Content uploaded by Aristotelis C. Papageorgiou
Author content
All content in this area was uploaded by Aristotelis C. Papageorgiou on Feb 28, 2014
Content may be subject to copyright.
Chapter 16 Conservation of Biodiversity in
Managed Forests: Developing an Adaptive
Decision Support System
Konstantinos Poirazidis, Stefan Schindler∗, Vassiliki Kati,
Aristotelis Martinis, Dionissios Kalivas, Dimitris Kasimiadis,
Thomas Wrbka and Aristotelis C. Papageorgiou
Abstract
Forest ecosystems provide several goods and services, but strategies for
the conservation of biodiversity are missing in traditional forest man-
agement schemes. In this paper we develope a decision support system
to optimize the conservation of biodiversity in managed forests, taking
Dadia National Park as a case study area, a local Mediterranean hotspot
of biodiversity in northeastern Greece. Using environmental niche fac-
tor analysis, we produced a series of spatially explicit habitat suitability
models for vascular plants, amphibians, small birds and raptors and an
overall model for total biodiversity. Further, we produced maps related
to timber production and investigated potential conflicts between conser-
vation of biodiversity and wood production. A decision support system
based on a conflict assessment was created using three management sce-
narios. It enables the establishment of integrated management strategies
and the assessment of their effects on biodiversity and timber production.
Habitat suitability models for selected groups of organisms were found
very effective to investigate the impact of the management on forests and
wildlife. Further evaluation of key indicator taxa on these models could
improve decision support systems and the sustainable management of
forests.
Keywords
Forest ecology, sustainable use, timber extraction, habitat suitability,
∗Stefan Schindler: Department of Conservation Biology, Vegetation & Landscape Ecology,
University of Vienna, Rennweg 14, A-1030. E-mail: stefan.schindler@univie.ac.at
C. Li et al. (eds.), Landscape Ecology in Forest Management and Conservation
© Higher Education Press, Beijing and Springer-Verlag Berlin Heidelberg 2011
16.1 Introduction 381
raptors, birds of prey, amphibians, vascular plants, Dadia National Park,
Greece.
16.1 Introduction
The increasing exploitation of forests is one of the main reasons of human-
induced loss of biodiversity (Lindenmayer et al. 2002; Foley et al. 2005). Al-
though the socio-economic value of biodiversity was underestimated until re-
cently (Costanza et al. 1997; Farber et al. 2002), its maintenance has become a
commonly accepted goal of sustainable forestry (United Nations 1992; Kohm
and Franklin 1997). The concept of ecosystem services provides a tool for com-
municating the importance of intact ecosystems for human well-being and a
framework for the evaluation of multiple functions of landscapes and forests
(Costanza et al. 1997; De Groot et al. 2002; Millennium Ecosystem Assess-
ment 2005; Boyd and Banzhaf 2007; Steffan-Dewenter et al. 2007). In forest
ecology, a major challenge is finding trade-offs between timber production and
conservation of biodiversity (Johns 1997; Putz et al. 2001; Foley et al. 2005;
Burke et al. 2008).
Forestry practices can enhance or reduce habitat for particular wildlife
species by altering structural features at the stand scale (Burke et al. 2008;
Rend´on-Carmona et al. 2009). Forest management that enhances the hetero-
geneity of forests has in general a positive impact on the local biodiversity
(Loehle et al. 2005; Gil-Tena et al. 2007; Torras et al. 2008; Kati et al. 2010;
Poirazidis et al. 2010a; Schindler et at. 2010), but forest management guide-
lines for the maintenance of biodiversity are mainly valid for site specific
conditions and can be rarely used as general directions (Loehle et al. 2005).
As it is impossible to measure and monitor the effects of various manage-
ment practices on the entire ecosystem, indicators are used as surrogates for
biodiversity (Lindenmayer et al. 2000). Taxon-based proxies include flagship,
umbrella and indicator species (Caro et al. 2004; Roberge and Angelstam
2004; Hess et al. 2006; Cabeza at el. 2008), while structure-based ones deal
mainly with stand complexity, connectivity and heterogeneity (Lindenmayer
et al. 2000; Schindler et al. 2008). Many researchers have explored the use
of particular taxa, especially vascular plants, arthropods and birds, as surro-
gates for biodiversity, but a general pattern has not yet emerged (Kati et al.
2004b; Sauberer et al. 2004; Sergio et al. 2005; Billeter et al. 2008; Cabeza
et al. 2008; Zografou et al. 2009). The importance of including several guilds
of taxa to represent adequately overall biodiversity is currently stressed by
several authors (Angelstam et al. 2004; Edenius and Milusinski 2006; Loehle
et al. 2006).
In this study, we developed a decision support system with the ultimate
goal of providing management guidelines and optimal solutions for the conser-
vation of biodiversity in managed forests. We considered Dadia National Park,
382 Chapter 16 Conservation of Biodiversity in Managed Forests
a Mediterranean forest mosaic in north-eastern Greece, as a case study. Using
available data sets from systematic scientific research in the area, a series of
habitat suitability models for groups of indicator species and for overall biodi-
versity was produced to discover potential conflicts between biodiversity and
timber production. Additionally, the effectiveness of different management
scenarios was assessed.
16.2 Methods
The following method section contains information about the study area, the
species data, and the applied statistical analyses. It further deals with the
methods of producing maps of habitat suitability, timber standing volume,
and forest management categories.
