A REVIEW OF THE METHODS FOR MONITORING ROE DEER EUROPEAN POPULATIONS WITH PARTICULAR REFERENCE TO ITALY

Page created by Carmen Norris
 
CONTINUE READING
Hystrix It.
Hystrix It. J.
            J. Mamm.
               Mamm. (n.s.)
                      19(2)19
                            2008:  103-12023-40
                              (2) (2008):

    A REVIEW OF THE METHODS FOR MONITORING ROE
    DEER EUROPEAN POPULATIONS WITH PARTICULAR
                 REFERENCE TO ITALY

     ALBERTO MERIGGI1,2, FRANCESCA SOTTI2, PAOLO LAMBERTI2,
                          NICOLA GILIO2
1
 Dipartimento di Biologia Animale, Università di Pavia, Piazza Botta 9, 27100 Pavia, Italy;
                                   e-mail: meriggi@unipv.it
 2
   Centro Interuniversitario per la Ricerca sulla Selvaggina e i Miglioramenti Ambientali a
               fini Faunistici, Piazzale delle Cascine 18, 50144 Firenze, Italy

                       Received 20 September 2008; accepted 22 December 2008

ABSTRACT - Throughout the last century, deer populations have shown a remarkable in-
crease both in North America and Europe. As a consequence, the estimate of roe deer den-
sity has become a matter of interest. We reviewed the available literature on the methods
used for monitoring roe deer populations in Europe from 1950 to 2004, with the aim of de-
tecting the trend of papers and distribution of census techniques by years, countries and
habitat types. Particular attention was paid to the census and monitoring methods adopted
in Italy and in Tuscany, which is the region where the roe deer is more carefully managed.
Published papers showed an increasing trend, as did the number of methods used and their
complexity. France, Italy, UK, and Spain were the countries with the richest literature and
the largest variety of methods applied. Eleven census methods have been applied in woods -
particularly line transects, pellet group counts, CMR and IKA - with only 6 in open coun-
try, mainly pellet group counts. In Europe vantage points are more commonly used for
planning culling programs, whilst in Italy, and particularly in Tuscany, the drive census and
spotlight counts are mainly used. Unfortunately, in Europe, harvesting programs are still
too much based on hunter knowledge and traditions. However the countries where the man-
agement of roe deer hunting is of more recent tradition make an exception to this rule. In
Italy and in Tuscany the methods of monitoring roe deer populations should be improved
towards less expensive and more accurate methods.

Key words: Capreolus capreolus, monitoring methods, culling, management, Europe, Italy

RIASSUNTO – Revisione dei metodi di monitoraggio delle popolazioni di capriolo in Eu-
ropa con particolare riferimento all’Italia. Le popolazioni di Cervidi hanno avuto
nell’ultimo secolo un notevole incremento sia in America settentrionale, sia in Europa. Di
conseguenza la densità delle popolazioni di capriolo è diventata oggetto di interessi diversi
e la sua stima suscita grande attenzione. Con lo scopo di evidenziare l’andamento delle
pubblicazioni e la distribuzione dei metodi di censimento per anno, Paese e tipo di habitat,
abbiamo analizzato la letteratura europea sui metodi di monitoraggio delle popolazioni di
capriolo dal 1950 al 2004. Particolare attenzione è stata data ai metodi normalmente adotta-
ti in Italia e in Toscana, che è la regione dove la gestione del capriolo si può considerare più
avanzata. I lavori pubblicati hanno avuto una tendenza all’aumento così come il numero di
metodi utilizzati e la loro complessità. La Francia, l’Italia, il Regno Unito e la Spagna sono

                                               103
Meriggi et al.

risultati i Paesi con la letteratura più ricca su questo argomento e anche quelli dove è stato
sperimentato il maggior numero di metodi. Undici metodi di censimento - soprattutto tran-
setti lineari, conteggi di pallottole fecali, cattura-marcatura-ricattura e indice chilometrico
d’abbondanza -, sono stati usati negli ambienti boscati e solamente 6 nelle aree aperte, in
particolare il conteggio delle pallottole fecali. In Europa il metodo più usato per pianificare
il prelievo con la caccia di selezione è risultato quello dei punti dominanti, mentre in Italia e
particolarmente in Toscana sono principalmente usati le battute su aree campione e i con-
teggi notturni. Sfortunatamente in Europa i piani di prelievo sono ancora troppo basati sulle
conoscenze e sulla tradizione dei cacciatori. Tuttavia, fanno eccezione a questa regola i Pa-
esi dove la gestione venatoria del capriolo è iniziata di recente. In Italia e in Toscana i me-
todi di monitoraggio delle popolazioni di capriolo dovrebbero essere migliorati, adottando
tecniche meno dispendiose e più accurate.

Parole chiave: Capreolus capreolus, metodi di monitoraggio, piani di prelievo, gestione,
Europa, Italia

INTRODUCTION                                        (Staines and Ratcliffe, 1987; Cederlund
                                                    et al., 1998; Duncan, 1998; Radeloff et
Throughout the last century, deer popu-             al., 1999).
lations have remarkably increased both              Roe deer density has become the object
in North America and Europe (Gill,                  of different interests: hunters, natural-
1990), probably as a consequence of                 ists and tourists that want higher densi-
the general improvement of habitat                  ties, and, on the other hand, stake-
quality and availability together with              holders, farmers and foresters that
that of game management (Gill et al.,               would prefer low or even zero densi-
1996; Cederlund et al., 1998). The                  ties. Nevertheless the roe deer, as all
European roe deer (Capreolus capreo-                the other ungulate species, must be also
lus) derived much benefit from the                  considered as an important prey for
widespread decrease of intensive stock-             threatened and protected carnivores
breeding and the abandonment of agri-               such as the wolf (Canis lupus) and lynx
culture on the hills and mountains, fol-            (Lynx lynx) (Mattioli et al., 1995;
lowed by an increase of woodland.                   Okarma, 1995; Meriggi and Lovari,
Moreover, in the last forty years, the              1996; Meriggi et al., 1996; Aenes et
game management of roe deer popula-                 al., 1998; Mattioli et al., 2004; Gazzola
tions has improved, providing shorter               et al., 2005; Meriggi et al., in press).
hunting seasons and limited quotas,                 Considering all these aspects, roe deer
food supply and habitat improvement,                management should gain a sort of bal-
reintroductions and restocking. On the              ance between harvesting and conserva-
other hand, if the consequent growth in             tion of habitats and species without af-
density and distribution of roe deer                fecting human activities. In other
populations has brought benefits to                 words, different strategies should be
hunters, the increasing damage to for-              adopted to adjust population density
estry and agriculture has enlarged the              according to different habitats and eco-
social component interested in the                  systems.
proper management of this species                   Several examples of different harvest

