Cent. Eur. J. Biol. • 8(1) • 2013 • 8-17 DOI: 10.2478/s11535-012-0108-2

Central European Journal of Biology

Autumn-winter diet overlap of fallow, red, and roe in forest ecosystems, Southern

Research Article Artur Obidziński1,*, Piotr Kiełtyk2, Jakub Borkowski3,4, Leszek Bolibok5, Karolina Remuszko3,6

1Department of Forest Botany, Warsaw University of Life Sciences, 02-776 Warsaw, Poland

2Faculty of Biology and Environmental Sciences, Cardinal Stefan Wyszyński University, 01-938 Warsaw, Poland

3Department of Forest Ecology, Forest Research Institute, 05-090 Sękocin Las, Poland

4Department of Forestry and Forest Ecology, University of Warmia and Mazury, 10-727 Olsztyn, Poland

5Department of Silviculture, Warsaw University of Life Sciences, 02-776 Warsaw, Poland

6Forest Directorate Wielbark, 12-160 Wielbark, Poland

Received 30 May 2012; Accepted 21 September 2012

Abstract: The wild population of fallow deer in Central Europe has grown considerably over the last decade. However, information on feeding habits of this alien species in relation to the indigenous or , in areas of their co-occurrence, is scarce. A prevailing view maintains that their food-niches are distinct, although direct comparative studies have not been carried out. Therefore, the aim of the research was to compare the diets of fallow, red, and roe deer feeding in the same . Research was based on the rumen contents of 242 hunted in the autumn-winter season in the forests of Southern Poland. The analyses demonstrated that fallow deer are moderate grazers in such conditions and eat more graminoids in comparison to red or roe deer (36.4% vs. 16.1% or 5.5%, respectively). On the other hand, it feeds on less browse (17.2% vs. 41.4%) or dwarf shrubs (8.4% vs. 19.0%) than red deer, and on less bramble (10.9% vs. 34.6%) or forbs (4.0% vs. 7.6%) in comparison to roe deer (P=0.05). Although the diets of the three deer species differ in terms of the proportion of each food type in their diet, overlapping of their food-niches is high (52.6%). Keywords: Ungulates • Feeding preferences • dama • elaphus • capreolus • Central Europe © Versita Sp. z o.o.

1. Introduction and regulation due to their role in the environment [e.g. 9-11]. Ungulates transform inhabited ecosystems to a great Roe deer, Capreolus capreolus, and red deer, Cervus degree. They influence species composition and elaphus, are the most common cervids in Europe. Their structure of plant communities [1-3]. They also have an ecology is researched and presented in various studies indirect impact on populations of invertebrates, birds, [e.g. 12,13]. Fallow deer, Dama dama, have been and even small [e.g. 4-6]. Therefore, the historically less common and less well-studied. This significant growth of the number of ungulates recorded in species was introduced in southern Europe from Asia Europe [7,8] requires adequate examination, monitoring, Minor in antiquity and is presently spreading intensely

* E-mail: [email protected] 8 A. Obidziński et al.

across Central Europe [14]. For example, the number of In response to the aforementioned factors, we fallow deer has risen in Poland nearly fourfold in the years focused our research on the study of the autumn-winter 2000–2010 – from approximately 6500 to 23300, in spite diet of fallow deer, in comparison with the diet of roe and of a simultaneous increase of the hunting harvest from red deer, in the forest environment of Central Europe. 1700 to 5100 specimens yearly (Central Statistical Office. The following hypotheses were formulated based on the Forestry 2010, http://www.stat.gov.pl/gus/5840_1540_ premises listed above: 1) graminoids constitute the main PLK_ HTML.htm, Last access 15.04.2012). component of fallow deer autumn-winter diet in central The growth of the fallow deer population in Europe Europe; 2) percentage share of particular food types in and its subsequent impact on the environment, the diet of fallow deer differs from that in the diets of roe especially on forests, has resulted in keen scientific or red deer, so the overlap is small; 3) the diet of fallow interest. Results of a number of studies suggest little deer is more opportunistic, so it is more varied than that impact of fallow deer on tree stands [e.g. 15-18]. while of roe or red deer. others demonstrate significant economic damage and unfavourable environmental changes [e.g. 19-23], including inhibition of the natural regeneration of stands 2. Experimental Procedures and changes in the direction of vegetation succession [e.g. 24,25]. Fallow deer most frequently feed on 2.1 Study area graminoids [15,17,26-32]. However, occasionally their The research was carried out in the Raciborskie Forest diet is dominated by tree browse [19,23,28,30], acorns over an area of ~40 000 ha situated in southern Poland: [15,16,33], dead leaves and lichens [34], or fruits [35]. N50º08’00”-23’10” and E18º11’50”-33’40”, at an altitude As fallow deer are an alien species in Central Europe, of 178–252 m above sea level (Figure 1). Arenosols it is of further value to establish whether there may be food prevail in the southern and western portions of the area. competition between fallow deer and indigenous cervids, They represent coniferous forest with Querco and if so, to what extent. This is especially important roboris–Pinetum potential vegetation. Cambisols prevail considering observations that indicate that the growth in the north and east, and they represent broadleaved of the fallow deer population may cause a decrease in forest habitats with Galio–Carpinetum and Melico– the roe deer population [36,37]. Comparative studies Fagetum potential vegetation (Matuszkiewicz, J. M. of the diets of the above-mentioned cervids inhabiting 2008. “Potential natural vegetation of Poland – numeric the same areas simultaneously are scarce. They map”, IGiPZ PAN, Warszawa, http://www.igipz.pan.pl/ demonstrate that fallow deer eat less forbs or browse and more graminoids than roe deer [23,27], as well as KĘDZIERZYN- more graminoids, forbs, or acorn and less browse than KOŹLE red deer [28,32,38]. However, no comparative studies of all three – fallow, roe, and red deer – living in the same place at the same time have previously been published. Existing information on the fallow deer’s diet and its comparison with roe or red deer’s diets comes mostly from areas with mild winters [e.g. 15,16,19,28,33]. Odra River Snowy winters usually cause an increase in feeding on browse in all cervids [e.g. 28,39,40]. However, even seasonal differences in nutritional content in plants can cause alter the composition of the diet [41]. Long lasting snow cover, typical for the study area in Central Europe, N RUDY can cause different feeding behaviour in fallow deer from W E that described in the south and west of the continent. S It can be supposed that limited access to forest floor POLAND vegetation, due to a thick layer of snow, should result in 0 5 10km an increase of browse in the fallow deer’s diet, and more intense competition between the three cervids species research area multiaged managed forest for access to food supplies. Results of our study of the fallow deer’s diet can be of practical significance for main rivers even-aged thickets decision making on this species’ population size control in this part of the continent. Figure 1. Location of the study area.

