ISSN 10674136, Russian Journal of Ecology, 2013, Vol. 44, No. 4, pp. 336–345. © Pleiades Publishing, Ltd., 2013. Original Russian Text © Yu.N. Litvinov, S.A. Abramov, V.V. Panov, 2013, published in Ekologiya, 2013, No. 4, pp. 300–309.
Significance of Rodent Population Dynamics for the Formation of LongTerm Community Structure Yu. N. Litvinov, S. A. Abramov, and V. V. Panov Institute of Systematics and Ecology of Animals, Siberian Branch, Russian Academy of Sciences, ul. Frunze 11, Novosibirsk, 630091 Russia; email:
[email protected] Received June 28, 2012
Abstract—Changes occurring with time in the structure of communities are evaluated and strategies of com munity functioning are described using the example of murine rodents from three areas of Western Siberia that differ in landscapes and geographical conditions. Many rodent species are common to all communities included in analysis. The populations of these species have different ranks in the dominance structure of com munities, depending on the natural climatic conditions of the landscape, and differently influence the com position of communities and time course of changes in their structure. Keywords: communities, rodents, population dynamics, community structure, phase portraits, dominants DOI: 10.1134/S1067413613040097
Rodent communities (sympatric population assemblages) as components of the animal population have peculiar landscaperelated features of species composition and quantitative composition, which are interconnected and vary in space and time, depending on basic natural factors. Changes occurring in such communities in time from year to year manifest them selves in the following processes: population dynamics of the total community abundance, changes of domi nants and reorganization of the structure of dominance, and changes (fluctuations) of information parameters. Analysis of numerically dependent population parame ters of the rodent community in a particular season makes it possible to reveal the main principles that determine the structure and function of the community as an integral system (Hansson, 1993; Williams, Marsh, and Winter, 2002; Yoccoz and Ims, 2004). Rearrangements in the species (population) struc ture of the community occur in a certain time sequence. The degree and smoothness of the responses of the community (with all species it includes) as a dynamic system to external influences in different years may be reproduced by the method of phase por traits, or stability portraits (Fedorov and Gil’manov, 1980; Ekologicheskie sistemy, 1981). This method reveals the pattern of system dynamics upon deviation from the stationary state (the state with zero rate of changes in the parameter). The degree of the influence of different factors on the composition and structure of communities can be estimated by factor analysis, which recognizes groups of factors that have an effect on temporary changes in communities and act in dif ferent directions relative to each other (i.e., are not correlated).
The purpose of this study was to estimate and char acterize the longterm structure and various forms and ways of functioning in rodent communities of three key areas with different landscapes and natural cli matic conditions by analyzing longterm dynamic processes (population dynamics) using mathematical methods (analysis of population dynamics, informa tion indices, ranks of species in the dominance struc ture, phase portraits of particular species in time, and information parameters of the community). MATERIALS AND METHODS This study was performed in three areas of Western Siberia differing in their landscapes and geographic conditions: the forest–steppe zone of the Baraba Low land (below, referred to as Northern Baraba); ribbon like pine forest on the right bank of the Ob River, in the parkforest zone of Akademgorodok, Novosibirsk (Akademgorodok); and Siberian stone pine–fir taiga forest with an admixture of deciduous trees near Teletskoye Lake in the environs of the Teletskoye Lake Station, Institute of Systematics and Ecology of Ani mals, Siberian Branch, Russian Academy of Sciences (Teletskaya Taiga). The rodents were captured by the standard method of trench cone pitfall traps (Naumov, 1955) from late June to early September. On the whole, field work amounted to 37000 trapdays in Northern Baraba, 27968 trapdays in Akademgorodok, and 25890 trap days in Teletskaya Taiga area; a total of 15520, 5400, and 5000 rodent individuals, respectively, were trapped and analyzed.