                                              104
Monitoring roe deer populations

plans are reported in the available scien-      different across Europe. In the Mediter-
tific and technical literature; most of         ranean area, for example, roe deer
them are focused on a single objective,         were, and somewhere still are, hunted
such as maximising harvest quotas or            by drives with hunting dogs without
trophy hunting. Harvest plans can be            any selective and sustainable harvest-
qualitative, when based on criteria such        ing criterion. Otherwise, in central
as the distribution of age or sex classes       Europe roe deer are traditionally hunted
and trophy size (Von Raesfeld, 1985;            for their trophies, so only adult males
Kurt, 1991), or quantitative, when              and individuals of poor quality (i.e. old,
based on the estimate of fecundity and          sick or injured ones) were culled, often
productivity (Ellenberg, 1975; Blant,           ignoring young and females.
1991).                                          Progressively, harvest plans and tech-
The adoption of one approach or the             niques have evolved everywhere to-
other may depend on the main man-               wards culling plans that respect the sex
agement objective - e.g. population             and age class proportions, according to
control vs. the rational exploitation of        the consciousness that each population
the population -, the information               has its own characteristics and parame-
needed varying drastically in relation to       ters. As a result, the census of deer
the different circumstances (Cederlund          populations has become a priority to
et al., 1998). In the case of a small har-      formulate culling plans, following the
vest quota, compared to the population          idea that the more accurate the census
abundance (e.g. trophy hunting), it is          the higher the probability of avoiding
sufficient to evaluate by direct counts         the degeneration of the population to-
the minimum number of adult males. If           wards either overabundance or scarcity.
the management objective is to reduce           In this review we collected the avail-
deer damage to forests or crops, it is          able information on roe deer monitor-
necessary to estimate the whole popula-         ing methods that have been applied in
tion size, particularly the proportion of       Europe and Italy, in particular in Tus-
adult females (Ratcliffe, 1987; Lubow           cany, where the species is widespread
et al., 1996). Finally, if the objective        and more carefully managed. Data
was to achieve the maximum sustain-             were obtained from the international
able exploitation, the population re-           and national scientific literature as well
cruitment should be known, as also the          as from technical reports and congress
intrinsic and external factors that influ-      proceedings.
ence it. In any case, especially for the
last target, a sound knowledge of the           MATERIALS AND METHODS
main parameters of population dynam-
ics represents the first step for a correct     To investigate the census methods usually
management of roe deer populations.             adopted for roe deer populations in Europe,
                                                we analysed the bibliography available for
Hunting techniques have changed over
                                                the period 1950 - 2004. From 1990 on-
the years as a consequence of increased         wards, the scientific indexed papers were
public awareness promoted by the ac-            searched for using the CILEA digital li-
quisition of scientific knowledge of roe        brary, by means of ISI Web of Science and
deer populations. This process has been         Current Contents Connect (CCC). The

                                          105
Meriggi et al.

Wildlife Service software was used to               the census techniques adopted in each
search for and consult the abstracts of all         game management unit.
the literature published before 1990; above
all, this database yielded a list of references     RESULTS
for non indexed journals, congress and
workshop proceedings. Moreover, we con-             1. Evolution of roe deer census methods
sulted books and reports on the conserva-
tion and management of ungulates, particu-
larly the roe deer, in Europe and Italy.            The first census methods adopted for
The following journals were consulted:              management and scientific purposes -
Journal of Animal Ecology (1968-2004),              the drive census and direct observations
Journal of Applied Ecology (1968-2004),             - derived from hunting practices (De
Acta Theriologica (1979-2004), Revue                Crombrugge, 1969; Boisaubert and
D’Ecologie (Terre e Vie) (1979-2004), The           Stoquert, 1975; Boisaubert et al., 1979;
Journal of Wildlife Management (1981-               Von Berg, 1979; Cederlund et al.,
2004), Gibier Faune Sauvage (1984-1998),            1984; Denis, 1985; Blant, 1987). These
Ethology Ecology & Evolution (1989-                 techniques have been largely used,
2004), Ecography (1994-2004), Oikos
                                                    even if some authors pointed out the
(1994-2004), Wildlife Biology (1995-2004).
The references of each paper of some inter-         risk of underestimating the actual roe
est were searched for and consulted, and            deer density (Cederlund et al., 1998;
the same was done for the two most impor-           Van Laere et al., 1998), whilst others
tant books about the roe deer (Danilkin and         authors asserted their accuracy
Hewison, 1996; Andersen et al., 1998).              (Boscardin, 1999). Particularly, with
Then all papers were ordered and listed per         the drive census technique it is very
year of publication, country, study period,         difficult to apply an adequate sampling
census techniques, census season and habi-          strategy and a sufficient number of rep-
tat. All these data were fully available only       licates (Van Laere et al., 1998). The
for some papers.
                                                    impossibility of replicating the census
The analyses were conducted in order to
obtain: a) the 1950-2004 trend of papers            in the same conditions prevents the cal-
dealing with fully explained census tech-           culation of confidence intervals and
niques, b) the distribution of papers per           standard errors (Cederlund et al.,
country and the frequency distribution of           1998), which are necessary to obtain
census techniques c) per year, d) country,          comparable data (Fattorini and Pisani,
e) habitat and f) season.                           1999; Fattorini, 2000).
Moreover, the census techniques adopted             Successively less laborious techniques
by the major European wildlife services             - spotlight road counts, sector census,
and research institutions have been investi-        vantage points census, line transects
gated to define the main methods used to
                                                    and pellet-group counts -, have been
note down culling plans. In particular, for
Italy, all the Regional and Provincial regu-
                                                    adopted, with the main aim of reducing
lations on the management of ungulates              the number of people involved, and
were collected, in order to find out the offi-      sample size. All these methods have
cial rules on this matter. Finally, a more          been scarcely used in the past, due to
intensive investigation was conducted in            environmental constraints and the lack
Tuscany by direct inquiries with game-              of scientific tests on roe deer popula-
keepers and wildlife managers, to define            tions (Cederlund et al., 1998), even if