9 Autumn-winter diet overlap of fallow, red, and roe deer in forest ecosystems, Southern Poland

geoekoklimat/roslinnosc/prn_mapa/home_pl.htm. Last analysed. Material for analyses was collected during access 15.04.2012). The mean temperature is -1.2ºC the September–February hunting season from the years in January, and 18.1ºC in July. The mean annual 2004/5 to 2010/11. 242 rumina of hunted animals were precipitation level is approximately 630 mm and the used for the study. Collected material included 84 fallow vegetation period lasts about 224 days [42]. About deer, 93 red deer, and 65 roe deer. 127 specimens 9000 ha of the Raciborskie Forest stands were burnt were killed in autumn and 115 in winter. 86 samples in the year 1992. Over 6000 ha were replanted, mostly were obtained from large thickets and 157 samples with Scots pine. The remaining area regenerated on its were obtained from regular managed forest (Table 1). own, mainly with silver birch [43]. In the period of study, Samples were collected across the entire area of study regeneration reached the stage of thickets 11–18 years with a close to regular distribution. The distribution in age. of the sexes in each species sample was 37 males, Scots pine (Pinus sylvestris) dominates the forest as 37 females, and 10 calves in fallow deer; 36 males, a whole, occupying 68% of its area. Silver birch (Betula 38 females, and 19 calves in red deer; and 11 males, 48 pendula) occupies 13%, and pedunculate oak (Quercus females, and 6 fawns in roe deer. The lower number of robur) occupies 9%. Norway spruce (Picea abies), male specimens among the roe deer sample (caused by European beech (Fagus sylvatica), oak (Quercus robur), hunting laws in Poland) does not need to be considered, and black cherry (Prunus serotina) prevail in the shrub as differences between the diets of opposite sexes of layer of the unburnt parts of the forest. Common bracken roe deer are generally insignificant [e.g. 39,45,46]. The (Pteridium aquilinum), bilberry (Vaccinium myrtillus), ages of hunted specimens were determined by dentine reed grass (Calamagrostis epigeios), wavy hair grass layers, but was not always available as hunters did not (Deschampsia flexuosa), quaking grass sedge (Carex consistently described samples. Age was recorded for briziodes), and bramble (Rubus fruticosus) dominate 54 out of 84 fallow deer, 42 out of 93 red deer, and 34 the forest floor. Pine with admixture of birch, European out of 65 roe deer specimens. Mean age of harvested larch (Larix decidua), and common aspen (Populus fallow deer was 4.14 years (±SD=2.00), while it was tremula) are the most common types of vegetation in the 2.29 years (±SD=2.36) for red deer, and 3.21 years thickets overgrowing burnt areas. Reed grass, purple (±SD=1.41) for roe deer. moor grass (Molinia caerulea), wavy hair grass, and About 500 cm3 of the rumina’s contents was common heather (Calluna vulgaris) [44] dominate the collected directly after the specimens were killed, and forest floor of the thickets. Crops and canola, as well as preserved in 5% formaline solution. Every sample some corn and root crops, available from neighbouring was labeled in the field. The date and location of fields (personal observation). 150 tons of fodder beet, collection, as well as the species, gender and age approximately 100 tons of silage, about 100 tons of each given specimen, was marked on the label. of crop grain (corn, oats, barley, and wheat), and roughly Samples were then washed in a sieve of 1 mm mesh 30 tons of hay were distributed as winter feed (based in the lab. Subsequently, three samples of 5 cm3 each on oral information from local hunting clubs in 2010). In were extracted from the fraction left in the sieve from 2009, the populations of roe deer, red deer, and fallow each rumen. They were then analysed using the ”point- deer in the surveyed area were approximately 10.5, frame” method [47], which involved mixing them with 14.8, and 9.6 individuals/km2, respectively (J. Borkowski water and smearing the samples on a tray with a grid of et al., unpublished data). lines forming 100 intersection points along the bottom, allowing for random selection of 100 particles of food. 2.2 Sample collection and data analysis The procedure was repeated three times to obtain a Food obtained from the rumina of animals harvested sample of 300 items, which is considered adequately by hunters according to annual hunting plans was representative [48]. Collected particles were identified

Species Red deer Fallow deer Roe deer

Season autumn winter autumn winter autumn winter

Forest 23 30 24 25 26 28

Thickets 18 22 26 9 10 1

Sum 41 52 50 34 36 29

Table 1. Number of samples of categories of animals analyzed.

