioning, rather than trying to control numbers directly.

The long-term harvest density of the mountain hare highlights four distinct conditions along the timeline: two phases at equilibrium density, a logistic growth phase, a rapid decline phase, and a low density phase, where the population is approaching critically low levels (Fig. 2.4).

Figure 2.4  The trace of mountain hare harvest density is quite telling. From the early 1940s to 1978 the species was the most common game in Sweden. In 1944 the mountain hare was the most common game in the forest habitat. Its harvest density exceeded the harvest densities of the moose and roe deer ≈ 18 times. The moose matched the number of harvested mountain hares in 1978 and the roe deer became the most common game after 1988 (see the gray lines). In 2023 this relationship was inverted, with roe deer harvest densities ≈ 11 times higher. The color coded sections suggest some ecologically important episodes. The blue shaded sections in the plot suggest that the mountain hare has been at or near equilibrium with the carrying capacity (K) from 1944 to 1979 and for a shorter while between 1984 and 1988. The ruggedness of the blue shaded sections indicates random variation in vegetation quality between years. Compare this ruggedness with the smooth undisturbed trends in the green and yellow sections. The fast growth in harvest densities was initially promoted by an increasing number of re-vegetated clear-cuttings in the mid-1960s and a lagged effect of intensive moose browsing that bolstered forage availability for the mountain hare. The roe deer (gray line) reacted somewhat later, with an unprecedented peak in harvest density about 26 times higher than in 1944. Finally, the red section shows a slower but steady and alarming decline with shallow yearly swings. This observation is crucial, as it may indicate an ongoing change in mountain hare habitat.

The long-term harvest density of the mountain hare may at first appear irregular and inconsistent. However, when viewed in relation to forage availability, a clear and coherent pattern emerges, reflecting how the species responds to changes in carrying capacity.

The mountain hare timeline is further divided into five stages (A–E), representing distinct ecological episodes (Fig. 2.5). Stage A reflects a gradual decline in carrying capacity as populations of moose and roe deer became less affected by random effects and began to increase in harvest densities. Stage B marks the era when modern forestry practices began to temporarily improve field-layer vegetation but the mountain hare still remained influenced by seasonal variability in forage abundance. Stage C represents the full effect of enhanced forage production, supporting unlimited growth as long as populations roam below or at carrying capacity. When forage does not limit physiological needs, offspring production is at its maximum and thus at maximum growth rate. After a short stay at or near equilibrium densities, a rapid fivefold decline in harvest densities followed that lasted for nine years as a result of overexploited forage resources (Stage D). Stage E sheds light on serious long-term consequences of sustained forage depletion, where the field layer fails to recover and mountain hare populations remain at very low levels that barely respond to seasonal variation in vegetation quality. The weakened response to seasonal variation and steady decline may relate to a changing vegetation in mountain hare habitat. A study on the effects of fenced fallow deer reports substantial changes in the composition of vascular plant species after four years of grazing. If the habitat has changed and no longer fully supports the mountain hare, the slow decline may indicate that this native species is in peril.

Effects of Seasonal and Climatic Conditions

As we have stressed above, the variation in harvest density ruggedness may reflect yearly random effects of ambient climate. We can see that the amplitudes are either large or small (Fig. 2.5). That led us to suggest that the ruggedness may reflect changes in habitat carrying capacity rather than just being errors in messy annual reported bag sizes. If messy data was the reason, we wouldn’t have the distinct difference between yearly swings of the amplitudes before and after 1979. So our reasoning is that the amplitude swings have to do with variation in availability of forage. One reason for such variation may be the duration and quality of vegetation seasons. The beginning of the vegetation season is defined as the first five consecutive 24-hour spring mean air temperatures above 5°C, and the end as the first five consecutive 24-hour autumn mean air temperatures below 8°C for each year. This provides a visual overview of the onsets and terminations of vegetation seasons from 1941 to 2016 (Fig. 2.6).

The most important aspect of the vegetation season is how long it can support growing conditions for herbaceous and woody plants. The duration of a vegetation season is simply how many days there are between onset and ending dates. To facilitate a better visual comparison, both season length and mountain hare harvest density were scaled (standardized). These scaled values were plotted by year, revealing a very telling pattern (Figure 2.7). Between 1941 and 1979, harvest densities and season durations seem to follow a similar track in magnitudes. Longer seasons yield larger bag sizes and vice versa for shorter vegetation seasons. However, from 1980 and onward, harvest densities and season durations seem decoupled. Harvest densities have become independent of the duration of vegetation seasons. The most likely reason for this is that forage conditions have changed drastically as a result of sustained overbrowsing and the use of herbicides to suppress deciduous vegetation.