16.2.1 Study area
This research was conducted within Dadia National Park (hereafter called
Dadia NP), a sub-mountainous area with a diverse landscape mosaic, domi-
nated by extensive pine (Pinus brutia, P. nigra) and oak (Quercus frainetto,
Q. cerris, Q. pubescens) forest, but containing also a variety of other habi-
tats such as pastures, cultivated land, torrents and stony hills (Schindler et
al. 2008; Poirazidis et al. 2010a). Dadia NP covers 43,000 ha in the prefec-
ture of Evros, northeastern Greece (Fig. 16.1), and was designed to protect
the diverse community of birds of prey, including the last breeding colony
of the Eurasian black vulture (Aegypius monachus) in the Balkan peninsula
(Poirazidis et al. 2004, 2010b; Skartsi et al. 2008). Almost 45% of the National
Park is managed mainly for timber production (Zone B1), while it has been
recognized during the last years that this specific zone is of great value for
many species (Grill and Cleary 2003; Kati et al. 2004a, b, c, 2007; Korakis et
al. 2006; Poirazidis et al. 2010a,b).
16.2.2 Species data
We used five datasets of indicator species groups as surrogates for the total
biodiversity in Dadia NP, systematically surveyed using appropriate sampling
techniques per group. Those comprised woody plants, non-woody vascular
plants, amphibians, small birds and birds of prey (Kati and Sekercioglu 2006;
Korakis et al. 2006; Poirazidis et al. 2009; Kret, Poirazidis, Kati, unpublished
data). For each sampling plot (the number of plots was ranging from 34 to 63
depending on the indicator species group) all present species were evaluated.
16.2 Methods 383
Fig. 16.1 Location and zoning of Dadia National Park, the case study area in north-
eastern Greece. Zone B1 (highlighted in grey) represents the forest management area
that was investigated in this study. A1, A2: strictly protected areas, B2: agroforestry
area, B3: grazing land, A1/B1: forest management area that changed recently to
strictly protected area.
The survey for vascular plants was based on fieldwork during the years 1999
and 2000, and the 62 sampling plots had been chosen in accordance to the sur-
vey for the Nature 2000 Network (Korakis et al. 2006). The sampling scheme
for the amphibians was based on the breeding phenology of the species occur-
ring in eastern Greece (Arnold 1978; Helmer and Scholte 1985), and each pond
of the study area was visited once per month from February to July during the
year 2007. The presence of amphibians was detected through a combination of
visual encounter, aural and dip net surveys, during the diurnal transects in the
banks of the ponds (Kret, Poirazidis, Kati, unpublished data). We excluded
finally the species Triturus cristatus as its presence was verified at two sites,
only. Similarly, a subset of the existing database for small birds (Kati and
Sekercioglu 2006) was used for analysis. As the conservation value was one
of the factors under evaluation, we included in our analysis only bird species
that are “Species of European Conservation Concern” (SPEC; BirdLife In-
ternational 2004). These included species with an unfavorable conservation
status, concentrated in Europe (SPEC 2) or not (SPEC 3), as well as species
with favorable conservation status, but concentrated in Europe (SPEC 4). Fi-
nally, for the small birds, the two species Dendrocopos syriacus and D. medius
were used as a combined dataset due to limited detections of D. medius.The
survey of birds of prey was based on a systematic monitoring of raptor ter-
ritories that was conducted from 2001 through 2005 (Poirazidis et al. 2009,
384 Chapter 16 Conservation of Biodiversity in Managed Forests
2010b), and we pooled the data of all five years and plotted the centers of the
yearly territories. The Black stork (Ciconia nigra), a species of conservation
priority in the area (Tsachalidis and Poirazidis 2006), was included in the
raptor dataset. A subset of the breeding raptor species was used in this study,
and the criterion for selection was the relatively high abundance in order to
produce stable habitat suitability models.
16.2.3 Habitat suitability maps and statistical analysis
Habitat suitability maps (HSM) have broad applicability within conserva-
tion biology and are of special interest to predict the distributions of wildlife
species for geographical areas that have not been extensively surveyed. The
methods for modeling habitat suitability can be classified into two groups:
those requiring presence-only data and those requiring presence-absence data
(Guisan and Zimmerman 2000). Here we prepare HSM using Ecological Niche
Factor Analysis (ENFA) provided by the software BIOMAPPER (Hirzel et
al. 2002). ENFA is a multivariate approach developed to predict habitat suit-
ability based on the likelihood of occurrence of the species when absence data
for the species are not available (Hirzel et al. 2002). Without absence data
some limitations on the accuracy of the habitat suitability maps are possible
(Hirzel and Le Lay 2008), and we reclassified the predictions into four robust
levels (=bins) of suitability to settle this problem (Hirzel et al. 2006). The
suitability is based on functions that define the marginality of the species,
i.e. how the species mean differs from the mean of the entire area, and the
specialization of the species, i.e. the ratio of the overall variance to the species
variance. Marginality lies between 0 and 1, with larger values indicating that
the focal species has habitat requirements that differ from the average avail-
able conditions. A high specialization value indicates that the focal species
has a particular requirement for certain habitat characteristics and occupies
a narrow range of variables compared to the overall range of variables within
the study area (Hirzel et al. 2002).
We used 23 environmental variables, classified into four groups to derive
potentially relevant predictors for species habitat selection (Table 16.1). This
database contained maps stored in both a vectorial and a raster format. All
species and habitat information was rasterized into a 50 ×50 m grid cell maps.