                                              106
Monitoring roe deer populations

sometimes their potential and applica-              (1953) exterminated a roe deer popula-
bility has been shown (Mitchell et al.              tion that demonstrated a three times
1985; Staines and Ratcliffe, 1987;                  more abundant estimate than had pre-
Ratcliffe and Mayle, 1992; Boscardin,               viously been recorded by direct obser-
1999; Crépin et al., 1999; Mayle et al.,            vations and drive counts. Successively,
2000).                                              Strandgaard (1972), Pielowski (1984)
Among them, line transects are becom-               and Ratcliffe (1987) reported similar
ing more and more used in the monitor-              cases, the solitary and secretive behav-
ing of wildlife populations (Burnham et             iour of roe deer, particularly in forested
al., 1980, 1981; Seber, 1992; Buckland              habitat, enhancing the probability of
et al., 1993). This method is based on              population underestimates (Gaillard et
the count of animals observed on both               al., 1993).
sides of a standardised transect of                 One of the first methods adopted to
known length; for each individual or                evaluate the accuracy of the different
group of individuals detected the per-              census techniques and for research pur-
pendicular distance from the transect               poses was the Petersen-Lincoln method
must be measured. In order to obtain                (Petersen, 1896; Lincoln, 1930), also
reliable population estimates and con-              called the Capture-Mark-Recapture
fidence intervals some assumptions                  (CMR) method (Andersen, 1962;
should be fulfilled: a) all the individu-           Strandgaard, 1967, 1972; Gaillard et
als on the transect must be observed, b)            al., 1986; Gill et al., 1996; Andersen et
the individuals are observed in their               al., 1995).
initial position, c) the observations are           This method is hardly suitable for man-
independent events. This technique is               agement, being time-expensive. Never-
still poorly adopted for roe deer, even if          theless, population estimates and confi-
some encouraging results have been de-              dence intervals are generally reliable if
rived from its application in various               the following assumptions are fulfilled
habitats (Zejda, 1984, 1985; Alvarez                (Strandgaard, 1972; Caughley, 1977;
Jimenez, 1988; Fandos et al., 1990;                 Seber, 1982): a) the population is
Gaillard et al., 1993). Particularly, roe           closed, b) both marked and unmarked
deer are easily detectable at night, so             individuals have the same probability
this method has been adopted with                   of being recaptured or re-sighted, c) the
good results using night-vision and                 number of marked animals in the popu-
thermal-imaging equipment; however                  lation is known at any time, d) the pro-
the costs of this equipment still repre-            portion of marked individuals is around
sent a limiting factor (Gill et al., 1997).         60%. Recently several software pack-
Line transects methods are increasingly             ages have been released that yield pre-
used by researchers, whilst managers                cise calculations, even when not all
are still reluctant because not all the as-         these assumptions are fulfilled (Pollock
sumptions can be verified and some-                 et al., 1990).
time few observations are recorded.                 The CMR method is considered the
Since the first time that roe deer census           reference one when other methods are
estimates have been checked, their in-              to be tested. Obviously this means that
accuracy has been clear. Andersen                   suitable experimental areas with a lot

                                              107
Meriggi et al.

of marked animal have to be available.            habitat suitability models that incorpo-
On the other hand, nowadays the need              rate even small habitat changes and
for methods that fit the various levels           population dynamics. The adoption of a
of management strategies and popula-              Graphical User Interface (GUI) makes
tion density is declared.                         them adaptable to any local situation, in
Recently, during the VI European Roe              order to enhance the accuracy of the
Deer Congress that took place the 23-             estimates of roe deer density through
26th April 2003 in Portugal, it emerged           the computation of annual habitat
that the more popular and widespread              changes and the culling effort of previ-
census methods (drive counts, line                ous years (Radeloff et al., 1999).
transects, pellet-group counts, etc.)
show limitations in some situations and           2. Trend and distribution of census
a solution for all density levels and             methods
habitats is still far from achieved. A
flexible and well-studied approach is             Sixty-one papers on roe deer census
necessary. Especially in France, a set of         methods in Europe were found. The
indicators is being developed for roe             published papers were unevenly dis-
deer populations and their habitats               tributed among years, only 4 before
(Ballon et al., 1991; Van Laere et al.,           1975 and a remarkable increase starting
1998). All these indexes are based upon           from the late 1970s until 1985, when
biological parameters that were se-               the number of published papers stabi-
lected for being easily measurable and            lized around a mean of 3 papers per
closely correlated with population den-           year, with a peak of 7 papers in 2001
sity and resource availability (Milner-           (Fig. 1). Paper distribution among
Gulland and Akcakaya, 2001). The                  European countries was overbalanced
most widely used are the kilometric in-           towards France (47.5%), Italy (14.8%),
dex of abundance (IKA, Vincent et al.,            the United Kingdom (9.8%) and Spain
1991), body mass (Maillard et al.,                (8.2%), while in the rest of Europe pa-
1989; Gaillard et al., 1996; Van Laere            pers were lacking or absent (Fig. 2).
et al. 1998), ovulation rate (Liberg et           The most common methods were line
al., 1991; Gaillard et al., 1992), fawn           transect or distance sampling (22.8%),
production (Boutin et al., 1987; Liberg           pellet group counts (17.5%), the CMR
et al., 1991; Vincent et al., 1995), jaw          (often involving re-sighting instead of
or hind foot length (Hewison et al.,              recapture; 14.0%) and the kilometric
1996; Van Laere et al., 1998) and                 index of abundance (IKA; 12.3%). The
browsing index (Mabille and Neet,                 other methods were less used even if all
1994; Ballon, 1994; Guibert, 1997).               the pooled ecological and biological
Unfortunately, none of them give in-              indexes reached 10.7% (Fig. 3). Con-
formation about the population struc-             sidering the frequency of use of the dif-
ture, limiting their application for an           ferent methods, a tendency to diversifi-
accurate culling plan.                            cation and experimentation clearly ap-
Finally, a new tendency in roe deer               peared through the years, with the pro-
management is to apply, by the Geo-               gressive abandonment of expensive and
graphical Information System (GIS),               time consuming methods such as the