10 A. Obidziński et al.

by the naked eye or enlarged with a 25x magnifying of food; 5) significance of each type of food in the diet; binocular. Particles were then divided into the following and 6) overlapping of the species’ diets. groups: browse and needles of coniferous trees (Pinus The niche breadth of food categories consumed 2 sylvestris, Picea abies, Larix decidua), browse and was assessed using Levins’ [51] measure: B=1/Σpi , leaves of broadleaf trees (mainly Prunus serotina and where B is the niche breadth and pi is the proportion of Quercus robur), dwarf shrubs (Vaccinium myrtillus, food category i in the diet. This index was standardized V. vitis-idaea, Calluna vulgaris), bramble (Rubus to a scale of 0−1 following formula of Hurlbert [52]: fruticosus s.l.), grasses (mainly Deschampsia flexuosa, Bsta=(B–1)/(Bmax–1), where Bmax is the total number Poa sp., Agrostis sp.), sedges (mainly Carex brizoides, of food categories consumed by at least one Carex ericetorum), dicotyledones forbs, mushrooms, species. acorns, crops (mainly rape leaves), additional feed Similarities of samples were established with the given (mainly fodder beet and crop grains), pollution, use of principal components analysis (PCA) based on and unidentified. Plants collected in the study area the correlation matrix of the variables [53]. Samples were used as reference material. Identified particles obtained from the specimens’ rumina defined according were dried at 60°C for 48 hours and weighed with an to percentage share of particular types of food were accuracy of ±0.001 g. Therefore, comparison of diets of used as statistical units. The PCA diagram (Figure 2A) analysed species was based on dry biomass of food. As represents classification of specimens for each food in each rumen was ruminated to a different degree, species. A separate diagram (Figure 2B) represents the percentage share of a particular type of food in the contribution of each type of food for the specimens’ the rumina was calculated based on dry mass of food ordination in reference to PC1 and PC2. Factor loadings and not on frequency of food particles. In the cases of expressing factor-variable correlation are presented in strongly ruminated samples, the easily-digested forbs Table 2. The absolute magnitude of their values reflects made up a lower percentage of the mass than other the strength of the relationship between a variable and food types [49]. To compensate for these altered values, particular axis. The signs + or – in their values indicate some authors [e.g. 50] suggest doubling the amount positive or negative correlation of a variable with a of forbs in the results. Forbs analysed in our research given axis. Calculations were done with the use of the were not doubled because samples were of varied Statistica 8.0 package (StatSoft Inc. 2008. Statistica ver. levels of digestion. It can be therefore assumed that the 8.0). actual percentage of forbs could be slightly higher than Percentage share of each type of food in the diets demonstrated. of the studied species were compared by using the Diets of analyzed species of animals were Kruskal-Wallis test with an adequate test of multiple compared in terms of: 1) number of types of food; 2) comparisons [54]. This method was chosen due breadth of food-niche (diet diversity); 3) similarities of to different numbers of animals in each group and sample contents and principal food types that samples impossibility of transforming data to normal distribution. differed by; 4) percentage of dry biomass of each type Implementation of the above-mentioned tests was done

1

2

Axis 2 graminoids

dwarf shrubs 0 forbs

deciduous 0 coniferous browse -2 browse feed

Axis 2; eig.=1.423 (15.8%) acorn and mushrooms -4 red deer cultivated plants roe deer bramble fallow deer Axis 1; eig.=1.613 (17.9%) Axis 1 -6 -1 -2 0 2 -1 0 1

Figure 2. Diversity of components of the autumn-winter diet of red deer and roe deer in the autumn-winter period in forest ecosystems (S Poland) acquired with the PCA method; a) ordination of samples of rumina content, b) ordination of type of food.

