The Rodents Population Cycle In Northern Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The rodents cycle is one of the important subjects in the studies of small mammals. It is necessary to study the population abundance each year for understands it. However, it is essential to study also different parameters, as the influence of the season and the year, the sex and age, and what is the ecology of these rodents. We studied these parameters by using snap trapping during four sessions in June and September 2011, 2012, and we concentrated on dominant specie: The bank vole (Microtus agrestis). We expected a lower proportion in 2012 but we found in September an important proportion of female as in 2011, whereas there was a large lower of rodent's proportion in June 2012. These observations might be explained by several factors as the predator-prey interaction, the snow cover, the quality and quantity of food, female behavior and diseases or does the cycle of the rodents becomes irregular?


Study area:

The study area for the snap trapping is a part of Øyer, a municipality in Oppland country (Figure 1), with a total size of 640 km2 km². This area is located in the south-east part of Norway.

The traps were filed in this Øyer area and in 23 transects that were of different kind. Indeed, they were in different altitudes: 300 to 1100 meters, in different type of landscape: open, forest or bush. And different trees and bush layers: deciduous, juniper, pine, spruce and also Field layer: bare ground or grass, berries, herbs, lichens or mousses.



Figure 1 a, b: Oppland Position in Norway (a) and the traps position in Øyer (b)

Snap trapping:

The snap trapping lasted for four sessions for three days in June, September 2011, and June, September 2012. We have as a first step set out traps in the first day. These traps were made with carrots with peanut butter for enticed the rodents and attached to the releaser; except in June 2011 because we used a small piece of rope soaked in soya oil, so perhaps this may be affected the results. We added a roof for protected the birds and we have fixed the traps in the soil. These in each area defined previously and containing one transect of 10 traps separated of several meters. Each position of traps that filed was registrated in the GPS for uploaded coordinates.

The next day, we have checked each trap and recovered the trapped rodents. At the same time we noticed if the traps were released with the carrot missing or not, if we caught animal or the trapped was moved, untouched or absent. The lure was also remitted in place. Thereby the next and last day we able to check again the traps and wrote of the same way the notes concerning those. Each animal caught was retained in a plastic bag with the date, numbers of the transect and traps, and at the day's end in the freezer.

As a second step, we did the identification of the trapped rodents. We detailed the specie, sex, age (juvenile or adult), the weight and height of the body and the tail. For the adult female we have specified if they had a litter and how many babies. As for the adult male we specified the testes length and the presence or not of the tubili, epididymis.

For the sex identification we were help with the sexual characteristics: penis, anus, scrotal testes, anus, perforate vagina, clitoris, nipples, and then we made a dissection for confirm or proceed the analysis. And we identified the specie with books references.


Eleven different species of small rodents were trapped: Myodes glareolus, Microtus agrestis, Microtus peconomus, Myodes rufocanus, Arvicola terrestris, Apodemus sylvaticus, Apodemus flavixollis, Sorex spp., Neomys fodiens, Myopus schisticolor, Lemmus lemmus, Sicista betulina. The bank vole (Microtus agrestis) is the dominant specie. We find it in general in grasslands and forest clearings. It is an herbivore and it constructs a network of runways among the grass clump on the surface of the ground. Its feeding is composed usually green vegetables, in majority grass and a high proportion of dicotyledonous plants , forb, and others food items like berries, seeds, lichens, fungi and sometimes insects ( Hansson 1971; Viro & Sulkava 1985; Ferns 1976; Myllymäki 1977 ).The majority of breeding season is in the early summer months and later in the autumn (Myllymäki 1977).

Data analysis:

We defined the number of animals that were caught in the four periods and for which species and the catch per unit effort. Then, we studied the difference in number of catches for the different seasons and years with a logistic regression. We used the p-value for known if the variables are significant or not. We concentrated the rest of our study on the bank vole, the difference of sex and age in the caught and still for the seasons and year. We did a logistic regression with the sex and age as response. Finally, the habitat selection of the bank vole, we used also a logistic regression, with as first step a stepwise backwards selection with the different habitats variables and we excluded variables with P > 0.05.


The first studies concerning the rodent's population cycle started in Finland and it has been demonstrated after many statistical analyses data (Henttonen et al. 1995, Hanski et al. 1999). During the last century, there was a lot of opinions concerning this cycle, but the existence of a regular and universal cycle appears as an evidence especially at high latitudes (Krebs and Myers, 1974)

In Northern latitudes, the rodent's population is characterizes by a cycle periodic of 4-5 years intervals (Hansson & Hanttonen 1985) and fluctuates in higher and lower proportion. A lot of study have been conducted for try to explain this cycle and many hypothesis were been suggested. That's why it is necessary to harvest data each year to continue the study of this cycle.