336
SIGNIFICANCE OF RODENT POPULATION DYNAMICS
Individual species in the communities were numer ically characterized by the parameter of annual aver age relative abundance per 100 trapdays. Shannon’s diversity index (H) was calculated from proportions of species (Begon, Harper, and Townsend, 1989). The phase portrait method is a convenient way to obtain an illustrative picture of spatial movements of objects comprising a system and analyze their trajecto ries in certain coordinates. We analyzed the average annual parameter of species abundance (for dominant or key species) and Shannon’s diversity index as a characteristic of community structure (Tereshchenko and Verbitskii, 1997). The main idea of the method (Efimov, Galak tionov, and Shushpanova, 1988) is transformation of a onedimensional time series into a multidimensional series with vectors that characterize the history of the process at moments separated by equal time intervals. Combining these vectors into a table yields a matrix, which is then processed by the principal component analysis and transforms into a new matrix of the same size. Each new component is a new time series repre senting a linear combination of previous time series; the behavior of any such component relative to any other component can then be analyzed. In this case, each state is represented by a point on a plane formed by the corresponding pair of components. When such points are joined in consecutive order (e.g., by cubic spline smoothing), they form the trajectory of the pro cess projected onto the plane of this pair of compo nents, or a phase portrait. Despite its orthogonality, not all components are independent, and shifting them relative to each other can reveal very high corre lation coefficients, which reflect the internal structure of the processes that have generated the series (Efimov and Galaktionov, 1983). This method is applicable to any time series, does not require stationarity (unlike, e.g., spectral analysis), and reveals trends (if any) without any assumptions about their nature or form (Bobretsov et al., 2000; Efimov and Kovaleva, 2007). Each matrix of longterm data on the abundance of species in each of the three studied areas was also pro cessed by the principal component analysis (with mean correction and normalization). The method allows either years or populations of particular species to be classified into groups with similar abundance dynamics (Williamson, 1975). On the basis of interde pendence between each of the components and vari able environmental factors, the resulting classification can well be matched with similar changes in the annual abundance of species. Although the results dis cussed here remain hypothetical, their classification allows more profound comprehension and representa tion of the phenomenon studied. RESULTS AND DISCUSSION Longterm dynamic processes taking place in com munities of wild rodents serve to support the existence of the entire systems of these communities. At the same RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
337
time, the populations of those species that are repre sented by large numbers of individuals are of the great est importance in the structure of any community. The characteristic aspect of any cenosis is determined by those animal species (key and dominant ones) that comprise the background elements of the cenosis. A complex of landscape elements creates conditions favorable for the living, reproductive success, and high abundance of such species; on the other hand, the ani mals themselves, due precisely to their ability to reach high levels of abundance, have a considerable effect on other elements and the entire aspect of the landscape (Kuzyakin, 1962). The key species, compared to domi nants, have a stronger influence on the structure of the community (Primak, 2002). Dynamics of the rodent community structure in Northern Baraba. The rodent community of the Northern Baraba area comprises nine species (exclud ing synanthropic rodents). One of the most important components of this community is the European water vole (Arvicola terrestris L.). Population outbreaks in this key species of the community take place approxi mately every 10 years and coincide in time with the “moist” phase of the climate (Panteleev, 1968; Maksi mov, 1984), during which large areas of different hab itats are occupied by populations of the water vole. During the “dry” phase, the abundance of this species decreases to a minimum, with local populations of this species rarely occurring only in river floodplains and other waterlogged biotopes. In