                                            108
Monitoring roe deer populations

drive census and CMR in favour of                each) and Italy (4). In the other Euro-
population indexes (Fig. 4).                     pean countries only one or two meth-
France was the country where the larg-           ods have been used (Fig. 5).
est number of methods has been ap-               Only for 35 out of 61 papers it was
plied (8), followed by UK, Spain (7              possible to define the habitat where the

Figure 1 - Trend of the papers on monitoring methods of roe deer populations in Europe.

Figure 2 - Distribution of papers on roe deer monitoring methods in the main European
countries.

                                           109
Meriggi et al.

census methods were applied. The dif-            in particular, line transects, pellet group
ferent habitat types were grouped into           counts, CMR and IKA were the most
two macro-habitats: open country
                                                 used methods in the former, whereas in
(crops, pastures, grassland, etc.) and
woods. Eleven methods were applied in            the latter pellet group counts prevailed
woods, whilst only 6 in open country;            (Fig. 6).

Figure 3 - Frequency distribution of the main roe deer monitoring methods.

Figure 4 - Percentage of occurrence of the different roe deer monitoring methods from 1950
to 2004.

                                           110
Monitoring roe deer populations

Figure 5 - Percentage of occurrence of the different roe deer monitoring methods in some
European countries.

Figure 6 - Percentage of occurrence of the different roe deer monitoring methods in open
and wooded habitats

                                         111
Meriggi et al.

3. Roe deer census methods in Europe                4. Roe deer census methods in Italy

The abundance of roe deer populations               In Italy the roe deer is present in 67 out
that are regularly culled has been esti-            of 103 provinces (65.0%) and hunting
mated by various methods: in some                   is allowed in 38 (36.9%), usually by
European countries no census has ever               selective criteria except for some north-
been applied, while in other countries              eastern provinces (Pordenone, Udine,
sophisticated methodologies have been               Treviso, Vicenza, Belluno) were the
used, even for research purposes. The               species is also hunted by drives with
most used methods in planning deer                  dogs (Pedrotti et al., 2001, Apollonio,
culling were the measure of body                    2004). Roe deer censuses are con-
weight and jaw length and age estima-               ducted in 34 provinces, that is 50.7% of
tion by tooth eruption and consump-                 the provinces where the species is pre-
tion. The census methods more com-                  sent and 89.5% of those where the spe-
monly applied were vantage points                   cies is hunted (Apollonio, 2004).
counts, either complete in open areas or            Considering both regional and provin-
according to a sampling design in                   cial regulations, only in five provinces
wooded areas (Tab. 1).                              belonging to two regions (Piedmont and

Table 1 - Monitoring methods of roe deer populations applied in some European countries.

Country             Census methods
Finland             snow-tracking
Sweden              snow-tracking, pellet groups
Norway              sanitary surveillance on culled animals
Denmark             drive census, line transect
United Kingdom      drive census, vantage points, pellet groups
France              vantage points, sector counts, populations indexes
Germany             vantage points
Austria             browsing index, direct counts, culling statistics, pellet groups, snow-tracking
Poland              drive census, snow-tracking, direct observations.
Czech Republic      sector counts in spring
Hungary             vantage points, IKA, culling statistics
Bulgaria            no census
Romania             no census
Serbia              vantage points, IKA
Croatia             vantage points, IKA.
Slovenia            vantage points
Spain               no census

                                              112
Monitoring roe deer populations

Tuscany) were the periods and methods               formed by hunters form March to May
for roe deer census provided. Usually,              and planned by the technicians of the
the normative simply refers to technical            Hunting Districts. The sampling area
report of the wildlife services or to               usually ranges from 4 to 7% of the total
agreement documents with the INFS                   wooded area of the Hunting District.
(National Institute for Wildlife).                  Vantage point censuses are performed
Vantage point counts was the most                   wherever the wood percentage is lower
used census method in all the provinces             than 50%, whilst in areas where the
where roe deer culling is allowed. The              previous two census techniques are not
drive census was mostly applied in                  suitable (e.g. Siena, Pisa and Grosseto
Liguria and Tuscany, whilst sector                  provinces) the IKA and population
counts and spotlight night counts pre-              structure are assessed by observations
vailed in alpine areas, where counts are            from transects (Tab. 2).
conducted using the roads in the valley             Roe deer are hunted also in private es-
bottom or those above the tree-line. In             tates; in this case shooting plans are de-
central Italy these techniques are com-             cided by the owner under the supervi-
monly used in hunting estates. In some              sion of the wildlife services of the
scientific researches line transects, pel-          provinces. Censuses are performed by
let group counts and capture-mark-                  the technicians of either the estate or
resight have been used (Focardi et al.,             the province, and usually the vantage
2002 a, b, c; Fattorini et al., 2004).              points technique or night counts are
                                                    adopted. In the province of Pisa, pri-
5. Roe deer census methods in Tuscany               vate estates adopt the IKA technique
                                                    and, once every three years, the drive
In Tuscany roe deer harvesting is con-              census.
ducted by means of quantitative and
qualitative shooting plans. The shoot-              DISCUSSION
ing plans, as reported in the regional
regulation, must be based on census                 In Europe, the scientific and technical
data and all hunters must take part in              literature on roe deer census methods
the census activities, under the control            and population monitoring has not been
of the wildlife services of the prov-               abundant in the past fifty-four years,
inces. Only qualified hunters can take              even if some papers may have been ne-
part in culling and all culled individuals          glected by mistake from this review,
must be recorded. Nevertheless, only in             especially the technical reports, and
some provinces are harvesting data                  PhD and degree theses that are not al-
analysed to obtain proper information               ways easy to locate.
on population structure and dynamics.               In the last fifty years, roe deer census
Despite these inconsistencies, popula-              techniques have taken two opposite
tion monitoring appears to be excellent             paths, as often happens when species
in Tuscany compared to the rest of It-              are the object of interest by hunters,
aly. The drive census is the most                   managers and researchers. Even if re-
adopted method in forest habitats (ex-              searchers have improved the accuracy
cept for the province of Prato). It is per-         and precision of census techniques,

                                              113
Meriggi et al.

managers have often ignored the up-               techniques that well fitted the purposes
date, preferring to keep the obsolete             of management.