Topographical data were directly obtained as quantitative variables. Variables
quantifying land cover, landscape and potential sources of disturbance were
transformed into frequency and distance variables. The forest cover categories
were reclassified into pure broadleaves, mixed pine-oak and pure pine forest,
but only the first two were used for the models, as the information from the
third was redundant. As ENFA does not work with multinomial data, these
qualitative maps were converted into several Boolean maps (i.e. one for each
variable). Frequency describes the proportion of cells from a given category
16.2 Methods 385
within a circle around the focal cell and it was derived using a circular moving
window. We varied the radius of the moving window to test the performance
of three different scales (200 m, 500 m and 1,000 m), but finally only the scale
of 1,000 m was used as it performed better than the others. The topographical
descriptors were averaged by means of a similar radius circular moving win-
dow. Spatial data analysis was conducted using ArcMap 9.0 and the Spatial
Analyst extension.
Correlations between all variables of the initial pool of predictors (Table
16.1) were calculated prior to the ENFA. When two or more predictors had
a correlation coefficient greater than 0.7, only the most proximal was kept
(Austin 2002). Topographic and frequency environmental layers were nor-
malized using the “box–cox” algorithm (Sokal and Rohlf 1981) and distance
variables by the “square root” algorithm. There are different algorithms avail-
able in BIOMAPPER to build habitat suitability maps by ENFA (Hirzel et
al. 2002) and following Hirzel and Arlettaz (2003) we used the geometric mean
Table 16.1 Environmental variables used in ENFA as predictors to define the
species’ ecological niche.
Environmental predictors Scales (m)
Topography -
1. Altitude 200, 500, 1000
2. 1 SD of altitude 200, 500, 1000
3. Slope 200, 500, 1000
4. Northness aspect 200, 500, 1000
Landscape/F orest attributes -
5. Relative richness index 200, 500, 1000
6. Fragmentation index 200, 500, 1000
7. Frequency of broadleaves 200, 500, 1000
8. Frequency of mixed forest (Pine-Oak) 200, 500, 1000
Other ecological metrics -
9. Frequency of openings 200, 500, 1000
10. Frequency of agricultural lands 200, 500, 1000
11. Frequency of permanent water 200, 500, 1000
12. Frequency of rocky area 200, 500, 1000
13. Distance to openings -
14. Distance to agricultural lands -
15. Distance to main river -
16. Distance to permanent water -
17. Distance to rocky area -
P otential disturbance metrics -
18. Frequency of paved roads 200, 500, 1000
19. Frequency of unpaved roads 200, 500, 1000
20. Frequency of urban area 200, 500, 1000
21. Distance to paved roads -
22. Distance to unpaved roads -
23. Distance to urban area -
386 Chapter 16 Conservation of Biodiversity in Managed Forests
algorithm to account for the density of the observations in environmental
space.
For the plants, the number of species was used as dependent variable per
plot and we created two multiple regression models (one for woody plants and
one for non-woody vascular plants) to predict species richness. The resulting
models were transformed with the “box-Cox byte” algorithm and combined
with equal weight (factor 0.5) to produce the overall “plant HSM”. For each of
the three groups of fauna, an overall HSM was created combining the specific
HSMs by user-defined weight per species (Eastman 2001), which depended
on the conservation value (Appendix). Finally, all HSMs per organism group
were combined into an overall biodiversity HSM applying a new user-defined
weight per group. The HSM for breeding Black vulture and Egyptian vulture
(Neophron percnopterus) – the species with the highest conservation value in
the area – were not included in the initial raptor HSM, but were used as
Boolean data in a later step (see below) to highlight the priority areas for
conservation of these two species.
16.2.4 Timber standing volume
We used the recent forest inventory for wood production of the local Forest
Service (2006-2016) to produce quantitative maps of the distribution of stand-
ing wood volumes (basal area) (Consorzio Forestale del Ticino 2006). We used
the stand level as spatial unit to summarize these data (417 sub-units of the
division of managed forest, with an average size of 46.5 ±18.9 ha). The tim-
ber volume was described as pine, oak and total volume (Consorzio Forestale
del Ticino 2006). We used only the managed area of Dadia NP (zone B1),
excluding the non-managed strictly protected areas (Fig. 16.1).
16.2.5 Establishment of the management scenarios
To obtain spatially explicit management plans at stand level, we reclassified
the biodiversity thematic maps into four bins representing habitat suitabil-
ity: (1) unsuitable, (2) marginal, (3) suitable and (4) optimal. We also re-
classified the timber maps into four bins representing the standing volume:
(1) minimum, (2) medium, (3) large and (4) maximum. We used the Nat-
ural Break method (ArcMap) for the biodiversity bin classification, and the
four timber volume bins were defined by values of total standing timber vol-
ume of <500 m3, 500-1,000 m3, 1,000-2,000 m3and >2,000 m3per stand.
We finally considered four possible general management actions at the stand
level, in order to integrate biodiversity values into the timber management:
(1) management without limitations (free forestry), (2) management with tem-
16.3 Results 387
poral restrictions, (3) management with temporal and spatial restrictions, and
(4) management focussing on the ecological values (ecological management).
In this study, we implemented three management scenarios. The “biodi-
versity scenario” focused on the maximization of the biodiversity value (max-
imum environmental profit) in the managed forest. It was defined by the bio-
diversity models with each bin of habitat suitability leading to related man-
agement actions (Table 16.2), e.g. biodiversity bin 1 “unsuitable” leading to
management action 1 “free forestry” and biodiversity bin 4 “optimal” to man-
agement action 4 “ecological management”.The“timber scenario”focusedon
the maximization of the economical benefits for the timber production (max-
imum economical profit) and was defined by the standing volume map with
each bin of timber density leading to inverse related management actions (Ta-
ble 16.2), e.g. timber volume bin 1 “minimal” leading to management action
4“ecological management” or timber volume bin 4 “maximum” to manage-
ment action 1 “management without limitations”. The third scenario was the
“trade off scenario”, which attempted to maximize the long-term net benefits
for both biodiversity and society. The established trade off matrix considered
both biodiversity and timber production at the same level, leading to the final
determination of the management action for each stand (Table 16.2).