These changes encompass three different episodes: rapid population growth, rapid population decline, and finally a slow but steady deterioration of the habitat suitable for mountain hare. What this suggests in ecological terms is that actual carrying capacity has been pushed beyond any random seasonal influence. The effect of re-vegetating clear-cuttings created unlimited forage conditions until reaching the carrying capacity, resulting in a rapid increase in population densities. At such densities, the depletion of forage caused a fast decline, far too rapid to be compensated by any seasonal effects. The last stage for the mountain hare seems to depict a changing habitat. The shallow amplitudes, despite favorable seasonal effects, suggest that the composition of suitable forage vegetation for the mountain hare has changed. Thus, the main take-out from Figure 2.7 is that any seasonal effects are only visible when a population is at or near equilibrium densities. That is when populations match carrying capacities, and any random swing in the length of the vegetation season causes a response in harvest density.

A closer look at the correlation between the duration of the vegetation season and mountain hare harvest density, confined to the leftmost section in Figure 2.6, indicates that habitat carrying capacity is connected to realized harvest density. Figure 2.7 shows that timeline. The relationship between harvest density and season duration is striking. A correlation test confirms that this relationship (r = 0.40) is statistically reliable (p = 0.01). The same correlation, based on the full time series (1944–2016), shows that the duration of the vegetation season decouples from harvest density when populations are not at carrying capacity (r = 0.04). The decline in harvest densities between 1945 and 1955, which we previously related to as an effect of increasing populations of moose and roe deer, may better be explained by a decline in the duration of the vegetation season.

So far, we have only used the length of the vegetation season as a predictor of potential forage availability, which, to some extent, reflects the magnitude of harvest density for the mountain hare. By adding average annual air temperature to a full multiple linear model, we now have additional dimensions that may improve our understanding of the relationship:

Harvest density = Duration × Mean temperature

The multiple linear regression statistics show that neither the length of the vegetation season (Duration) nor mean annual temperature alone has a clear effect on harvest density. However, when combined (the multiplicative interaction term), these have a statistically distinct influence (p < 0.01). In warmer years, a longer vegetation season leads to higher harvest densities. In colder years, a longer vegetation season instead leads to lower harvest densities (Figure 2.9). This effect may explain why, in some years, harvest density does not match the observed length of the vegetation season (see Figure 2.7).

It is important to realize that the length of the vegetation season and yearly mean temperatures are not the only factors that matter in explaining the quality of a single vegetation season. Including variables such as rainfall, number of sunshine hours, and snow cover would likely improve the model’s ability to explain variation in harvest density.

As of now, the model explains 36.2% of the variation in harvest density (R² = 0.362). We look forward to including these factors in future analyses, as access to such data would likely improve the results and give a more complete picture of harvest density based on encounter rate.

Relevance of fox predation on mountain hare

Figure 2.10 highlights a key difference between mountain hare and fox in the magnitude of their harvest density amplitudes. The mountain hare exhibits large, asymmetric swings—rapid increases followed by steep declines—indicating a population tightly constrained by changes in carrying capacity. In contrast, fox harvest densities show lower amplitudes and smoother trajectories, consistent with a generalist predator less directly limited by field-layer resources.

A randomized re-sampling of oscillation amplitudes shows that fluctuations in mountain hare harvest density are roughly twice as large as those observed for fox (0.54 ± 0.119 SE versus 0.25 ± 0.040 SE; p < 0.05). This clear difference indicates that mountain hare abundance responds more strongly to changes in climatic factors than fox abundance. The mountain hare is more closely linked to variation in forage availability than to fox predation.

However, previous studies point to sarcoptic mange as an important effect of the coinciding rapid increase in harvested mountain hares. It is fair to say that the decline in fox populations, to some extent, has contributed to the observed hare peak in harvest density but the main reason was improved foraging conditions. If mountain hare were a preferred prey, we would not expect fox densities to increase while mountain hare were already entering a phase of decline in the late 1980s, followed by a more shallow and irregular decline from 1996 onward, despite a period of generally prolonged vegetation seasons between 1997 and 2016. This suggests that the habitat no longer supports higher mountain hare densities.

Summary

• The mountain hare is not managed in a traditional meaning by preset numbers. Instead, harvest success is based on encounter rates. To put it plainly - bag size depends on abundance.

• The variation in encounter rates primarily reflects variation in foraging conditions.

• Variation in foraging conditions mirrors the quality of vegetation seasons and thereby the reproductive output.

• Fluctuations in harvest densities are only detectable when population densities are at or near carrying capacity. When populations are far below or above carrying capacity any seasonal variation in forage abundance will be hidden.

• The effect of variation in length of vegetation seasons also depends on yearly mean air temperatures.

• The mountain hare shows four main events between 1941 and 2023. Phases where at or near equilibrium with current carrying capacity, substantially below carrying capacity, substantially above carrying capacity and finally a phase where it barely responds to actual quality of vegetation seasons.

• The phase of rapid increase was mainly an effect of changed forestry routines and to some extent an effect of the outbreak of sarcoptic mange among foxes.

• The extremely rapid decline in mountain hare coincided with an equally fast paced increase in the roe deer population.

• The last phase (from 1997 onward) is alarming. In spite of steadily prolonged vegetation seasons, the mountain hare population keeps on declining. This suggests a habitat that no longer supports viable populations of mountain hare.