Table 2 - Census methods of roe deer adopted by the province Administrations in Tuscany
region.

Province            Type of hunting area          Census method              Period
                                                  Drive census             March-May
Arezzo              Hunting districts
                                                  Vantage points           March-May
                                                  Drive census                  -
                    Hunting districts
                                                  Vantage points           Late spring
Firenze
                                                  Vantage points           Late spring
                    Private estates
                                                  Spotlight census              -
                                                  Drive census                April
                    Hunting districts
                                                  Vantage points              April
                                                  Drive census             April-June
Pisa
                                                  Vantage points          March-April
                    Private estates
                                                  Spotlight census          February
                                                  Line transects              April
                                                  Drive census            March-April
Livorno             Hunting districts
                                                  Vantage points          March-April
Pistoia             Hunting districts             Drive census                  -
                                                  Drive census                  -
Massa-Carrara       Hunting districts
                                                  Vantage points                -
                                                  Drive census                  -
Lucca               Hunting districts
                                                  Vantage points                -
                                                  Drive census                April
                    Hunting districts
                                                  Vantage points              April
                                                  Drive census             April-June
Siena
                                                  Vantage points          March-April
                    Private estates
                                                  Spotlight census          February
                                                  Line transects              April
                                                  Drive census            March-April
                    Hunting districts             Vantage points              April
                                                  Line transects              April
Grosseto
                                                  Drive census            March-April
                    Private estates               Vantage points              April
                                                  Spotlight census       February-March
Prato               Hunting districts             Vantage points                -

                                            114
Monitoring roe deer populations

After the pioneering papers from the              selective pressure and suffer high mor-
Danish school in the fifties, which               tality rates; as a matter of fact the sex-
highlighted the high risk of underesti-           ratio is strongly shifted towards fe-
mating deer populations by the most               males in many natural population, even
frequently used census methods, a pe-             if local variability is high (Clutton-
riod of stasis lasted until the second            Brock, 1991). Therefore reliable census
half of the 1980s. Then the frequency             data are fundamental to prevent selec-
of publications increased and became              tive hunting from altering the structure
considerable in the first years of the            and life-history traits of natural popula-
1990s. At the same time, the census               tions (Gadgil and Bossert, 1970;
methods adopted and verified increased            Dhondt, 1991).
in number, and biological indicators              In some countries scientific research
have been developed for monitoring                has been greater compared with the
roe deer populations.                             European mean. In particular, 80% of
This trend could be explained by a                papers were published in only four
combination of different factors: a) the          countries - France, Italy, the United
increase of the European deer popula-             Kingdom and Spain. These countries
tions and the consequent rise of man-             are those where selective hunting is of
agement and economic interest, b) the             recent tradition and deer populations
increase of knowledge on roe deer bi-             have showed a marked increase, proba-
ology and of the concern about their              bly stimulating the research.
proper management, c) the increasing              Notwithstanding the increasing knowl-
consciousness about the unreliability of          edge of roe deer biology and the ex-
traditional census methods for hunting            perimentation of improved monitoring
management, d) the necessity of har-              methods, it seems that in management
vest programs for reducing deer dam-              practices innovations have been rarely
age and browsing pressure and e) the              adopted and accepted throughout
need of less expensive and more accu-             Europe (e.g. France and United King-
rate census methods.                              dom). Unfortunately, harvesting pro-
In the last ten years, managers have              grams are still too much based on hunt-
discovered that every harvest program             ers’ knowledge and traditions and re-
that does not respect the natural struc-          ferred to low-scale geographical units,
ture, in terms of age and sex classes, of         without adopting objective census
each population can have huge effects             methods (Pollock et al., 2000).
on population dynamics and recruit-               In Italy, in spite of the lack of proper
ment (McIntosh et al., 1995). Recently,           regulations, it seems that roe deer man-
population genetics have shown that a             agement has reached an acceptable
high percentage of individuals are not            level, all shooting plans being based on
involved in reproduction, only few                census data. Unfortunately, no census
adults playing an important role in an-           is performed where roe deer hunting is
nual      recruitment    (Clutton-Brock,          forbidden. Among all regions, Tuscany
1988).                                            is top ranking in the organization and
Moreover, adult males, as happens in              diffusion of roe deer censuses, even if
all polyginic species, are under a strong         there is high local variability. A large