Table 16.2 Forest management categories determined by biodiversity and timber
production under the scenarios biodiversity,timber and trade off.
Scenario Biodiversity Timber Trade Off
Timberbins 123412341234
Biodiversity bins 8
>
<
>
:
1FF FF FF FF EM TSR TR FF FF FF FF FF
2TR TR TR TR EM TSR TR FF TR TR FF FF
3TSRTSRTSRTSREMTSRTRFFTSRTSRTR TR
4EM EM EM EM EM TSR TR FF EM EM TSR TSR
FF: free forestry, TR: temp oral restrictions, TSR: temporal and spatial restrictions, EM: ecolog-
ical management. Biodiversity bins: 1 unsuitable, 2 marginal, 3 suitable, 4 optimal; timber bins:
1 minimal, 2 medium, 3 large, 4 maximal.
We applied each scenario to each biodiversity data set as well as to the
overall biodiversity HSM. For each scenario at the last step, we used the suit-
able and optimal areas for Eurasian black vulture and Egyptian vulture as
Boolean variables as such: suitable and optimal areas for Eurasian black vul-
ture were upgraded to the Management action “4” (ecological management)
and for Egyptian vulture to the Management action “3” (temporal and spatial
restrictions).
16.3 Results
In the following section, we present the resulting maps regarding habitat suit-
ability, timber standing volume, and forest management categories. We fur-
ther present the evaluation of the effectiveness of the different management
388 Chapter 16 Conservation of Biodiversity in Managed Forests
scenarios in conserving biodiversity.
16.3.1 Habitat suitability maps
The species richness of vascular plants (351 plant species in 63 plots) was
modeled using the eco-geographical variables as independent variables. The
resulting regression model for woody plants was “Y = 4.3 + 2.01 northness
– 10.29 frequency of openings + 2.53 frequency of mixed forest + 0.001 fre-
quency of rocks + 0.001 distance to agricultural lands”, while for non-woody
plants it was “Y = 30.4 + 0.24 slope – 0.23 relative richness index + 5.02
frequency of mixed forest”. Both models were significant at the level p=0.05
and were combined equally to the overall HSM for plants (Fig. 16.2a)
Fig. 16.2 Habitat suitability maps for (a) plants, (b) amphibians, (c) small birds,
(d) raptors and (e) overall biodiversity in Dadia NP.
Amphibians (10 species in 53 plots) showed a pronounced specialization for
certain habitats as their mean global marginality was 0.94 (range 0.63-1.35)
and their specialization was 4.37 (range 1.59-12.56). Both groups, small birds
16.3 Results 389
and raptors, showed intermediate sensibility and differentiation of habitat
use. The mean global marginality of small birds was 0.70 (range 0.35-1.05)
and the specialization was 3.23 (range 1.13-6.93). For the raptor HSM, ten
species of breeding raptors plus the Black stork had a relative abundance
that enabled stable models. The mean global marginality for raptors was 0.63
(range 0.17-1.64) and the specialization was 2.05 (range 1.03-6.05). Finally,
separate HSM were created for each taxon-group of animals (Fig. 16.2b,c,d)
using species specific weights (Appendix). The combined overall biodiversity
HSM resulted (Fig. 16.2e), applying the weights of 0.5 to raptors HSM, 0.25
to amphibians HSM, 0.15 to small birds HSM, and 0.1 to plants HSM.
16.3.2 Standing volume distribution maps
The mean pine wood volume was 1,533.2 m3±1,424.1 (sd) per stand, with
a maximum value of 7,380.8 m3while the mean oak wood volume was 731.5
±658.1 m3with a maximum value of 4,785.3 m3. The total timber volume
ranged from 69 to 8,094 m3(Fig. 16.3), while the total volume per ha was
49.2 m3±26.2 and ranged per forest stand from 2 m3/ha to 131 m3/ha.
Fig. 16.3 Total timber standing volume of the managed forest area in Dadia NP.
390 Chapter 16 Conservation of Biodiversity in Managed Forests
16.3.3 Establishment of the management scenarios
We produced three thematic maps of spatially explicit management plans,
based on the desired forestry policy in the management area (Fig. 16.4). At
the timber scenario, where conservation priorities are considered exclusively in
areas without economical value for timber, only 6% of the area was proposed
for ecological management and 46% for free forestry. On the other hand, in
the biodiversity scenario, where the most suitable areas remain unexploited,
18% of the managed forests were proposed for ecological management and
Fig. 16.4 Spatial forest management plans, presenting the distribution of the four
forest management categories under the timber, trade off and biodiversity scenario.
16.4 Discussion 391
11% for free forestry. The trade off scenario, taking into account both tim-
ber and biodiversity, lies in between, proposing 9% of the area for ecological
management and 32% for free forestry.
The trade off scenario served both ecosystem services, biodiversity values
and timber production (Fig. 16.5). In this scenario, 91% of the area with low
suitability for biodiversity (bins unsuitable and marginal)wascoveredbythe
management category “free forestry”, while the areas of high suitability for
biodiversity (bins suitable and optimal) were intensively covered by the man-
agement categories “temporal and spatial restrictions” (47%) and “ecological
management” (25%). For comparison, in the timber scenario, only 60% of the
low biodiversity area was dedicated to free forestry and more importantly only
42% and 4% of the high biodiversity areas were classified as “temporal and
spatial restrictions” and “ecological management”, respectively (Fig. 16.5).