11 Autumn-winter diet overlap of fallow, red, and roe deer in forest ecosystems, Southern Poland

Food Axis 1 Axis 2 roe deer and red deer were significant (Kruskal-Wallis test P=0.05). forbs 0.3591 0.1718 The food-niche breadth of the examined animals graminoids 0.5129 0.5295 is characterised by high variability (0.01–0.38). Fallow bramble -0.0409 -0.7776 deer and red deer demonstrated similar mean food- deciduous browse 0.3792 0.0129 niche breadth (0.16±0.08 and 0.17±0.07, respectively) and at the same time were significantly (Kruskal-Wallis coniferous browse -0.7882 0.1280 test P=0.05) higher than that of roe deer (0.12±0.07). dwarf shrubs -0.5402 0.3096 Graminoids and forbs were the types of food acorn and mushrooms 0.3017 -0.1919 that distinguished the diet of the fallow deer from the feed 0.2560 0.0088 other two species’ diets. Bramble was the distinctive cultivated plants 0.0770 -0.5993 type of food for roe deer, and needles, coniferous browse, and dwarf shrubs were distinctive for red Table 2. Results of principal components analysis (PCA); factor deer (Figures 2A,B). More differences were found loadings expressing correlations of variables (diet between fallow deer and roe deer than between red components) with PCA axes. deer and either of the other two species. Each axis in the principal components analysis explained a limited with procedures from the agricolae library written by amount of the entire variability. Eigenvalues of the nine de Mendiburu (2009, Agricolae: Statistical Procedures extracted axes amounted to 1.6131, 1.4230, 1.2155, for Agricultural Research. R package version 1.1012, 1.0147, 0.9216, 0.8748, 0.8331, and 0.0031, 1.0-7. Vienna. Austria: The R Foundation for Statistical which correspond to the following percentages of total Computing) for R package (R Development Core Team. variability demonstrated by each axis: 17.92%, 15.81%, 2008. R: a language and environment for statistical 13.51%, 12.24%, 11.27%, 10.24%, 9.72%, 9.26%, and computing. R Foundation for Statistical Computing. 0.03%, respectively. Axis 1 was the most negatively Vienna. Austria. URL http://www.R-project.org, on-line correlated with coniferous tree browse and needles access 15.01.2008). Percentage share of mass of (r=–0.788) and dwarf shrubs (r=–0.540), whereas it was chosen category of food in a single rumen examined positively correlated with graminoids (r=0.513) (Table 2). was used as statistical unit in calculations. The level of Axis 2 was the most negatively correlated with bramble significance in statistical analyses was set at α=5%. (r=–0.778) and crops (r=–0.599), whereas it was Index of importance of food type in the diet was set positively correlated with graminoids (r=0.529). after Bruno and Apollonio [19]: W=100Ni/N x 100Mi/M, Frequency of type of food identified in the rumina where Ni = number of rumina where food i was found, of examined species was high, reaching on average N = number of all rumina analysed, Mi = mass of food i in more than 50% (mean 59.2%) with the exception of all rumina, and M = mass of all type of food in all rumina. acorns (27.3%), mushrooms (10.3%), fodder supplied This index indicates, in a most complete manner, the by hunters (18.5%), and crops (4.5%), as well as, in the significance of particular types of food for each species case of fallow and roe deer, browse (38.6%). Graminoids of . However, it unfortunately does not allow for (97.8%), needles (92.2%), and dwarf shrubs (91.4%) establishment of the significance of differences between were found to be most abundant in the rumina of red obtained values. deer. Graminoids (97.6%), needles (97.6%), and forbs Dietary overlap was calculated using Schoener’s (91.7%) were found to be most prevalent in the rumina of index [55]: fallow deer. Bramble (87.7%), graminoids (86.2%), and O=100(1-0,5∑|Xi-Yi|) [%], needles (81.5%) were found to dominated the rumina of where Xi and Yi are the proportions of each food roe deer. Dwarf shrubs and browse were found in higher category in the rumina of species X and Y. frequency in rumina of red deer than in the other two species, whereas, mushrooms and crops were of higher frequency in the rumina of roe deer than in the other two 3. Results species (Table 3). The percentage share of dry mass of each type Food type analysis demonstrated the presence of 2–11 of food in the diets of analysed species differed out of 13 types of food in any particular rumen. The mean significantly. Needles, dwarf shrubs, and graminoids co- (±SD) number of types of food was highest in the rumina dominated in the diet of red deer (20.3%, 19.0%, and of red deer (8.1±1.2), lower in fallow deer (7.7±1.2), and 16.1%, respectively). Graminoids (36.4%) composed lowest in roe deer (7.3±1.8). Only differences between the highest percentage share in the diet of fallow deer,

12 A. Obidziński et al.

Frequency % share of dry mass ± SD Importance Food Red Fallow Roe Red Fallow Roe Red deer Fallow deer Roe deer deer deer deer deer deer deer

forbs 86.0 91.7 75.4 1.7±2.4 b 4.5±6.6 a 7.6±15.2 b 74.8 308.5 367,8

graminoids 97.8 97.6 86.2 16.1±16.9 b 36.4±24.1 a 5.5±12.1 c 689.7 2842.3 263,2

bramble 82.8 86.9 87.7 9.0±13.1 b 10.9±14.5 b 34.6±29.7 a 380.2 578.1 2329,6

deciduous leaves 89.2 77.4 80.0 9.113.6 a 6.4±12.0 a 7.9±12.6 a 398 415.8 453,6

coniferous needles 92.5 97.6 81.5 20.3±18.9 a 10.7±13.9 b 14.4±22.1 b 1145 1000.2 1200,7

