Landscape-related differences in diet, food supply and distribution pattern of the pine marten, Martes martes in the transitional mixed forest of northern Belarus

1. What they did

The researchers compared pine martens in two areas of natural forest in northern Belarus that differ mainly in their soil. In the Gorodok “clay” area, rich soils support varied, productive forest with plenty of prey. In the Polotsk “sandy” area, poor dry soils carry mostly pine forest with sparse ground plants, and richer habitats are found only along the valleys of glacial lakes and rivers. They asked how diet, numbers and distribution of pine martens differ with the food on offer, between landscapes and between seasons.

• Diet. They analysed 1222 pine marten scats, mostly found on marten tracks or at known latrines, collected over seven years (1995 to 2001) in the “clay” area and five years (1995 to 1999) in the “sandy” area. They identified 3095 prey items and other foods from teeth, hair, feathers, bones, scales, insect remains and plant material.
• Food supply. They snap-trapped small mammals in the main habitats (about 21,000 trap nights), counted hare and wild ungulate trails in the snow, and counted overwintering perching birds along set routes.
• Habitat use. They radio-collared seven pine martens in the “clay” area (7549 locations between May 1996 and February 2001), and in both areas followed marten trails in the snow, measuring how much of each trail ran through each habitat.
• Numbers. Twice each winter, early and late, they counted marten trails crossing standard routes in the snow: twenty winters (1984 to 2004) in the “clay” area and seven (1997 to 2004) in the “sandy” area. They turned the counts into densities with Priklonsky’s formula, adjusted with daily travel distances measured by snowtracking and radiotracking.

2. What they found

Food supply

• The bank vole was the most common small mammal in both areas (70% and 82% of the small mammal community in autumn). Small mammals were about ten times more abundant in the “clay” area (27.1 versus 2.8 per 100 trap nights).
• Small perching birds were 3 times, hares 11 times and wild ungulates 14 times more abundant in the “clay” area.
• In the “sandy” area, prey was concentrated along streams and glacial lakes; in the “clay” area it was spread much more evenly.
• Bank vole numbers changed with the seasons but showed no cycles of outbreaks and crashes.

Diet

• “Clay” area, warm season: a varied diet, mostly mammals (47.7% of biomass eaten), especially bank voles (27.5%), with birds (25.1%) and berries and other plant food (13.6%). Shrews were common but rarely eaten.
• “Clay” area, cold season: similar, but carrion rose about fourfold (to 21.6% of biomass) and birds fell.
• “Sandy” area, warm season: mostly berries (37.8% of biomass in the text, with blueberries 24.0%) and rodents (23.5%), with more insects and reptiles than in the “clay” area. The diet was narrower.
• “Sandy” area, cold season: dominated by mammal carrion (61.2% of biomass), about 84% of it from wild ungulate carcasses. The diet was very narrow.
• Pine martens showed no preference for any rodent species; they ate them in proportion to what was available.

Numbers

• Early in winter, an average of 7.6 marten trails per km per day crossed the routes in the “clay” area, against 0.5 in the “sandy” area. This means about 7.1 pine martens per 10 km² of woodland in the “clay” area (5 to 11 between years) and 1.2 in the “sandy” area (0 to 6).
• Numbers varied much more from year to year in the “sandy” area, and the two areas did not rise and fall together.
• In both areas, late-winter marten numbers were closely linked to bank vole numbers the previous autumn. In the “sandy” area they were also linked to wild ungulate numbers.

Habitat use

• Pine martens mostly used forest (70.0% to 76.5% of activity), favouring spruce and mixed spruce forest over 60 years old, forest edges next to open ground, and stream banks and lake shores.
• They largely avoided recent clearcuts and pine bogs, and mostly used black alder swamps, dry pine woods, dry meadows and grassy marshes less than their availability.
• In the “clay” area, their habitat use matched the make-up of the woodland. In the “sandy” area it did not: there they lived mainly in the valleys of lakes, rivers and brooks.
• In the warm season they used older clearcuts with regrowth four times more, and swampy areas less, than in winter.

3. What they concluded

• In this transitional forest the pine marten is a generalist. They hunt rodents, birds and other prey of suitable size, and also gather fruit (mostly berries) and scavenge ungulate carrion.
• In the poor woodland on sand they specialise: on carrion in the cold season and on berries in the warm season.
• The richer food supply on clay supports about six times more pine martens, spread fairly evenly through the woodland. On sand they live mainly in the valleys.
• Autumn bank vole numbers drive pine marten numbers through the winter. Where prey is scarce, wild ungulate carrion is also crucial to winter survival, shaping numbers in late winter just before the breeding season.
• Compared with Białowieża Forest (Poland and Belarus), martens in northern Belarus ate few Microtus voles, probably because these voles are scarce in the forest here and martens avoid foraging in open habitats.

4. How tracking was or was not used

Tracking was central to this study:

• Scat collection: most scats were found on pine marten tracks or at known latrines.
• Snowtracking (trailing): in both areas the researchers followed marten trails in the snow and recorded the length of trail in each habitat (109 km of trails from 22 animals in the “clay” area, 88 km from 17 animals in the “sandy” area). Snowtracking, combined with radiotracking, also gave the daily travel distances needed to estimate density.
• Track counts: marten trails crossing standard routes, counted twice each winter for up to twenty winters, gave the abundance index (trails per km per day since the last snowfall). Hare and wild ungulate trails were counted the same way to measure food supply.

5. Tracker or observer certification

Not reported. The paper does not say who collected scats, followed trails or counted tracks, or how their identifications were checked. The diet, habitat and density results all rest on correct identification of pine marten tracks and scats.

6. Other technologies, skills or innovations

• Radio collars (Biotrack Ltd.) and receivers (Telonics Inc.); martens caught in wooden box traps.
• Microscopic identification of hair in scats, and biomass estimates using digestibility coefficients.
• Snap trapping to measure small mammal supply.
• Priklonsky’s formula to convert winter trail counts into densities, adjusted for local snow conditions.