Fig. 16.5 Management and conservation of areas of differing suitability of biodi-
versity under the scenarios “Biodiversity”, “Trade off”, and “Timber”. Black bars:
forest stands of high suitability for biodiversity (bins suitable and optimal), white
bars: forest stands of low suitability for biodiversity (bins unsuitable and marginal);
FF: free forestry, TR: temporal restrictions, TSR: temporal and spatial restrictions,
EM: ecological management.
16.4 Discussion
In the following section we discuss the need of integrating biodiversity into for-
est management, and several aspects regarding multi-taxa indicators, decision
support systems, and the scenarios applied in this study.
16.4.1 Integrating biodiversity into forest management
New environmental policies call for increased attention to biodiversity issues in
forest management planning, given that the loss and fragmentation of mature
forest together with the structural diversity decline have threatened forest-
392 Chapter 16 Conservation of Biodiversity in Managed Forests
dependent species (Andr´en 1994; Siitonen 2001; Thompson et al. 2003; An-
gelstam et al. 2004; Poirazidis et al. 2004). Sustainable forestry and deadwood
supply have recently emerged as two of the twenty-six headline indicators to-
wards halting further biodiversity loss in Europe (European Environmental
Agency 2007). In this frame, the approach developed in this study provides a
useful tool for forest managers. We established biodiversity priority areas into
the managed areas, providing a guideline for effective management strategies.
We also developed habitat suitability models based on environmental features
and we identified habitat associations that provide an important source of in-
formation for general habitat management issues. These models quantifying
relationships between species and their habitats are considered nowadays one
of the most efficient tools for forest management (Edenuis and Mikusinsky
2006). Sustainable forest management should be efficient, satisfying on one
hand conservation goals while minimizing on the other hand socio-economic
costs and the area removed from timber production (Pressey et al. 1997; Mon-
tigny and McLean 2005).
16.4.2 Species selection and multi-taxa indicator species
We modeled in this research habitat suitability for several groups of organ-
isms, using totally 351 taxa of vascular plants, 10 species of amphibians and
23 species of birds for the assessment. For a successful use of habitat suit-
ability models in forest biodiversity management an appropriate selection of
species is required and multi-taxa bio-indication has several advantages (King
et al. 1998; Angelstam et al. 2004; Rempel et al. 2004; Wrbka et al. 2008).
Ecologically different taxa can show different patterns of biodiversity and it is
assumed that even several species of one single taxon or guild are not enough
for being representative (Schulze et al. 2004; Billeter et al. 2008; Cabeza et
al. 2008). Also Edenius and Mikuszinski (2006) stressed the need for multi-
species selection procedures in their recent review on the use of HSM in forest
management. They have found only one study (out of 55 reviewed ones) that
followed a multi-taxa approach, and only five papers of the review (9%) could
be attributed to indicator species in the species selection procedure.
The indicator species approach has been criticized on conceptual grounds,
such that no species share the same ecological niche, as well as on empirical
grounds, i.e. untested or unverified relationships between the indicator and the
species or species groups that the indicator supposedly covers (Lindenmayer
et al. 2000; Rempel et al. 2004; Roberge and Angelstam 2004; Edenius and
Mikuszinski 2006). In our study we used vascular plants, amphibians, small
birds and raptors as indicator groups in habitat suitability models. Recent
research confirmed that plants and birds are well performing surrogate taxa
for overall biodiversity in Dadia NP (Kati et al. 2004b; see also Sauberer et al.
2004 for a Central European case study). Amphibians, due to their very spe-
16.4 Discussion 393
cific habitat needs and life cycle, are important for being complementary and
good indicators of habitat matrix permeability (Ray et al. 2002; Kati et al.
2004a, 2007; Cabeza et al. 2008). Raptors are top predators; requiring enough
prey, large areas and limited disturbance, they indicate ecosystem health and
perform well as indicators of biodiversity (Sergio et al. 2005; Sekercioglu 2006;
but see also Cabeza et al. 2008). Raptors are also focal species of conserva-
tion efforts in the reserve, as their populations in Dadia NP are of regional
importance (Poirazidis et al. 2004, 2007, 2010b; Skartsi et al. 2008).
16.4.3 Decision Support Systems and comparison of scenarios
Concerning limited funding and limited data sources, adaptive management is
a useful tool for fast implementations (Angelstam et al. 2004; Duff et al. 2009).
Ideally, an active adaptive management approach with iterated assessment
and corrective action should be applied through continuous mutual learning
by scientists, policymakers, managers and other actors until the targets are
reached (Simberloff 1999; Brown et al. 2001; Angelstam et al. 2004; Steffan-
Dewenter et al. 2007; Duff et al. 2009). The three scenarios, presented in this
case study, are adaptive in terms of their main objectives and regarding their
simplicity. The timber scenario is a simple approach to integrate conservation
of biodiversity into forest management when timber production has the main
priority. In this scenario more restrictive conservation management will be
done only in forest stands with little timber. The biodiversity scenario can be
followed when conservation is the key issue. Restrictions are proposed, where
habitat suitability reaches maximum values, the performance regarding con-
servation is optimal, but the socio-economic benefits remain totally unused
in forest stands with a high level of biodiversity. The trade off scenario as
an alternative solution proved very useful to integrate timber extraction and
nature conservation and an optimization of the benefits for society and biodi-
versity could be achieved. Compared with the timber scenario, free forestry is
encouraged where habitat suitability is lower but forest stands of high biodi-
versity have more restrictions. A decision support system can be an effective
mechanism to support technological and managerial decision making (Mal-
czewski 2006) as it can combine multiple sources of information (models and
data) into a single system that provides a tool to manipulate the information.