dwarf shrubs 91.4 78.6 76.9 19.0±21.1 a 8.4±13.6 b 10.9±18.1 b 986.8 531.7 914,8

deciduous shoots 58.1 46.4 43.1 5.9±11.8 a 3.0±8.4 b 1.6±4.1 b 206.6 104.7 85,1

coniferous shoots 64.5 28.6 38.5 6.1±12.9 a 2.3±6.1 b 2.8±5.3 b 323.5 78.2 111,2

acorn 21.5 31.0 26.2 3.3±11.5 a 5.2±11.5 a 8.3±22.6 a 105.1 213.7 601,5

mushrooms 3.2 8.3 20.0 0.5±5.0 b 1.0±4.9 b 1.8±7.5 a 0.5 5.8 34,3

feed 22.6 31.0 3.1 6.6±15.3 a 7.4±16.1 a 0.9±5.1 b 111.8 395.1 2,5

cultivated plants 3.2 1.2 9.2 0.5±3.1 ab 0.4±3.4 b 1.7±6.4 a 1.4 0.2 12,4

other 96.8 94.0 96.9 2.0±2.7 b 3.5±3.6 a 1.9±2.2 b 103 232.4 142,3

Table 3. Frequency, percentage share, and importance of food types in the autumn-winter diet of fallow deer, roe deer, and red deer in forest ecosystems (S Poland). Red deer N=93, fallow deer N=84, roe deer N=65. The same symbol of the alphabet is used for marking means that do not differ according to the level of significance set as P=0.05 for the same type of food. followed by a significantly lower percentage share of red deer roe deer bramble (10.9%) and needles (10.7%). The diet of roe 8.3 deer was dominated by bramble (34.6%), while needles (14.4%) and dwarf shrubs (10.9%) demonstrated 21.3 32.0 significantly lower percentage share. Red deer ate more needles, dwarf shrubs, and browse in comparison with 52.6 the other two species. Fallow deer ate more graminoids, and roe deer ate more bramble, forbs, mushrooms, and 17.8 7.1 crops (Kruskal-Wallis test P=0.05; Table 3). The value distribution of importance indices of types of food was to a great degree similar to the percentage 22.5 share of food mass (Table 3). Needles (1145), dwarf shrubs (987), and graminoids (690) reached the highest values in the diet of red deer. Graminoids (2842) played fallow deer the most important role in the diet of fallow deer, followed Figure 3. Overlap of the autumn-winter diets of red deer, fallow by still high though lower percentage share of needles deer, and roe deer in forest ecosystems (S Poland), (1000) and bramble (578). Bramble (2330) dominated based on mean percentage share of food mass. the diet of roe deer, followed by the still important though smaller role played by needles (1201) and dwarf shrubs 4. Discussion (915). In spite of significant differences in percentage share of particular types of food in the diets of fallow The diet of the fallow deer analysed is highly varied. deer, red deer, and roe deer, 52.6% of the diet was Although graminoids constitute the largest part (over common between all three species. Dietary overlap 1/3) of the diet, they were less abundant than in the reached 70.4% between red deer and fallow deer, 59.7% cases where fallow deer behaved as typical grass- between fallow deer and roe deer, 60.9% between roe eating species [e.g. 15,27-29,35]. On the other hand, deer and red deer (Figure 3). percentage share of browse in their diet was too small

13 Autumn-winter diet overlap of fallow, red, and roe deer in forest ecosystems, Southern Poland

to consider fallow deer a browser species – as was Differences between the diets of the studied demonstrated in a few other places [19,23,30,56]. species generally resemble results obtained by earlier Food supplies present in the study area described studies of diet in these species. Similar to the results in the literature [44] encompassed eight categories: presented here, fallow deer were previously found to deciduous branches, coniferous branches, deciduous feed on more graminoids and forbs and less browse seedlings, coniferous seedlings, dwarf shrubs, bramble, with respect to red deer [28,32,38]. On the other hand, forbs, graminoids, and mosses. The percentage share fallow deer demonstrated smaller consumption of dwarf of the dry mass of food found in the rumina was shrubs than the other two species. Also, similarly to compared with that of their share in the environment other authors, we found fallow deer to feed on more with the Ivlev’s electivity index [57]. We found that graminoids and less forbs, browse, leaves, and needles fallow deer preferred dwarf shrubs and bramble while in comparison to roe deer [23,27]. However, contrary they avoided tree leaves, forbs, and mosses. They to the findings of other authors, we recorded a lower ate graminoids and woody browse proportionally to consumption of bramble, mushrooms, and crops, and these food types’ share in the environment. The diet greater consumption of fodder provided by hunters, as of red deer analyzed in this study was rather varied well as a lack of differences with respect to consumption and had a medium percentage share of plants rich of browse. However, it should be emphasised that food in fibre. In comparison with the food acquired by red availability in a given environment has an important deer in coniferous forests in other locations [58], it impact on the results [58,59]. contained more needles, woody browse, and bramble, Differences in diets of fallow deer, red deer, and less forbs, and similar amounts of dwarf shrubs and roe deer seem to be a constant predisposition as they tree leaves. With respect to the food supply available are based on the animals’ different structure and the in the study area [44], red deer preferred dwarf shrubs functioning of the digestive tract. The digestive tract and avoided forbs and mosses. Red deer acquired of the roe deer best digests forbs, while fallow deer other types of food according to their prevalence in - graminoids, and the red deer’s digestive system lies the environment. The diet of roe deer analyzed in somewhere in between these two [60,61]. It is further this study was dominated by plants of high nutritive supported by the fact that in spite of a close evolutionary value. In comparison to food acquired by roe deer in relationship between of these three species, the fallow coniferous forests in other locations [59], it contained deer’s diet is sometimes more similar to the diet of significantly more bramble, less dwarf shrubs and typical grazing species, such as bovines or even equids graminoids, and a similar share of forbs and coniferous [27]. browse. With respect to the food supply available in the The overlap of the autumn-winter diet of fallow study area [44], roe deer preferred bramble and dwarf deer with that of indigenous cervids across the study shrubs and they avoided graminoids, tree leaves, and area was higher than in the west of the continent. It mosses. They fed on needles and forbs accordingly was higher by 13.3% in the case of fallow deer and to their prevalence in the environment. Roe deer roe deer than in England, as calculated from Putman’s demonstrated the highest selectivity while fallow deer data [27]. In the case of fallow deer and red deer, it demonstrated the lowest. At the same time, the diet of was higher by 17.4% than in Spain, as calculated from the fallow deer was more similar to that of the red deer the data of Garcia-Gonzales and Curtas [28]. Differing in terms of contents and variety than to the diet of the development of vegetation resulting from the diverse roe deer. The impact of additional feeding and crops climates of these areas must be taken into account in available on neighboring fields on the composition of the interpretation of these differences. The latter may the diet of all three species may be considered to be result from lower autumn-winter food diversity in the small. The percentage share of cultivated plants in the environment with longer and harder winters (Poland). A rumina of all three species did not exceed 2% and in change in diet can be caused not only by lack of access winter was close to zero. Although fallow deer are to certain food types in given seasons, but also to a often considered to be more eager to wander out of change in the nutritive properties of the same plants the forest and onto fields than red deer, the diet does over different seasons [41]. Both of these factors can not support this view. The percentage share of fodder cause the overlap of food-niches between cervids to be given by hunters did not exceed 7.5% of the diet of any higher in winter than in other seasons [48,62]. Particular of the analysed species in the period of study, and in climatic conditions may cause food competition between winter it constituted a maximum of 10.5% of dry mass fallow deer and indigenous cervids [62], especially in a of their food. It is worth mentioning that similar feeding situation when populations are too high or winter lasts is practiced all over Poland. longer than usual.