A recently studies revealed at first in Finland a change in the cycle, more stable and irregular, in particular a lengthening of the period, and also in others part of Fenniscandia (Lindstrom & Hornfeldt 1994; Steen, Ims & Sonerud 1996; Laaksonen, Korpimaki & Hakkarainen 2002). For example in central and southern Fennoscandia the cycles' periodicities change with 7 years periods whereas no study of this kind of long cycles exists for the moment (Sundell et all. 2004), but it was too recently for concluding of a real change in the cycle.

The bank vole is the dominant species in our area, that's why we interested in particular at this specie, its abundance and also its ecology, in particular its favorite habitat. In 2012 with the regular cycle of the rodents, we expected a low period of abundance population than in 2011.This support by one recent study did previously of red fox diet, which suggests in their scats a decreasing proportion of small rodents. But our results didn't show that. This irregularity of the cycle is the principal subject of our study.


We caught 154 animals, in the proportion shown : 80% were a bank vole, 8% a shrew, 6% a Grey red-backed vole, 2% a Yellow-necked mouse and 1% a Field vole, Wood lemming, Root vole, Lemming and Wood mouse (Table 1). And (Table 2)

Number of caught



Bank vole



Field vole



Grey red-backed vole






Root vole






Wood lemming



Wood mouse



Yellow-necked mouse



Total général




June 2011


September 2011


June 2012


September 2012


Table 1: number of animals caught and for which species

Table 2: the catch per unit effort per species

We analyzed the difference in number of catches for the different seasons and years.

The season is a significant factor for the number of caught (p <0, 01, chi² = 16,11) and also the interaction between the season and the year (p = 0,001, chi² = 10,2283), but the year wasn't a factor significant (p>0,005, chi²= 1,36) (table 3).We might see the effect between these two factors with the Figure 2 . There wasn't a large difference between June and September in 2011, but in 2012 there was a great difference between these two seasons. In fact, in June 2012 the caught was very lower than September 2012. Finally in September 2012 the caught was about in the same proportion of 2011, quite high, even higher than 2011.














Table 3: logistic regression for the number of caught response with the season and year

Figure 2: The effect between the season and the year for the caught

The bank vole was the dominant species, then, we continue our study with this species. We studied the proportion of female and male, juvenile and adult (Figure 3). As regarding the breeding, in general in summer months our results showed those expected. In fact, in June for the two years, the adults were more important and in September it was the juveniles, because of the generation cycle. But in September 2012 there was a surprisingly high proportion of adult female. We also noticed in June 2011 a tendency of high proportion of females than male. Furthermore, as view previously we might see in June 2012 an important lower of catches proportion and a totally absence of male juvenile.

Figure 3: The bank vole proportion of male adult (M- A) and juvenile (M- J), and female adult (F-A) or juvenile (F-J) for the four sessions.

To further the analysis of the difference in sex and age, we did a logistic regression. The year and the season weren't significant for the sex (year : p=0,21, chi² =1,53 and season : p=0,13, chi² =2,27), also for the year/season interaction (p=0,16, chi²=1,98) (Table 4); but when we observed the effect plot we notice one difference in September between 2011 and 2012 with a positive slope (Figure 4).

Then, we studied the effect of the year on the sex only for September. We found a correlation not significant (p = 0,07, chi² = 3,23), but even so we might say there was a tend to a low proportion of female in 2011.












Table 4: logistic regression of the season and year effect on the sex

Figure 4: The interaction effect plot of year and season on the sex (with sex01 = female)

As regards the Age, the season was significant with a negatively slope (p= 1,24e-10, chi² =41,404) (Table and Figure 5), which confirms the generation cycle June/September noticed before with the figure 2. And year and year*season weren't significant (p > 0.05) (Table 5).













Table 5: logistic regression of the year effect on the sex in September

Figure 5: effect plot of the season on the age

Finally we studied the habitat preference. We described for each habitat the number of trap had been set and the percentage of those having trapped a bank vole (Figure 6 a,b,c). We have did a first conclusion on where they most likely to be caught and thereby what are their favorite habitat and their diet. In first step we may say they prefer an open and bush habitat. Then, if we interested for the tree layer for begin, the captures were more important in tree layer with spruce or none tree. For the field layer, the catches was in great majority in grass. Finally the bush layer, the caught was more divided, but the dominants layers were the spruce and the deciduous bush.

After these results we did the statistics. We tested for each habitat if there was a relation with the caught of the bank vole, (table 6). The altitude, the dominant habitat and the tree layer were significant for capture the bank vole (p<0,001).

There was another factor in the dominant habitat: the altitude. For the higher altitude (1000 and 1100 meters) none bank vole was caught, then we might conclude they doesn't live in this type of habitat. The most altitude was in 500 meter and we noticed in 400m a lower proportion than the others altitude (Figure 7).

Figure 6 a, b, c, d: the percentage of bank vole trapped for each habitat, dominant habitat (a), tree layer (b), field layer (c), bush layer (d)















Table 6: logistic regression of the relation between the habitat and the caught of bank vole

Figure 7: the caught of bank vole according to the altitude


The foundation of our study was the irregularity of the rodents' cycle and the favorite habitat of the bank vole.