With these capabilities, it supports decision makers in cognitive tasks that
involve choices, judgment and decisions, in recognizing needs and identifying
objectives, as well as in formulating and evaluating different courses of action
(Garcia and Armbruster 1997). In the case of sustainable forest management,
these actions are forest management scenarios, i.e. collections of rules and
strategies regarding harvest scheduling and forest regeneration (Van Damme
et al. 2003).
Timber harvesting and conservation of biodiversity are not necessarily
394 Chapter 16 Conservation of Biodiversity in Managed Forests
mutually exclusive and some rules of temporal and spatial restrictions can
optimize their coexistence (L˜ohmus 2005; Brown et al. 2007). Integrating
different data sources to a decision support system for spatial forest manage-
ment planning can increase clearly the sustainability of forest management.
Viable populations of indicators species and a high level of biodiversity can be
maintained, without losing the socio-economic benefits of professional timber
production. At the local scale, a selective targeting approach that identifies
forest stands of potential high biodiversity and nature conservation value is
essential. Once identified, these areas can be highlighted for inclusion in fu-
ture local targets and management prescriptions altered accordingly (Bayliss
et al. 2005). As maps of habitat suitability were initially created for individual
species, our approach provides also a further resource for species specific con-
servation management. We recommend applying habitat suitability modeling
to selected groups of indicator organisms to develop spatial management plans
for managed forests. This enhances the sustainability of the management and
promotes monitoring and evaluation of its effects on wildlife. The inclusion
of further taxa as indicators of overall biodiversity into the existing decision
support system is a prerequisite for continuous improvements of a sustainable
forest management.
Acknowledgements
This research was financed by the Greek project “EPEAEKII-PYTHAGORAS
II: KE 1329-1” and co-funded by the European Social Fund & National Re-
sources. We thank Giorgos Korakis and Elzbieta Kret for providing the data
sets for plants and amphibians, respectively, Giancarlo Graci for computing a
specific GIS extension, and Christa Renetzeder for her helpful comments on
the manuscript.
References
Andr´en H (1994) Effects of habitat fragmentation on birds and mammals in land-
scapes with different proportions of suitable habitat: A review. Oikos 71: 355-
366.
Angelstam P, Boutin S, Schmiegelow F et al. (2004) Targets for boreal forest bio-
diversity conservation – a rationale for macroecological research and adaptive
management. Ecol Bull 51: 487-509.
Arnold EN, Burton JA (1978) A Field Guide to the Reptiles and Amphibians of
Britain and Europe. Collins, London.
Austin MP (2002) Spatial prediction of species distribution: An interface between
ecological theory and statistical modelling. Ecol Model 157: 101-118.
Bayliss JL, Simonite V, Thompson S (2005) The use of probabilistic habitat suitabil-
ity models for biodiversity action planning. Agric Ecosyst Environ 108: 228-250.
References 395
Billeter R, Liira J, Bailey D et al. (2008) Indicators for biodiversity in agricultural
landscapes: A pan-European study. J Appl Ecol 45: 141-150.
BirdLife International (2004) Birds in Europe. Population estimates, trends and
conservation status. BirdLife Conservation Series No. 12, Cambridge.
Boyd J, Banzhaf S (2007) What are ecosystem services? The need for standardized
environmental accounting units. Ecol Econ 63(2-3): 616-626.
Brown GS, Rettie WJ, Brooks RJ et al. (2007). Predicting the impacts of forest
management on woodland caribou habitat suitability in black spruce boreal
forest. For Ecol Manag 245: 137-147.
Brown NR, Noss RF, Diamond DD et al (2001) Conservation biology and forest
certification: Working together towards sustainable forestry. J. For 99(8): 18-
25.
Burke DA, Elliott KA, Holmes SB et al. (2008) The effects of partial harvest on
the understory vegetation of southern Ontario woodlands. For Ecol Manag 255:
2204-2212.
Cabeza M, Arponen A, van Teeffelen A (2008) Top predators: Hot or not? A call
for systematic assessment of biodiversity surrogates. J Appl Ecol 45: 976-980.
Caro T, Engilis Jr A, Fitzherbert E et al. (2004) Preliminary assessment of the
flagship species concept at a small scale. Anim Conserv 7: 63-70.
Consorzio Forestale del Ticino (2006) Management study of the Dadia-Lefkimi-
Souflion public forest, Soufli forest service, Evros prefecture. Ministry of Agri-
culture, General Direction of Forests and Natural Environment.
Costanza R, d’Arge R, de Groot RS et al. (1997) The value of the world’s ecosystem
services and natural capital. Nature 387: 253-260.
De Groot RS, Wilson MA, Boumans RMJ (2002) A typology for the classification,
description and valuation of ecosystem functions, goods and services. Ecol Econ
41(3): 393-408.
Duff G, Garnett D, Jacklyn P et al. (2009) A collaborative design to adaptively
manage for landscape sustainability in north Australia: Lessons from a decade
of cooperative research. Landsc Ecol 24: 1135-1143.
Eastman JR (2001) Idrisi32, Guide to GIS and Image Processing, Volume 1 &
Volume 2. Clark University, Worcester, USA.
Edenius L, Mikusinski G (2006) Utility of habitat suitability models as biodiversity
assessment tools in forest management. Scand J For Res 21(Suppl 7): 62-72.
European Environmental Agency (2007) Halting the loss of biodiversity by 2010:
Proposal for a first set of indicators to monitor progress in Europe. EEA Tech-
nical report No 11/2007. EEA, Copenhagen.