14 A. Obidziński et al.

On the other hand, the overlap of the diet does not diversity of these supplies. Competition can occur necessarily result in interspecific competition. Given between these species, and it may grow stronger in that a preferred type of food is supplied in abundance, Europe with increasing of geographic longitude and two species can co-exploit it without detriment to one latitude, especially with further growth of fallow deer another [63]. In order for exploitation competition to take populations. place, an overlap in habitat use and in diet must occur The results of this study provide reasons for setting simultaneously and the shared dietary resource must be permissible densities of fallow deer populations in order limited [e.g. 48,64,65]. to prevent competition between this species and red or Food competition between fallow deer and the other roe deer in different ecosystems or climatic regions. It two deer species has not been previously recorded in is worth mentioning that setting the right numbers for Poland. However, in view of the rapid growth of the fallow deer populations should take into account the fallow deer population, the question whether this status body mass ratio of 1:2 of fallow deer to roe deer and 2:1 quo will remain or not is valid. This concern is supported of fallow deer to red deer [67]. It also seems advisable by observations made in of decreasing roe deer to leave a number of hunting areas without fallow deer populations in areas where fallow deer settled [36,37], as control plots for monitoring the relations between this as well as observations suggesting the same from species and the indigenous cervids co-inhabiting with England [66]. Food competition between red deer and fallow deer in other locations. fallow deer is suggested by a number of studies [32,66]. Rapid growth of the fallow deer population gives cause for concern in terms of imminent increase in damage in Acknowledgements young tree stands as well, after introduction of fallow deer next to roe and red deer in Central European We would like to thank hunters from the following forests. Although fallow deer generally avoid browse, hunting clubs: Azoty, Daniel, Darz Bór, Łoś, Odyniec, they feed on it during the winter. Ostoja, Ponowa, and Ryś, and particulary to Mr. The results of this study and those in the published Ryszard Kot and Mr. Zygmunt Tomaszek for collecting literature indicate that fallow deer feed on their the rumen content samples. We are also grateful to preferred food – graminoids – whenever possible. On Prof. Jacek Goszczyński of Warsaw University of Life the other hand, they are a flexible species and adjust to Sciences for helpful suggestions to the first version of the food supply available in the environment. Overlap this paper and to anonymous Referees for their valuable of food niches between fallow deer and the indigenous and encouraging remarks. Research was financed cervids is small in environments and seasons with by the Faculty of Forestry WULS and by the General diverse food-plant supplies. Overlap of their food Directorate of Polish State Forests with research grant niches increases in environments and seasons of low No. BLP-267. Translated by Bogna J. Obidzińska.

References

[1] Bradshaw R.H.W., Hannon G.E., Lister A.M., A long- [5] Fuller R.J., Responses of woodland birds to term perspective on ungulate-vegetation interactions, increasing numbers of deer: a review of evidence For. Ecol. Manage., 2003, 181, 267–280 and mechanisms, Forestry, 2001, 74, 289–298 [2] Cote S.D., Rooney T.P., Tremblay J.P., Dussault [6] Flowerdew J.R., Ellwood S.A., Impacts of woodland Ch., Waller D.M., Ecological impact of deer deer on small ecology, Forestry, 2001, 74, overabundance, Annu. Rev. Ecol. Evol. Syst., 277–287 2004, 35, 113–47 [7] Fuller R.J., Gill R.M.A., Ecological impacts of [3] Martin J.-L., Stockton S.A., Allombert S., Gaston increasing numbers of deer in British woodland, A.J., Top-down and bottom-up consequences of Forestry, 2001, 74, 193–199 unchecked ungulate browsing on plant and animal [8] Milner J.M., Bonenfant Ch., Mysterud A., Gaillard J.M., diversity in temperate forests: lessons from a deer Csányi S., Stenseth N.Ch., Temporal and spatial introduction, Biol. Invasions, 2010, 12, 353–371 development of red deer harvesting in Europe: biological [4] Stewart A.J.A., The impact of deer on lowland and cultural factors, J. Appl. Ecol., 2006, 43, 721–734 woodland invertebrates: a review of the evidence [9] Bugmann H., Weisberg P.J., Forest – ungulate and priorities for future research, Forestry, 2001, interactions: monitoring. modeling and 74, 259–270 management, J. Nat. Conserv., 2003, 10, 193-201