                                            115
Meriggi et al.

difference is evident comparing the               Andersen J. 1953. Analysis of the Danish
provinces where selective hunting                     roe deer population based on the ex-
started first (e.g. Siena and Arezzo)                 termination of the total stock. Danish
with the others. In the former, a larger              Rev. Game Biol., 2: 127-155.
                                                  Andersen J. 1962. Roe deer census and
portion of territory is sampled, many
                                                      population analysis by means of a
hunters participate in the census activi-             modified marking-release technique.
ties, biometrics and fecundity data of                In: Le Cren E.D. and Holdgate M.W.
culled animals are collected, while in                (eds), The exploitation of natural ani-
the provinces where selective hunting                 mal population. Blackwell Scientific
started in recent years census opera-                 Publications, Oxford, 72-82.
tions should be improved and biomet-              Andersen R., Duncan P. and Linnell J.D.C.
rics data are not collected.                          1998. The European roe deer: the biol-
The census methods adopted appear to                  ogy of success. Scandinavian Univer-
                                                      sity Press, Oslo, 373 pp.
be effective for management purposes,
                                                  Andersen R., Linnell J.D.C. and Aenes R.
even if sometimes they may be difficult               1995. Roe deer in an agricultural land-
to organize and apply. This is particu-               scape. Final report. NINA Fagrapport,
larly true for the drive census that is               10: 1-80.
time-expensive and scarcely efficient in          Apollonio M. 2004. Gli ungulati in Italia :
terms of the area covered.                            status, gestione e ricerca scientifica.
                                                      Hystrix It. J. Mamm., 15: 21-34.
ACKNOWLEDGEMENTS                                  Ballon P. 1994. Relations forêts/cervidés -
                                                      Vers une meilleure gestion. Informa-
                                                      tions techniques du Cemagref, 96(5):
This research was part of the project                 1-8.
“La verifica e innovazione dei metodi             Ballon P., Guibert B., Maizeret C. and
di censimento del capriolo” that was                  Boutin J.M. 1991. Contribution à la
funded by Tuscany Region and by the                   mise au point de nouvelles méthodes
Provinces of Arezzo and Pisa, and co-                 de gestion des populations de che-
ordinated by M. Apollonio (University                 vreuil. Application au massif des
of Sassari).                                          Landes de Gascogne. "Etudes" du
                                                      CEMAGREF, Département Forêt, sé-
                                                      rie forêt n° 6, Annales 90: 173-185.
REFERENCES
                                                  Blant M. 1987. Dynamique de population,
                                                      condition et constitution du chevreuil
Aenes R., Linnell J.D.C., Perzanowski K.,             (Capreolus capreolus L. 1758) dans
    Karlsen J. and Odden J. 1998. Roe                 les cantons de Neuchatel et Vaud
    deer as a prey. In: Andersen R., Dun-             (ouest de la Suisse). Unpublished
    can P. and Linnell J.D.C. (eds), The              Ph.D. Thesis, University of Neuchâtel,
    European roe deer: the biology of suc-            Switzerland.
    cess. Scandinavian University Press,          Blant M. 1991. Reproduction in roe deer
    Oslo, 139-159.                                    populations in Western Switzerland.
Alvarez Jimenez G. 1988. Problemas                    Proceedings of the 18th IUGB Con-
    associados a la aplicacion del transecto          gress, Krakow, 185-188.
    lineal para el cenzo de las problacion        Boisaubert B. and Stoquert M. 1975. Recen-
    de cervidos en un biotopo medi-                    sement de cervidés. Unpublished re-
    terraneo (Quintos de Mora, Montes de               port, Office National de la Chasse, Pa-
    Toledo). Ecologia, 2: 233-249.                     ris, 6 pp.

                                            116
Monitoring roe deer populations

Boisaubert B., Vassant J. and Delorme D.                   tions de chevreuils. Cemagref Note
    1979. Contribution a la mise au point                  technique n° 51, 64 pp.
    d’une méthode de recensement appli-              Clutton-Brock T.H. 1988. Reproductive
    cable à l’espèce chevreuil (Capreolus                 success. Chicago University Press,
    capreolus) vivant en milieu forestier.                Chicago.
    Bull. Mens. ONC, N°sp.sc.Tech, 93-               Clutton-Brock T.H. 1991. Sort and wise
    205.                                                  use of ungulate populations. Gibier
Boscardin Y. 1999. Les méthodes de dé-                    Faune Sauvage, 8: 309-317.
    nombrement des populations de che-               Crépin F., Bouteiller R., Gaillard J.M. and
    vreuils. Bull. Mens. ONC, 244: 17-21.                 Delorme D. 1999. Dénombrement des
Boutin J.M., Gaillard J.M., Delorme D.                    chevreuils en milieu ouvert: résultats
    and Van Laere G. 1987. Suivi de                       et application à la gestion des popula-
    l'évolution de la fécondité chez le                   tions. Bull. Mens. ONC, 244: 78-82.
    chevreuil par l'observation des                  Danilkin A.A. and Hewison A.J.M. 1996.
    groupes familiaux. Gibier Faune Sau-                  Behavioural ecology of Siberian and
    vage, 4: 255-265.                                     European roe deer. Chapman & Hall,
Buckland S.T., Anderson D.R., Burnham                     London.
    K.P. and Laake J.L. 1993. Distance               De Crombrugge S.A. 1969. Modes de re-
    sampling: Estimating Abundance of                     censement du cerf (Cervus elaphus L.)
    Biological Populations. Chapman and                   en Belgique et portée pratique. Pro-
    Hall, London.                                         ceedings of the IX th IUGB Congress,
Burnham K.P., Anderson D.R. and Laake                     Moscow, 298-306.
    J.L. 1980. Estimation of density from            Denis M. 1985. Quelques méthodes pra-
    line transect sampling of biological                  tiques pour l’estimation de l’effectif
    populations. Wildl. Monogr., 72: 1-202.               d’une population de chevreuils (Ca-
Burnham K.P., Anderson D.R. and Laake                     preolus capreolus L.). Proceedings of
    J.L. 1981. Line transect estimation of                the XVIIth IUGB Congress, Brussels,
    bird population density using a Fou-                  979-989.
    rier series. Studies in Avian Biology 6:         Dhondt A.A. 1991. Density dependence,
    446-482.                                              teta selection and the exploitation of
Caughley G. 1977. Analysis of vertebrate                  animal populations. Gibier Faune
    populations. John Wiley and Sons,                     Sauvage, 8: 297-307.
    London.                                          Duncan P. 1998. Managing roe deer and
Cederlund G., Bergqvist J., Kjellander P.,                their impact on the environment: max-
    Gill R.M.A., Gaillard J.M., Boisa-                    imizing the net benefits to society. In:
    ubert B., Ballon P. and Duncan P.                     Andersen R., Duncan P. and Linnell J.
    1998. Managing roe deer and their                     D. C. (eds), The European roe deer:
    impact on the environment: maximiz-                   the biology of success. Scandinavian
    ing the net benefits to society. In: An-              University Press, Oslo, 337–372.
    dersen R., Duncan P. and Linnell                 Ellenberg H.1975. Beitrage zur Okologie des
    J.D.C. (eds), The European roe deer:                  Rehes. Phd Thesis, Munich, 275 pp.
    the biology of success. Scandinavian             Fandos P., Fernandez J.M., and Palomero
    University Press, Oslo, 337–372.                      G. 1990. Censo de corzo en un sector
Cederlund G., Bergqvist J., Kjellander P.,                de la reserva nacional de caza de
    Gill R.M.A., Gaillard J.M., Boisa-                    Sajaliebana. Informe por la diputacion
    ubert B. and Ballon P. 1984. Mé-                      regional da Cantabria, Universidad de
    thodes de recensement des popula-                     Cantabria.