some years, the water vole dominates in the rodent community of Northern Baraba, but on the longterm average it is not a domi nant species in the area. This is why we regard this structureforming species as a key species of this com munity. The dynamic processes in the rodent community of Northern Baraba have the following aspect (Fig. 1a). The total abundance of all species in the community is usually higher in years with strongly increased abun dance of the species that are dominant (1981, 1985, 1986). Medium and low values of total abundance are recorded in years with increased evenness of species distribution in the community, indicated by high val ues of Shannon’s diversity index (H) (1978–1980, 1984–1985, 1988–1990). A population outbreak in dominant species, especially the water vole, followed by decline in the next year in a strong decrease in the Shannon’s index (1981, 1986) (Fig. 1a). The phase portrait of the population of the water vole, the key structureforming species in this rodent community, is an almost perfect circle (Fig. 2a, plot A). Years corresponding to different phases of abundance are grouped in different parts of the graph. The factors that determine changes in numerical parameters of the species from year to year account for the smooth time course of these parameters and contribute to the cyclic movement of the system. These sustained oscillations give evidence for stable population cycles, which have been known previously (Maksimov, 1984). 2013
338
LITVINOV et al. (а)
Abundance, ind./100 trapdays
H 2.5
100 2.0 80 1.5
60
1.0
40
0.5
20
0 1990 9 H
0 1978 1980 1 2 3
1982 4
1984 1986 1988 5 6 7 8
(b)
35
2.5
30
2.0
25 20
1.5
15
1.0
10 0.5
5
0
0 1992 1994 1996 1998 2000 2002 2004 2006 1 9
100 90 80 70 60 50 40 30 20 10 0 1984
2 10
3 11
4 12
(c)
5 H
6
7
8
2.0 1.6 1.2 0.8 0.4
0 1988 1992 1996 2000 2004 1986 1990 1994 1998 2002 1 2 3 4 5 6 7 8 9 10 11 H
RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
2013
SIGNIFICANCE OF RODENT POPULATION DYNAMICS
339
Fig. 1. Annual variations in relative rodent abundance (per 100 trapdays) and Shannon’s diversity index (H) in different com munities. (a) Northern Baraba: (1–3) mice: (1) Sicista betulina, (2) Apodemus agrarius, (3) Micromys minutus; (4–9) voles: (4) Microtus gre galis, (5) Microtus oeconomus, (6) Microtus agrestis, (7) Clethrionomys rutilus, (8) Clethrionomys glareolus; (9) Arvicola terrestris. (b) Akademgorodok: (1–4) mice: (1) Sicista betulina, (2) Apodemus agrarius, (3) Apodemus peninsulae, (4) Micromys minutus; (5–12) voles: (5) Clethrionomys rufocanus, (6) Clethrionomys glareolus, (7) Clethrionomys rutilus, (8) Arvicola terrestris, (9) Microtus gregalis, (10) Microtus oeconomus, (11) Microtus agrestis, (12) Microtus arvalis. (c) Teletskaya Taiga: (1–5) mice: (1) Sicista betulina, (2) Apodemus agrarius, (3) Apodemus uralensis, (4) Apodemus peninsulae, (5) Micromys minutus; (6–10) voles: (6) Clethrionomys rutilus, (7) Clethrionomys rufocanus, (8) Clethrionomys glareolus, (9) Micro tus oeconomus, (10) Microtus agrestis; (11) Myopus schisticolor.
The dominant species in the rodent community of Northern Baraba, the striped field mouse (Apodemus agrarius Pall.), has a phase portrait similar to that of the water vole (Fig. 2a plot B). As shown previously (Litvinov and Panov, 1998), the time course of its abundance in Northern Baraba is in antiphase to that of the water vole and correlates with the time courses of abundance of several other species in the commu nity. The figure shows that the processes of population decline and growth in the population of the striped field mouse are slow, with the growth phase being 3– 4 years long. The phase portraits of subordinate species (such as the bank vole, Clethrionomys glareolus Schreber) (Fig. 2a, plot C) and of species rare in the Northern Baraba rodent community are similar in aspects to the two portraits described above: they show relatively smooth changes in abundance, usually forming 3 to 5year cycles typical of many rodent species (Sadykov and Benenson, 1992). On the whole, the portraits reflect stable oscillations in the average annual abundance of species comprising the community, determined by the stable amplitudes of these oscillations (Romanovskii, Stepanova, and Chernavskii, 1971). The phase portrait of the diversity index (Shan non’s index) for the whole community is a spiral spin ning inwards (Fig. 2Ad). For model communities, such phase portraits give evidence of the movement of the system towards balance with a considerable reset time. It can be suggested that the time cycle of the diversity index