15 Autumn-winter diet overlap of fallow, red, and roe deer in forest ecosystems, Southern Poland

[10] Nugent G., McShea W.J., Parkes J., Woodley [24] Stromayer K.A.K., Warren R.J., Are overabundant S., Waithaka J., Moro J., et al., Policies and deer herds in the eastern United States creating management of overabundant deer (native or alternate stable states in forest plant communities? exotic) in protected areas, Anim. Prod. Sci., 2011, Wildlife Soc. B., 1997, 25, 227–234 51, 384–389 [25] Husheer S.W., Frampton Ch.M., Fallow deer [11] Putman R., Apolonio M., Andersen R., Ungulate impacts on Wakatipu beech forest, New Zeal. J. management in Europe. Problems and practices, Ecol., 2005, 29, 83–94 Cambridge University Press, Cambridge, 2011 [26] Wehausen J.D., Elliott H.W., Range relationships [12] Clutton-Brock T.H., Guiness F.E., Albon S.D., Red and demography of Fallow and Axis Deer on Point deer: behavior and ecology of two sexes, University Reyes National Seashore, California Fish and of Chicago Press, Chicago, 1982 , 1982, 68, 132–145 [13] Danilkin A., Hewison A.J.M., Behavioural ecology [27] Putman R.J., Competition and coexistence in a of Siberian and European roe deer, Chapman & multi-species grazing system, Acta Theriol., 1986, Hall, London, 1996 31, 271–291 [14] Apollonio M., Andersen R., Putman R. [Eds.], [28] Garcia-Gonzalez R., Cuartas P., Food habits of European ungulates and their management in pyrenaica, Cervus elaphus, and Dama the 21st century, Cambridge University Press, dama in the Cazorla Sierra (Spain), Mammalia, Cambridge, 2010 1992, 56, 195–202 [15] Jackson J., The annual diet of the Fallow deer [29] Putman R.J., Culpin S., Thirgood S.J., Dietary (Dama dama) in the , , as differences between male and female fallow deer determined by rumen content analysis, J. Zool., in and in allopatry, J. Zool., 1993, 229, 1977, 181, 465–473 267–275 [16] Caldwell J.F., Chapman D.I., Chapman N., [30] Poli B.M., Focardi S., Tinelli A., Composition and Observations on the autumn and winter diet of fallow metabolizable energy of feed used by fallow deer deer (Dama dama), J. Zool., 1983, 201, 559–563 (Dama dama) in a coastal Mediterranean ecosystem, [17] Borkowski J., Obidziński A., The composition of the Small Res., 1996, 22, 103–109 autumn and winter diets in two Polish populations [31] Chapman D., Chapman N., Fallow deer: their of fallow deer, Acta Theriol., 2003, 48, 539–546 history, distribution and biology. The 2nd ed., Coch- [18] Heinze E., Boch S., Fischer M., Hessenmöller y-bonddu Books, Machynlleth, 1997 D., Klenk B., Müller J., et al., Habitat use of large [32] Azorit C., Tellado S., Oya A., Moro J., Seasonal and ungulates in northeastern Germany in relation to specific diet variations in sympatric red and fallow forest management, For. Ecol. Manage., 2011, deer of southern Spain: a preliminary approach 261, 288–296 to feeding behaviour, Anim. Prod. Sci., 2012, 52, [19] Bruno E., Apollonio M., Seasonal variations in the 720–727 diet of adult male fallow deer in a submediterranean [33] Morse B.W., McElroy M.L., Miller K.V., Seasonal coastal area, Rev. Ecol.-Terre Vie, 1991, 46, 349– diets of an introduced population of fallow deer on 362 Little St. Simons Island. Georgia, Southeast. Nat., [20] Kay S., Factors affecting severity of deer browsing 2009, 8, 57–58 damage within coppiced woodlands in the south of [34] Nugent G., Forage availability and the diet of fallow England, Biol. Conserv., 1993, 63, 217–222 deer (Dama dama) in the Blue Mountains, Otago, [21] Moore N.P., Hart J.D., Langton S.D., Factors New Zeal. J. Ecol., 1990, 3, 83–95 influencing browsing by fallow deer Dama dama [35] Chapman D.I., Chapman N., Observations on the in young broad-leaved plantations, Biol. Conserv., biology of fallow deer (Dama dama) in . 1999, 87, 255–260 Essex. England, Biol. Conserv., 1969, 2, 55–62 [22] Gill R.M.A., Morgan G., The effects of varying deer [36] Focardi S., Aragno P., Montanaro P., Riga F., Inter- density on natural regeneration in woodlands in specific competition from fallow deer Dama dama lowland Britain, Forestry, 2010, 83, 53–63 reduces habitat quality for the Italian roe deer [23] Trabalza-Marinucci M., Capecci A., Riganelli Capreolus capreolus italicus, Ecography, 2006, 29, N., Acuti G., Antonini C., Olivieri O., Dietary 407-417 preferences and ruminal protozoal populations in [37] Feretti F., Sforzi A., Lovari S., Behavioural roe deer (Capreolus capreolus), fallow deer (Dama interference between ungulate species: roe are not dama) and ( musimon), Ital. J. Anim. on velvet with fallow deer, Behav. Ecol. Sociobiol., Sci., 2005, 4, 401–403 2011, 65, 875-887