                                               117
Meriggi et al.

Fattorini L. 2000. Statistical estimation of                  winter in two contrasting populations.
     abundance in wild ungulate popula-                       J. Wildl. Manage., 60: 29-36.
     tions. Proceedings of the Congress                  Gaillard J.M., Lebreton J.D., Pontier D.
     “Gestione degli ungulati selvatici:                      and Landry P. 1992. Demographic
     problemi e soluzioni”. Università de-                    sensitivity and population manage-
     gli Studi di Perugia, Perugia, 50-64.                    ment: an application to roe deer
Fattorini L. and Pisani C. 1999. Metodi di                    (Capreolus capreolus). Proceedings of
     campionamento per le indagini am-                        the 18th IUGB Congress, Krakow, 2,
     bientali. Facoltà di Economia "R. Go-                    547-550.
     odwin", Università degli Studi di Sie-              Gazzola A., Bertelli I., Avanzinelli E.,
     na, Siena, 213 pp.                                       Tolosano A., Bertotto P. and Apollo-
Fattorini L., Pisani C. and Sforzi A. 2004.                   nio M. 2005. Predation by wolves
     The estimation of wild ungulate                          (Canis lupus) on wild and domestic
     abundance using sample area surveys:                     ungulates of the western Alps, Italy. J.
     an application to Maremma Regional                       Zool., 266: 205-213.
     Park. Satistical Methods & Applica-                 Gill R.M.A. 1990. Monitoring the status of
     tions, 13: 197-212.                                      European and North American cervids.
Focardi S., Isotti R., Pelliccioni E.R. and Ian-              GEMS Information Series N°8, Global
     nuzzo D. 2002a. The use of distance                      Environment Programme, Nairobi,
     sampling and mark-resighting to estimate                 Kenia, 277 pp.
     the local density of wildlife populations.          Gill R.M.A., Johnson A.L., Francis A.,
     Environmetrics, 13: 177-186.                             Hiscoks K. and Peace A.J. 1996.
Focardi S., Isotti R., Tinelli A. 2002c. Line                 Changes in roe deer (Capreolus ca-
     transect estimates of ungulate popula-                   preolus L.) population density in re-
     tions in a Mediterranean forest. J.                      sponse to forest habitat succession.
     Wildl. Manage., 66: 48-58.                               For. Ecol. Manage., 88: 31-41.
Focardi S., Pelliccioni E.R., Petrucco R.,               Gill R.M.A., Thomas M.L. and Stocker D.
     Toso S. 2002b. Spatial patterns and                      1997. The use of portable thermal im-
     density dependence in the dynamics                       aging for estimating deer population
     of roe deer population in central Italy.                 density in forest habitats. J. Appl.
     Oecologia, 130: 411-419.                                 Ecol., 34: 1273-1286.
Gadgil M. and Bossert W. 1970. Life his-                 Guibert B. 1997. Une nouvelle approche
     tory consequences of natural selec-                      des populations de chevreuils en forêt;
     tion. American Naturalist, 104: 1-24.                    l’indice de pression sur la flore. ONF
Gaillard J.M., Boisaubert B., Boutin J.M.                     Bull. Techn., 32: 5-13.
                                                         Hewison A.J.M., Vincent J.P., Bideau E.,
     and Clobert J. 1986. L'estimation d'ef-
                                                              Angibault J.M. and Putman R.J. 1996.
     fectifs à partir de capture-marquage-
                                                              Variation in cohort mandible size as an
     recapture: application au chevreuil
                                                              index of roe deer (Capreolus capreo-
     (Capreolus capreolus). Gibier Faune                      lus) densities and population trends. J.
     Sauvage, 3:143-158.                                      Zool., 239: 573-581.
Gaillard J.M., Boutin J.M. and Van Laere                 Kurt F. 1991. Das Rehwild in der Kultur-
     G. 1993. Dénombrer les populations                       landschaft. Verlag Paul Parey. Hambo-
     de chevreuils par l'utilisation du Line                  urg and Berlin. 284 pp.
     transect. Etude de faisabilité. Rev.                Liberg O., Johansson A., Lockowandt S.
     Ecol. (Terre Vie), 48: 73-85.                             and Wahlström K. 1991. Density ef-
Gaillard J.M., Delorme D., Boutin J.M.,                        fects in roe deer demography. Tran-
     Van Laere G and Boisaubert B. 1996.                       sactions of the XXth IUGB Congress,
     Body mass of Roe Deer fawns during                        Gödöllö, Hungary, 1: 125-130.