is associated with the population dynamics of the key species, the water vole. The decreased range of the oscillations coincides with high and medium peaks of the water vole abundance, which have a considerable effect on the diversity parameters of the whole rodent community of Northern Baraba and serve as a strong structureforming factor for the community (Litvinov and Panov, 1998). Analysis of the annual average values of abundance for nine rodent species by the principal component method yielded the following results (Fig. 3a). The years with increased moisture supply to the area had positive loadings on the first principal component (1981, 1982, 1986, 1987; Table 1); in those years the abundance of the water vole declined. The strongest negative loadings on this component were accounted for by “drier” years, when the abundance of the water vole was low. The position of the species along the first component axis indicates their preference for particu RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
lar biotopes along a drytomoist gradient. The aspect of the rodent community is especially strongly struc tured in years of high moisture supply; this effect is associated with the strong biotopic preferences of the rodent species of this community. The position of the species along the second prin cipal component (35.4% of the variance explained) indicates the longterm average values of their abun dance, from low to high. The loadings of all years (except 1990) on this component were approximately equal and positive. The species especially outstanding in its position along the second component is the bank vole (Fig. 3a, point 8). The year with the greatest load ing on this component was 1990 (Table 1), when this species was dominant in the community. Therefore, the rodent community of the Northern Baraba is especially strongly structured in years of increased moisture level in the area. Average rank values (for the period from 1978 to 1990) for rodent species of Northern Baraba (Fig. 4a) show that the dominant group includes species with fairly low average ranks and medium coefficients of Table 1. Coefficients of correlation with principal compo nents for parameters of rodent species abundance in North ern Baraba in different years Principal component Character (year) 1
2
3
1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990
–0.47 –0.48 0.62 0.91 0.85 –0.47 –0.62 0.25 0.87 0.88 –0.42 –0.64 –0.48
–0.61 –0.81 –0.74 –0.34 –0.47 –0.84 –0.55 –0.52 –0.46 –0.43 –0.88 –0.49 0.06
0.49 0.30 –0.03 –0.17 –0.14 0.17 –0.52 –0.14 –0.11 –0.09 0.07 –0.40 –0.87
Proportion of variance, %
41.7
35.4
12.6
2013
340
LITVINOV et al. (a)
A 1984
0
1989
–1
1982
–1
0
C
1985
1
2 PC 1
–2 2
–1 –1
0
1996
0
–1
PC 3
PC 2
1999
1 20072003
1997 2006 2004 2005 2002 1998 1995 2001
0
–1
0
2000
0 1998
2006 1997 2004 2002 1996 2005 1999
0
2006
2007 1998
2000
–1 D
2
1995 1996 2002 2005 2006 1997 2003 2004 2001
1
1
1999 2003
0
PC 3
–1
–1 1 PC 1
0
(c)
A
–1
1 PC 2 2 C
2 PC 2
1992 1988
1986
–1
0
1 1987
0
1992
PC 3
–2
1989 1986
0
1997 2004
1990
1 1990 1987
–1
–1 B
1988
1 0 1991
2007 1996
–2
PC 2
–2
2 PC 2
2005 1998 2002 2001
1995
2000 1999
–2
2
2 PC 1
1
B 2 2001 1995 2003 1 2007
2 PC 1
1
PC 3
PC 2
–1
1982
1988
–2 –1 C
0
1981
1983
–1
–2 1
1984
–2
1 PC 1
1 PC 1
1990 1987 1989
1985
–2 (b)
A 2 2000
0
1986
0
–1
–2 –2
–1
D
1
1986
1981 1987 1989 1988 1982 1983
1984
1984
–2
1
1990
0
1988 1981 1989 1982 1990 1983
0
–1
1987
1983
2 PC 2
1981 1986
1988
1986 1987
1985
1 PC 2
PC 2
1
B
2
1985
1990
PC 2
2
1989
–2
–1
0
1991
1
PC 1
1988
1 0 1987
1986 1991 1990
–1
1992
–2
–1
1989
0 1 PC 2
2
Fig. 2. Phase portraits of longterm dynamics of species abundance and Shannon’s diversity index (H) in rodent communities of (a) Northern Baraba, (b) Akademgorodok, and (c) Teletskaya Taiga. Northern Baraba: (A) Arvicola terrestris, (B) Apodemus agrarius, (C) Clethrionomys glareolus, (D) Shannon’s diversity index. Akademgorodok: (A) Clethrionomys rufocanus, (B) Apode mus agrarius, (C) Sicista betulina, (D) Shannon’s diversity index. Teletskaya Taiga: (A) Clethrionomys rutilus, (B) Microtus oecono mus, (C) Shannon’s diversity index. RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
2013
SIGNIFICANCE OF RODENT POPULATION DYNAMICS (а)
(b) 4
1
5
4
2
1
3
2
7
1
0 9
6 2
9
12 5
8
1
–1
8
–2
10 2
0 PC 2
1 2 3
1
2
11
1
0 PC 2 –1–2
–1
0 PC 1
7
3
PC 3 0
6
PC 3 –1
341
–1
–2
0 PC 1
1
2
(c) 1 9
7 5 8 10 4 2
0
3 11
6
–1 PC 3 –2 1
–3 2 1 0 PC 2 –1
–2
0
–1
1 PC 1
2
Fig. 3. Distribution of rodent species in the space of the first three principal components (PC) of population dynamics. For des ignations, see Fig. 1.