16 A. Obidziński et al.

[38] Kerridge F.J., Bullock D.J., Diet and dietary quality [51] Levins R., Evolution in changing environment, of red deer and fallow deer in late summer, J. Zool., Princeton University Press, Princeton, 1968 1991, 224, 333–337 [52] Hurlbert S.H., The measurement of niche overlap [39] Gębczyńska Z., Food of the roe deer and red deer and some relatives, Ecology, 1978, 59, 67–77 in the Białowieża Primeval Forest, Acta Theriol., [53] Sneath P.H.A., Sokal R.R., Numerical . 1980, 25, 487–500 The principles and practice of numerical [40] Dumont A., Ouellet J., Crête M., Huot J., Winter classification, W.H. Freeman & Company, San foraging strategy of white-tailed deer at the northern Francisco, 1973 limit of its range, Ecoscience, 2005, 12, 476-484 [54] Conover W.J., Practical Nonparametric Statistics. [41] Verheyden-Tixier H., Renaud P.-C., Morellet N., Third Edition, John Wiley & Sons, New York, 1999 Jamot J.,•Besle J.-M., Dumont B., Selection for [55] Schoener T.W., The Anolis lizards of Bimini: nutrients by red deer hinds feeding on a mixed resource partitioning in a complex fauna, Ecology, forest edge, Oecologia, 2008, 156, 715–726 1968, 49, 704–726 [42] Matyjaszczyk M., Forest Management Plans of [56] Butts G.L., Anderegg M.J., Armstrong W.E., Harmel Forest District Rudziniec for period from 2006.01.01 D.E., Ramsey C.W., Sorola S.H., Food habits of to 2015.12.31, vol. 1. General Description of five exotic ungulates on Kerr Wildlife Management Forests. Analysis [Plan Urządzania Lasu dla Area, Texas Parks and Wildl. Dep. Tech. Ser., Nadleśnictwa Rudziniec na okres od 1 stycznia 1982, 30 2006r. do 31 grudnia 2015 r. tom I. Ogólny opis [57] Ivlev V.S., Experimental ecology of the feeding lasu. Elaborat], Biuro Urządzania Lasu i Geodezji of fishes, Yale University Press, New Haven, Leśnej w Brzegu, RDLP Katowice, 2006 (in Polish) 1961 [43] Miścicki S., Szerszenowicz A., Szerszenowicz K., [58] Gebert C., Verheyden-Tixier H., Variations of diet Szukiel E., The state of post-fire forest regeneration composition of red deer (Cervus elaphus L.) in and damage by cervids in the Rudy Raciborskie Europe, Mammal Rev., 2001, 31, 189–201 Forest District. Sylwan, 2002, 146, 43–55, (in [59] Cornelis J., Casaer J., Herm M., Impact of Polish with English summary) season, habitat and research techniques on diet [44] Gmitrzuk K., Obidziński A., Do large-scale forest composition of roe deer (Capreolus capreolus): a regenerations offer to deer more atractive fodder review, J. Zool., 1999, 248, 195–207 than mature forest: a case of burn out forest in [60] Prins R.A., Geelen M.J.H., Rumen characteristics Rudy Raciborskie, Sylwan, 2007, 4, 36−43, (in of red deer, fallow deer and roe deer, J. Wild. Polish with English summary) Manage., 1971, 35, 673–680 [45] Mysterud A. Lian L-B., Hjermann D.Ø., Scale- [61] Hofmann R.R., Evolutionary steps of dependent trade-offs in foraging by European roe ecophysiological adaptation and diversification of deer (Capreolus capreolus) during winter, Can. J. : a comparative view of their digestive Zool., 1999, 77, 1486–1492 system, Oecologia, 1989, 78, 443–457 [46] Ruckstuhl K.E., Neuhaus P., Sexual segregation in [62] Latham J., Staines B.W., Gorman M.L., Comparative ungulates: a comparative test of three hypotheses, feeding ecology of red (Cervus elaphus) and roe Biol. Rev., 2002, 77, 77–96 deer (Capreolus capreolus) in Scottish plantation [47] Chamrad A.D., Box T.W., Point frame for sampling forests, J. Zool., 1999, 247, 409-418 rumen contents, J. Wildlife Manage., 1964, 28, [63] Pianka E.R., Niche overlap and diffuse competition, 473–477 Proc. Nat. Acad. Sci., 1974, 71, 2141–2145 [48] Storms D., Aubry P., Hamann J-L., Said S., Fritz [64] De Boer W.F., Prins H.H.T., Large H., Saint-Andrieux C., Seasonal variation in diet that strive mightily but eat and drink as friends, composition and similarity of sympatric red deer Oecologia, 1990, 82, 264-274 Cervus elaphus and roe deer Capreolus capreolus, [65] Bertolino S., Di Montezemolo N.C., Bassano B., Wildl. Biol, 2008, 14, 237-250 Food–niche relationships within a guild of alpine [49] Gaare E., Sörensen A., White R.G., Are rumen ungulates including an , J. Zool., samples representative of the diet? Oikos, 1978, 2009, 277, 63–69 29, 390–395 [66] Dolman P.M., Wäber K., Ecosystem and [50] McInnis M.L., Vavra M., Krueger W.C., A competition impacts of introduced deer, Wildlife Comparison of four methods used to determine the Res., 2008, 35, 202–214 diets of large herbivores, J. Range Manage., 1983, [67] Putman R., The natural history of deer, Comstock 36, 302–306 Pub. Associates, Ithaca N.Y, 1988

17