                                                   118
Monitoring roe deer populations

Lincoln F.C. 1930. Calculating waterfowl              bility in northern Italy. Ecography, 19:
    abundance on the basis of banding re-             287-295.
    turns. U.S. Department of Agriculture,       Meriggi A., Brangi A., Schenone L. and
    Circular 118, 4 pp.                               Signorelli D. In press. La dieta del lu-
Lubow B.C., White G.C. and Anderson                   po (Canis Lupus) in Italia in relazione
    D.R. 1996. Evaluation of a linked sex             alle variazioni d’abbondanza degli un-
    harvest strategy for cervid popula-               gulati selvatici. Atti del Convegno
    tions. J. Wildl. Manage., 60: 787-796.            “Ricerca scientifica e strategie per la
Mabille A. and Neet C. 1994. Roe Deer                 conservazione del lupo in Italia” Bolo-
    density estimation by the line transect           gna 24 Novembre 2006.
    method. In: Wotschikowsky U (ed.),           Milner-Gulland E.J. and Akcakaya H.R.
    Proceedings of the 2nd European roe               2001. Sustainability indices for ex-
    deer Meeting, Brixen, 133-136.                    ploited populations. Trends Ecol. &
Maillard D., Boisaubert B. and Gaillard               Evol., 16:686-692.
    J.M. 1989. La masse corporelle: un           Mitchell B., Rowe J.J., Ratcliffe P.R. and
    bio-indicateur possible pour le suivi             Hinge M. 1985. Defecation frequency
    des populations de chevreuil. Gibier              in roe deer (Capreolus capreolus) in
    Faune Sauvage, 6: 57-68.                          relation to the accumulation rates of
Mattioli L., Apollonio M., Mazzarone V.               faecal deposits. J. Zool., 207: 1-7.
    and Centofanti E. 1995. Wolf food            Okarma H. 1995. The trophic ecology of
    habits and wild ungulate availability             wolves and their predatory role in un-
    in the Foreste Casentinesi National               gulate communities of forest ecosys-
    Park, Italy. Acta Theriol., 40: 387-              tems in Europe. Acta Theriol., 40: 335-
    402.                                              386.
Mattioli L., Capitani C., Avanzinelli E.,        Pedrotti L., Dupré E., Preatoni D. and Toso
    Bertelli I., Gazzola A. and Apollonio             S. 2001. Banca dati ungulati. Status,
    M. 2004. Predation by wolves (Canis               distribuzione, consistenza, gestione,
    lupus) on roe deer (Capreolus capreo-             prelievo venatorio e potenzialità delle
    lus) in north-eastern Apennine, Italy.            popolazioni di ungulati in Italia.
    J. Zool., 264: 249-258.                           Biologia e Conservazione della Fauna,
Mayle B.A., Putman R.J. and Wyllie I.                 109: 1-128.
                                                 Petersen C.G.J. 1896. The yearly immigra-
    2000. The use of trackway counts to
                                                      tion of young plaice into the Limfjord
    establish an index to deer presence.
                                                      from the German sea. Rep. Dan. Biol.
    Mammal Rev., 30: 233-237.
                                                      Station 1895, 6: 1-77.
McIntosh R., Burlton F.W.E. and
                                                 Pielowski Z. 1984. Some aspects of popu-
    McReddie G. 1995. Monitoring the
                                                      lation structure and longevity of field
    density of roe deer Capreolus capreo-             roe deer. Acta Theriol., 29: 17-33.
    lus population subjected to heavy            Pollock K.H., Nichols J.D., Brownie C. and
    hunting pressure. For. Ecol. Manage.,             Hines J.E. 1990. Statistical inference
    79: 99-106.                                       for capture–recapture experiments.
Meriggi A. and Lovari S. 1996. A review               Wildl. Monogr., 107: 1-97.
    of wolf predation in southern Europe:        Pollock K.H., Nichols J.D., Simons T.R. and
    does the wolf prefer wild prey to li-             Sauer J.R. 2000. The design of large
    vestock? J. Appl. Ecol., 33: 1561-                scale wildlife monitoring studies. Pro-
    1571.                                             ceedings Atti del V Convegno Interna-
Meriggi A., Brangi A., Matteucci C. and               zionale su Metodi Quantitativi per le
    Sacchi O. 1996. The feeding habits of             Scienze Applicate - Inferenza su Popo-
    wolves in relation to large prey availa-          lazioni Biologiche. 1-15.

                                           119
Meriggi et al.

Radeloff V.C., Pidgeon A.M. and Hostert                   Biol., 7: 1-205.
      P. 1999. Habitat and population mod-            Van Laere G., Maillard D., Boutin J.M. and
      elling of roe deer using an interactive             Delorme D. 1998. Le suivi des popula-
      geographic information system. Ecol.                tions de chevreuils des méthodes tradi-
      Model., 114: 287-304.                               tionnelles d’estimation aux indicateurs
Ratcliffe P.R. 1987. Red deer population                  biologiques. Proceedings XXI Col-
      changes and the independent assess-                 loque francophone de Mammalogie,
      ment of population size. Symp. Zool.                Amiens, Oct. 1997.
      Soc. Lond., 58: 153-165.                        Vincent J.P., Bideau E., Hewison A.J.M.
Ratcliffe P.R. and Mayle B.A. 1992. Roe                   and Angibault J.M. 1995. The influ-
      Deer Biology and Management. For.                   ence of increasing body weight, kid
      Comm. Bull., 105: 1-28.                             production, home range and winter
Seber G.A.F. 1982. The estimation of an-                  grouping in roe deer. J. Zool., 236:
      imal abundance and related parame-                  371-382.
      ters. Griffin, London.                          Vincent J.P., Gaillard J.M. and Bideau E.
Seber G.A.F. 1992. A review of estimating
                                                          1991. Kilometric index as biological
      animal abundance. IInd International
                                                          indicator for monitoring forest roe deer
      Statistical Review, 60: 129-166.
                                                          populations. Acta Theriol., 36: 315-
Staines B.W. and Ratcliffe P.R. 1987. Es-
                                                          328.
      timating the abundance of red deer
      (Cervus elaphus L.) and roe deer (Ca-           Von Berg F.C. 1979. Zahltreiben zum Er-
      preolus capreolus L.) and their cur-                fassen von Wilddbichten. Allgemeiner
      rent status in Great Britain. Symp.                 Forstzirtschrifte, 44/45: 1200-1201.
      Zool. Soc. Lond., 58: 131-152.                  Von Raesfeld F. 1985. Das Rehwild. Ver-
Strandgaard H. 1967. Reliability of the Pe-               lag Paul Parey, Hambourg and Berlin,
      tersen method tested on a roe-deer                  392 pp.
      population. J. Wildl. Manage., 4: 643-          Zejda J. 1984. Road strip transects for es-
      651.                                                timating field roe deer density. Folia
Strandgaard H. 1972. The roe deer (Ca-                    Zool., 33: 109-124.
     preolus capreolus) at Kälø and the fac-          Zejda J. 1985. Field transects for roe deer
     tors regulating its size. Dan. Rev. Game             census. Folia Zool., 34: 209-215.

                                                120
You can also read