their variation: the striped field mouse, bank vole, and northern birch mouse (Sicista betulina Pall.). These three are followed by a species whose position sharply changes from year to year (the water vole), which fol lows from considerable interannual variation in its sta tus and the high value of the variation coefficient. The ranks of the other species represented in the figure are fairly constant, which is confirmed by their coeffi cients of variation. Dynamics of rodent community structure in the environs of Akademgorodok, Novosibrisk. No sharp population outbreaks have been recorded in any spe cies of this community, although their total abundance fluctuates considerably (Fig. 1b). Rearrangements in the structure of dominance in the community, against RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
the background of relatively high species richness (12 species, excluding synanthropic rodents), are minor, which is indicated by relatively small interan nual changes in the diversity index (H) of the commu nity (Fig. 1b). An even dominance structure is recorded in years with high total abundance of the whole community. The evenness of the community is rather high; therefore, in our opinion, the dominant species are also the key species. The phase portrait of the dominant species, the gray redbacked vole (Clethrionomys rufocanus Sund.), in the plane of the first two principal components is characterized by smooth movement along circular tra jectories of various diameters (which indicate differ ences in the range of the oscillations) (Fig. 2b, plot A). 2013
342
LITVINOV et al.
Table 2. Coefficients of correlation with principal compo nents for parameters of rodent species abundance in the parkforest zone of Akademgorodok in different years Principal component Character (year) 1
2
3
1992
0.60
0.64
0.05
1993
0.66
0.66
0.21
1994
0.74
0.47
–0.16
1995
0.56
0.44
–0.57
1996
0.62
0.46
–0.34
1997
0.67
–0.56
–0.30
1998
0.69
0.14
–0.39
1999
0.82
0.26
0.44
2000
0.76
–0.19
0.07
2001
0.76
–0.39
–0.12
2002
0.68
0.21
0.61
2003
0.88
–0.33
–0.17
2004
0.72
–0.58
–0.05
2005
0.83
–0.42
0.02
2006
0.88
–0.13
0.02
2007
0.64
–0.26
0.58
Proportion of variance, %
52.6
17.6
10.6
The duration of one cycle is about 3 years. The period icity of the action of groups of factors that determine the dynamics is similar from year to year and gives evi dence for the stability of this process. The trajectory of the abundance of the striped field mouse, a subdominant species in this community (Fig. 2b, plot B), is similar to that of the dominant species. The phase portrait of the northern birch mouse, another species included in the dominant group, is mostly a circular trajectory with points (years) more or less evenly distributed on it and a 9year cycle of oscilla tions (Fig. 2b, plot C). In this case, the phase portraits show a community of species with different longterm population dynamics. The longterm phase portrait of Shannon’s diver sity index for the whole community lies in the plane of the second and third principal components (Fig. 2b, plot D) and is shaped as a relatively smooth spiral with a 4year cycle. The evenness of the cyclic changes of the diversity index, which accompany minor rear rangements of the dominance structure, gives evi dence for balanced courses of abundance in all species comprising the community.