Farber S, Costanza R, Wilson M (2002) Economic and ecological concepts for valu-
ing ecosystem services. Ecol Econ 41: 375-392.
Foley JA, DeFries R, Asner GP et al. (2005) Global consequences of land use.
Science 309: 570-574.
Garcia LA, Armbruster M (1997) A decision support system for evaluation of wildlife
habitat. Ecol Model 102: 287-300.
Gil-Tena A, Saura S, Brotons L (2007) Effects of forest composition and structure
on bird species richness in a Mediterranean context: Implications for forest
ecosystem management. For Ecol Manag 242: 470-476.
Grill A, Cleary D (2003) Diversity patterns in butterfly communities of the Greek
nature reserve Dadia. Biol Conserv 114: 427-436.
Guisan A, Zimmermann NE (2000) Predictive habitat distribution models in ecol-
ogy. Ecol Model 135(2-3): 147-186.
Helmer W, Scholte P (1985) Herpetological research in Evros, Greece. Proposal
for a biogenetic reserve, Report by Research Institute for Nature Management
Arnhem and Department of Animal Ecology, Catholic University Toernooiveld,
396 Chapter 16 Conservation of Biodiversity in Managed Forests
Nijmegen.
Hess GR, Bartel RA, Leidner AK et al. (2006) Effectiveness of biodiversity indicators
varies with extent, grain, and region. Biol Conserv 132: 448-457.
Hirzel AH, Arlettaz R (2003) Modelling habitat suitability for complex species dis-
tributions by the environmental-distance geometric mean. Environ Manag 32:
614-623.
Hirzel AH, Le Lay G (2008) Habitat suitability modelling and niche theory. J Appl
Ecol 45: 1372-1381.
Hirzel AH, Hausser J, Chessel D et al. (2002) Ecological-Niche Factor Analysis:
How to compute habitat-suitability maps without absence data? Ecology 83:
2027-2036.
Hirzel AH, Le Lay G, Helfer V et al. (2006) Evaluating the ability of habitat suit-
ability models to predict species presences. Ecol Model 199: 142-152.
Johns AG (1997) Timber Production and Biodiversity Conservation in Tropical
Rain Forests. Cambridge Studies in Applied Ecology and Resource Management
(RU). Cambridge University Press, Cambridge.
Kati V, Sekercioglu CH (2006) Diversity, ecological structure, and conservation of
the landbird community of Dadia reserve, Greece. Divers Distrib 12: 620-629.
Kati V, Devillers P, Dufrˆene M et al. (2004a) Hotspots, complementarity or repre-
sentativeness? Designing optimal small-scale reserves for biodiversity conserva-
tion. Biol Conserv 120: 471-480.
Kati V, Devillers P, Dufrˆene M et al. (2004b) Testing the value of six taxonomic
groups as biodiversity indicators at a local scale. Conserv Biol 18: 667-675.
Kati V, Dufrˆene M, Legakis A et al. (2004c) Conservation management for the
Orthoptera in the Dadia reserve, Greece. Biol Conserv 115: 33-44.
Kati V, Foufopoulos J, Ioannidis Y et al. (2007) Diversity, ecological structure and
conservation of herpetofauna in a Mediterranean area (Dadia National Park,
Greece). Amphib-Reptil 28: 517-529.
Kati V, Poirazidis K, Dufrˆene M et al. (2010) Toward the use of ecological het-
erogeneity to design reserve networks: A case study from Dadia National Park,
Greece. Biodiv Conserv 19: 1585-1597.
King JR, Andersen AN Cutter AD (1998) Ants as bioindicators of habitat distur-
bance: Validation of the functional group model for Australia’s humid tropics.
Biodiv Conserv 7: 627-638.
Kohm K, Franklin JF (1997) Forestry in the 21st Century. Island Press, Covelo,
California.
Korakis G, Gerasimidis A, Poirazidis K et al. (2006) Floristic records from Dadia-
Lefkimi-Soufli National Park, NE Greece. Flora Mediterr 16: 11-32.
Lindenmayer DB, Margules CR, Botkin DB (2000) Indicators of biodiversity for
ecologically sustainable forest management. Conserv Biol 14: 941-950.
Lindenmayer DB, Cunningham RB, Donnelly CF et al. (2002) On the use of land-
scape surrogates as ecological indicators in fragmented forests. For Ecol Manag
159: 203-216.
Loehle C, Wigley TB, Rutzmoser S et al. (2005) Managed forest landscape structure
and avian species richness in the Southeastern US. For Ecol Manag 214: 279-
293.
Loehle C, Van Deusen P, Wigley TB et al. (2006) A method for landscape analysis of
forestry guidelines using bird habitat models and the Habplan harvest scheduler.
For Ecol Manag 232: 56-67.
L˜ohmus A (2005) Are timber harvesting and conservation of forest-dwelling raptors
always mutually exclusive? Anim Conserv 8: 443-450.
Malczewski J (2006) GIS-based multicriteria decision analysis: A survey of the
literature. Int J Geogr Inf Sci 20(7): 703-726.
References 397
Millennium Ecosystem Assessment (2005) Ecosystems and Human Well-being: Syn-
thesis, World Resources Institute, Island Press, Washington DC.
Montigny MK, MacLean DA (2005) Using heterogeneity and representation of
ecosite criteria to select forest reserves in an intensively managed industrial
forest. Biol Conserv 125: 237-248.