In fact, our case the results showed an irregularity. It is important to noticed an uncertainly of the results obtained caused by the different lure used in June 2011. It had perhaps affected the caught, but we don't know how.

The rodents cycle of 3 to 5 year periods, expected in 2012 a lower proportion of rodents' population than 2011 but in general we obtained a number of caught not really affected by the year which it was not a significant factor, although rather by the season and the interaction of year/season, even if we have noticed in June 2012 a great lower of the proportion of caught than 2011. But in this same year, in September the caught was higher and approximately in the same proportion that the caught in 2011. If we use the observation of the bank voles proportion which the dominant specie in our case, we reach the same results. In fact, in June 2012 the caught of bank vole was very low. In addition, once more the results revealed in September 2012 a surprisingly proportion of adult female bank vole (Figure 3).

Our results showed that the cycle seem not as expected, it is earlier to draw conclusions and it is necessary to continue the study and collect more data. But we might begin to make some assumptions.

As regarding the notable low proportion in June 2012, it was an observation in accordance with the low year of rodent's proportion expected in 2012 than 2011. But the high proportion found in September 2012 show it would seem 2012 is not a low period than 2011 as expected. That might be explain by several factors.

In the first instance we might suggest several hypotheses. Either there is a low period in 2012 as expected and September 2012 is an isolated case or, there is not a low period and June 2012 is particular.

One of these factors is the predator-prey interaction. That suggest the predators might impact the rodent's cycle. It is a factor already study for explain the multiannual cycle of the rodents, but this factor might also affect the cycle more in particular. A recently study showed that the specialist predators might change the cycle multiannual cycles to annual fluctuation and the generalist might changing the period and the amplitude of these fluctuations (Korpimäki et al. 2002). Moreover, the generalist predators might stabilizing the rodent population and also turn the cycle to a non-cycle (Hanski, Hansson, Heikki & Henttonen 1991) Then, we may be suggesting that this factor would might affect the proportion of rodents in our case. Therefore it would be necessary to study the population of these predators.

In addition to this factor, the variable of snow cover might play an important role. Indeed, the rodents are more vulnerability to predation when there is an important depth of snow, as with the predator red fox (Lindström & Hörnfeldt 1994).We might hypothesis that in the winter 2011 the snow cover affected the rodents and then in june 2012 the population was low.

We showed the season is a significant factor for the proportion of rodents. It is important to study the difference in season and why this affected the population of rodents. Other studies have provided some answers. The effect of food on the density dependent structure of the rodent. Indeed, there exists a multiannual oscillation in plant quantity and quality. The availability of food might a dynamic variable which might change the properties of the cycle (Turchin & Batzli 2001).

In Sweden, Hanson (1971) studied the effect food and the fructification on the granivorous and herbivorous species. He suggested that the rodents' fluctuations might be affected by the plant production. In addition, another study following this idea, Laine and Henttonen (1983) conducted a study on the role of plant production in microtine cycles in nothern Fennoscandia. They studied the variation in food availability of microtine in interaction with the climate and the plant growth. And they suggest that the factors beginning the cycle decrease are the qualitative and the quantitative changes in the food availability of the rodents. In addition, they showed that in northern latitudes, plants needs several summers to complete the cycle and accumulate reserve. With this climate and the plants cycle factor we might make an assumption, the needs food for the rodents (berries, seeds ect) was limited and insufficient, that might be affected the proportion of rodents in june 2012. The cover snow might play also an important role for the food availability (Hansson,1971).

Furthermore another factor has not been studied in our case, the distribution of female depends of the food availability (Jonsson, Hartikainen, Koskela, & Mappes, 2002).The female because of the energy necessary during the reproductive cycle are needs an important quantity of food (Jonsson, Hartikainen, Koskela, & Mappes, 2002). The female bank vole usually territorially (Koskela, Mappes & Ylönen 1997) might move more place if the foods are not sufficient. In this case, this might explain the high proportion of adult female in September 2012 by displacement.

The diseases epidemic is also a factor to take into account. The small rodents have a large number of macros parasites and micro parasites which are responsible in high mortality and diseases (Hanski & Henttonen ). For example, a study conducted in England showed a high mortality of rodents after an epidemic in 1934, because of disease encephalitis caused by Toxoplasma (Ferns 1976).

Finally, as regarding the habitat dominant of the bank vole, the result was those expected with their ecology and their preference of food. We might assume that we couldn't found the bank vole in high altitude because of an inhospitable climate and the majority of trees and others plants that provides their needs food, aren't grow in this kind of climate, but rather between 300 and 900 meters and more in 500m where we found the most bank vole. Likewise, the small rodents are among the species where landscape configuration might have an influence in the population dynamics because of its inhabiting habitat patches with various size, location and quality. Moreover the social interactions and mortality during the dispersal may affect the population (Hansson, 2001).