Component analysis shows an even distribution of species along the first principal component, from low to high abundance (Fig. 3b). The second component is associated with the years in which the diversity index of the community decreased because of decline in populations of rare species and changes in the domi nance structure resulting from growing abundance of the dominant species (1997, 2001, 2004, 2005). These years had the strongest negative loadings on the second principal component (Table 2). The third component distinguishes the years of high abundance of the Eur asian harvest mouse, Micromys minutus Pall. (2002, 2007) (Fig. 1b). These two years had the greatest posi tive loadings on the third component (Table 2). Many of the species in this community are charac terized by changes in abundance synchronous with those in the dominant group of rodents (Figs. 1b, 3b). This group includes voles of the genus Clethrionomys and mice. The population dynamics in voles of the genus Мicrotus and the water vole are in antiphase with those in the dominant group. It can be suggested that the similarity in the course of changes in abundance between species of the second group is determined by the level of moisture supply, because this group com prises mostly species feeding on vegetative organs of plants or living near water (the root vole, field vole Мicrotus agrestis Pall., and water vole). The group of dominants and species of medium abundance in the rodent community of the Aka demgorodok area has an even distribution of ranks and their coefficients of variation (Fig. 4b). This group includes nine rodent species whose abundance and rank evenly decrease in the plane of these two param eters. Species that are rare in this community have a constantly low rank. Dynamics of rodent community structure in Teletskaya Taiga. This community comprises 11 spe cies. The root vole and the northern redbacked vole (Clethrionomys rutilus Pall.) are dominant, and other species are represented by only a few trapped individ uals (Litvinov et al., 2007). In some years, the abun dance of the northern birch mouse and Korean field mouse (Apodemus peninsulae Thomas) can increase in some biotopes. The annual abundance of different species fluctuates synchronously or in antiphase. In years of high total abundance of the population, the dominant species play a paramount role (Fig. 1c); these species serve at the same time as key structure forming species of the community. The longterm dynamics of diversity index (H) for this community show that this parameter decreases or increases mainly because of changes in the dominance structure, rather than longterm changes in the abun dance of species comprising the community (Litvinov, 2001). Years with relatively high H values (1984, 2002; Fig. 1c) differ in values of the relative abundance of species, but are similar in dominance structure, which includes approximately five dominant species with evenly decreasing average annual abundance. Low val ues of the diversity index typically fall on years with a
RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
2013
SIGNIFICANCE OF RODENT POPULATION DYNAMICS
dominance structure characterized by one or two strongly prevailing species and relatively low propor tions of the other species; sometimes, these are years of high total abundance of the rodent community (1994, 1997, 2003; Fig. 1c). The phase portraits of the two species dominating in the rodent community of Teletskaya Taiga, the root vole (Fig. 2c, plot A) and northern redbacked vole (Fig. 2c, plot B) show asynchronous population dynamics with 3 to 4year cycles. The considerable range of variations in different turns of the spiral in the plane of the two principal components gives evidence of a low stability of the system, which is capable of oscillations at long distances from the initial point. The phase portrait of Shannon’s diversity index (Fig. 2c, plot C) also shows an “irregular” periodicity with cycles of various lengths. The dominance struc ture of the community is subject to serious rearrange ments between years of short periods (three years sim ilar in dominance structure) and long periods (struc ture considerably varying from year to year for 5 years). Analysis of annual average values of abundance for the 11 rodent species of this community by principal components analysis (Fig. 3c; Table 3) has shown that the first three components explain 96.7% of the total variance explained. The distribution of species along the first principal component (78.1% of variance) cor responds to longterm average values of abundance, from low to high. The loadings of all years on this com ponent are large, similar, and positive, indicating sim ilar responses of species to the main factors that deter mine the species abundance in the community. The longterm average parameters of abundance combine all species with medium and low abundance into one group and distinguish two dominant species: the root vole and northern redbacked vole. The greatest positive loading on the second compo nent was that of years with high abundance of the main dominant species, the root vole, and low abundance of the second dominant species, the northern redbacked vole (1991, 1994, 1997, and 2004). By contrast, the greatest negative loading was that of years with rela tively high abundance of the northern redbacked vole and low abundance of the root vole (1985, 1989, 2001; Table 3). The distribution of species along the second principal component (Fig. 3c) indicates their prefer ence for particular biotopes, from fairly dry and for ested to relatively moist. The greatest differences along the second component are found in the two dominant species, the northern redbacked vole and root vole. Therefore, the second component reflects mainly variation in the community structure depending on the abundance of the two dominant species, in accor dance with their different biotopic specializations. The third principal component distinguishes one more species that sometimes reaches high abundance and enters the dominant group, namely, the northern birch mouse (Fig. 3c). The greatest negative loading on this component was that of 2004, the year of an especially high abundance of this species. RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
343
(а)
CV 0.9
9 0.7 2 8 0.5
6 3
1
0.3
4
7 5
0.1 1
2
3
4
5
6
8
7
(b)
0.8 5 2
0.6
7
1
12 4
6
0.4
10
11 0.2
9
3 0
2
4
6
8
8
10
12
14
(c) 9 6
0.4
1
4 0.3
8
10 7
0.2
11 5
0.1 0
2
4 6 Average rank value
2
3 8
10
Fig. 4. Average rank values and coefficients of their varia tion (CV) for different rodent species in different commu nities. For designations, see Fig. 1.
The dominance structure of the community changes from year to year, although the dominant species retain their rank or status in the community (Fig. 4c). The species with the highest longterm ranks in the community (the northern redbacked vole, root vole, and northern birch mouse) are also the most “cyclic” ones, whereas the species of medium rank show a medium range of longterm changes in this parameter. The species of low abundance in this community are 2013
344
LITVINOV et al.
Table 3. Coefficients of correlation with principal compo nents for parameters of rodent species abundance in Teletskaya Taiga in different years Principal component Character (year) 1 1984 1985 1986 1987 1988 1989 1990 1991 1992 1994 1997 1999 2000 2001 2002 2003 2004 Proportion of variance, %
0.98 0.75 0.95 0.98 0.98 0.86 0.93 0.81 0.92 0.84 0.87 0.95 0.88 0.74 0.97 0.92 0.63 78.1
2
3
0.04 –0.55 –0.31 –0.12 –0.16 –0.50 0.31 0.51 0.32 0.52 0.46 –0.04 –0.39 –0.62 –0.16 0.29 0.39
–0.08 –0.36 0.02 0.10 –0.09 0.07 0.17 –0.11 0.18 0.03 0.09 0.02 0.20 –0.06 0.08 0.15 –0.67
14.2
4.5
year, and steadiness of cyclic processes. In populations of the bank vole, striped field mouse, and northern birch mouse, changes in abundance are synchronous with those of other species; in the second group (voles of the genus Microtus and the European water vole), the time course of these changes does not correlate with that in the first group. It can be suggested that similarity in the time course of abundance between species within these groups is accounted for by their preference for a certain level of moisture supply. The group of dominant species and the group of species of medium abundance are evenly distributed with respect to rank composition and its variance, and it is these species that determine the longterm structure of the community. Analysis of the principal dynamic parameters of the rodent community of Teletskaya Taiga has revealed a medium value of total abundance and the highest vari ation in this parameter, at a medium value of summa rized information parameters. The phase portrait shows alternating long and short periods of system resets to the initial state (in annual diversity parame ters). The dominants of this community, the northern redbacked vole and root vole, show no synchronism in the time course of annual abundance (Litvinov et al., 2007). The rodent community of Teletskaya Taiga is dominated by populations of species with a high longterm rank and the highest coefficients of variation in this parameter. Populations of species with a medium or low rank and low variation in this param eter are rather scant in this community.
characterized by the lowest ranks and the lowest long term average coefficient of variation in rank.
REFERENCES
CONCLUSIONS Many species common to all three key areas have been recorded in the analyzed communities. The pop ulations of these species have different positions in the dominance structure of communities and different effects on their composition and dynamics, depending on the natural climatic parameters of the landscape. The rodent community of Northern Baraba is characterized by strong changes in abundance from year to year, long periods of the system reset to a bal anced state, and strong rearrangements in the com munity structure from year to year, depending on the natural climatic factors. The time courses of change in the abundance of the dominant species (the striped field mouse) and the key species (the European water vole) are in antiphase. The longterm dominance structure of the community is formed as follows: spe cies of “cyclic” rank change this structure, while spe cies of “noncyclic” rank are conductive to the stabili zation of this structure. The rodent community of Akademgorodok is char acterized by a low average value of total abundance and low variation of this parameter, a high level of spe cies diversity, which remains constant from year to
Begon, M., Harper, J.L., and Townsend, C.R., Ecology: Individuals, Populations, and Communities, Oxford: Black well, 1986. Translated under the title Ekologiya: Osobi, pop ulyatsii i soobshchestva, Moscow: Mir, 1989. Bobretsov, A.V., Beshkarev, A.B., Basov, V.A., et al., Zakonomernosti poluvekovoi dinamiki bioty devstvennoi taigi Severnogo Predural’ya (Trends in the HalfCentury Dynamics of the Biota in Virgin Taiga Forests of the North ern Cisural Region), Syktyvkar: Goskomstat Respubliki Komi, 2000. Efimov, V.M. and Galaktionov, Yu.K., On the possibility of predicting cyclic changes in the abundance of mammals, Zh. Obshch. Biol., 1983, no. 3, pp. 343–352. Efimov, V.M. and Kovaleva, V.Yu., Mnogomernyi analiz bio logicheskikh dannykh (Multivariate Analysis of Biological Data). GornoAltaisk: RIO GAGU, 2007. Efimov, V.M., Galaktionov, Yu.K., and Shushpanova, N.F., Analiz i prognoz vremennykh ryadov metodom glavnykh komponent (Analysis and Prognosis of Time Series by the Method of Principal Components), Novosibirsk: Nauka, 1988. Ekologicheskie sistemy. Adaptivnaya otsenka i upravlenie (Ecosystems: Adaptive Assessment and Management), Moscow: Mir, 1981. Fedorov, V.D. and Gil’manov, T.G., Ekologiya (Ecology), Moscow: Mosk. Gos. Univ., 1980.
RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
2013
SIGNIFICANCE OF RODENT POPULATION DYNAMICS Hansson, L., Development and application of landscape approach in mammalian ecology, Abstr. 6th Int. Theriol. Congr., Sydney, July 4–10, 1993, Rensington, 1993, p. 126. Kuzyakin, A.P., Zoogeography of the Soviet Union, in Bio geografiya (Biogeography): Uch. Zap. MOPI im N.K. Krupskoi, Moscow, 1962, pp. 3–82. Litvinov, Yu.N., Soobshchestva i populyatsii melkikh mleko pitayushchikh v ekosistemakh Sibiri (Communities and Pop ulations of Small Mammals in Ecosystems of Siberia), Novosibirsk: TsERIS, 2001. Litvinov, Yu.N. and Panov, V.V., Structural connections as an element of biodiversity in rodent communities of North ern Baraba, Usp. Sovrem. Biol., 1998, vol. 118, no. 1, pp. 101–108. Litvinov, Yu.N., Abramov, S.A., Krivopalov, A.V., Novikov, E.A., and Chechulin, A.I., Structural–temporal organization of a rodent community in the Teletskaya Taiga, Altai Moun tains, Russ. J. Ecol., 2007, no. 6, pp. 413–416. Maksimov, A.A., Mnogoletnie kolebaniya chislennosti zhi votnykh, ikh prichiny i prognoz (LongTerm Fluctuations in the Abundance of Mammals: Causes and Prognosis), Novosibirsk: Nauka, 1984. Naumov, N.P., Studies on mobility and abundance of small mammals using trench traps, in Voprosy kraevoi, obshchei i eksperimental’noi parazitologii i meditsinskoi zoologii (Problems in Regional, General, and Experimental Parasitology and Medical Zoology), vol. 9, Moscow: Medgiz, pp. 179–202.
RUSSIAN JOURNAL OF ECOLOGY
Vol. 44
No. 4
345
Panteleev, P.A., Populyatsionnaya ekologiya vodyanoi polevki i mery bor’by (The Water Vole: Population Ecology and Control), Moscow: Nauka, 1968. Primak, R.K., Osnovy sokhraneniya bioraznoobraziya (Fun damentals of Biodiversity Conservation), Moscow: NUMTs, 2002. Romanovskii, Yu.M., Stepanova, N.V., and Chernavskii, D.S., Chto takoe matematicheskaya biofizika (What Is Mathemat ical Biophysics?), Moscow: Prosveshchenie, 1971. Sadykov, O.F. and Benenson, I.E., Dinamika chislennosti melkikh mlekopitayushchikh (Population Dynamics in Small Mammals), Moscow: Nauka, 1992. Tereshchenko, V.G. and Verbitskii, V.B., Method of phase portraits for analyzing structural dynamics of hydrobiont communities, Biol. Vnutr. Vod, 1997, no. 1, pp. 23–31. Williams, S.E., Marsh, H.A., and Winter, J.N., Spatial scale, species diversity, and habitat structure of small mam mals in Australian tropical rain forest, Ecology (USA), 2002, vol. 83, pp. 1317–1329. Williamson, M., The Analysis of Biological Populations, London, 1972. Translated under the title Analiz biolog icheskikh populyatsii, Moscow: Mir, 1975. Yoccoz, N.G. and Ims, R.A., Spatial population dynamics of small mammals: Some methodological and practical issues, International Conference EURING 2003 “The Quantitative Study of Marked Individuals in Ecology, Evo lution, and Conservation Biology,” Radolfzell, 2003, Anim. Biodivers. Conserv, 2004, vol. 27, no. 1, pp. 427–435.
Translated by P. Petrov
2013