Poirazidis K, Goutner V, Skartsi T et al. (2004) Modelling nesting habitat as a
conservation tool for the Eurasian black vulture (Aegypius monachus)inDadia
Nature Reserve, northeastern Greece. Biol Conserv 118: 235-248.
Poirazidis K, Goutner V, Tsachalidis E et al. (2007) Comparison of nest-site selec-
tion patterns of different sympatric raptor species as a tool for their conserva-
tion. Anim Biodiv Conserv 30: 131-145.
Poirazidis K, Schindler S, Ruiz C et al. (2009) Monitoring breeding raptor popula-
tions – a proposed methodology using repeatable methods and GIS. Avocetta
33 (in press).
Poirazidis K, Kati V, Schindler S et al. (2010a) Habitat heterogeneity and biodiver-
sity. In: Catsadorakis G (ed) The Dadia-Lefkimi-Soufli National Park, Greece:
Biodiversity, Management and Conservation. WWF Greece, Athens: 103-114.
Poirazidis K, Schindler S, Kakalis E et al. (2010b) Diurnal birds of prey in Dadia-
Lefkimi-Soufli National Park: Long-term Population Trends and Habitat. In:
Catsadorakis G (ed) The Dadia-Lefkimi-Soufli National Park, Greece: Biodi-
versity, Management and Conservation. WWF Greece, Athens: 151-168.
Pressey RL, Possingham HP, Day JR (1997) Effectiveness of alternative heuristic
algorithms for identifying minimum requirements for conservation reserves. Biol
Conserv 80: 207-219.
Putz FE, Blate GM, Redford KH et al. (2001) Tropical forest management and
conservation of biodiversity: An overview. Conserv Biol 15: 7-20.
Ray N, Lehmann A, Joly P (2002). Modeling spatial distribution of amphibian
populations: A GIS approach based on habitat matrix permeability. Biodiv
Conserv 11: 2143-2165.
Rempel RS, Andison DW, Hannon SJ (2004). Guiding principles for developing an
indicator monitoring framework. For Chron 80: 82-90.
Rend´on-Carmona H, Mart´ıez-Yr´ızar A, Balvanera P et al. (2009) Selective cutting
of woody species in a Mexican tropical dry forest: Incompatibility between use
and conservation. For Ecol Manag 257: 567-579.
Roberge J-M, Angelstam P (2004) Usefulness of the umbrella species concept as a
conservation tool. Conserv Biol 18(1): 76-85.
Sauberer N, Zulka KP, Abensperg-Traun M et al. (2004) Surrogate taxa for biodi-
versity in agricultural landscapes of eastern Austria. Biol Conserv 117: 181-190.
Schindler S, Poirazidis K, Wrbka T (2008) Towards a core set of landscape metrics
for biodiversity assessments: A case study from Dadia National Park, Greece.
Ecol Indic 8: 502-514.
Schindler S, Poirazidis K, Papageorgiou AC, et al. (2010) Landscape approaches
and GIS as a prerequisite for biodiversity management in a Mediterranean forest
landscape. In: Andel J, Bicik I, Dostal P et al. (eds) Landscape Modelling: Ge-
ographical Space, Transformation and Future Scenarios, Urban and Landscape
Perspectives Series Vol 8, Springer-Verlag, 207-220.
Schulze CH, Waltert M, Kessler PJA et al. (2004) Biodiversity indicator groups
of tropical land-use systems: Comparing plants, birds, and insects. Ecol Appl
14(5): 1321-1333.
Sekercioglu CH (2006). Increasing awareness of avian ecological function. Trends
Ecol Evol 21: 464-471.
Sergio F, Newton I, Marchesi L (2005) Top predators and biodiversity. Nature 436:
92.
398 Chapter 16 Conservation of Biodiversity in Managed Forests
Siitonen I (2001) Forest management, coarse woody debris and saproxylic organ-
isms: Fennoscandian boreal forests as an example. Ecol Bull 49: 11-42.
Simberloff D (1999) The role of science in the preservation of forest biodiversity.
For Ecol Manag 115: 101-111.
Skartsi Th, Elorriaga J, Vasilakis D et al. (2008) Population size, breeding rates
and conservation status of Eurasian black vulture in the Dadia National Park,
Thrace, NE Greece. J Nat Hist 42: 345-353.
Sokal RR, Rohlf FJ (1981) Biometry, 2nd ed. Freeman, San Francisco.
Steffan-Dewenter I, Kessler M, Barkmann J et al. (2007) Tradeoffs between income,
biodiversity, and ecosystem functioning during tropical rainforest conversion
and agroforestry intensification. Proc Nat Acad Sci 104: 4973-4978.
Thompson ID, Baker JA, Ter-Mikaelian M (2003) A review of the long-term effects
of post-harvesting on vertebrate wildlife, and predictive models, with an em-
phasis on boreal forests in Ontario, Canada. Forest Ecology and Management
177: 441-469.
Torras O, Gil-Tena A, Saura S (2008) How does forest landscape structure explain
tree species richness in a Mediterranean context? Biodiv Conserv 17: 1227-1240.
Tsachalidis EP, Poirazidis K (2006) Nesting habitat selection of the Black Stork
Ciconia nigra in Dadia National Park, Northeastern Greece. In: Manolas E
(ed) Proceedings, 1st International Congress on Sustainable Management and
Development of Mountainous and Island areas. Naxos Island, Greece, September
2006. Volume 2. University of Crete, Heraklion, Greece, 147-153.
United Nations (1992) Agenda 21: the United Nations programme of action from
Rio. United Nations, New York.
Van Damme L, Russel JS, Doyon F et al. (2003) The development and application
of a decision support system for sustainable forest management on the Boreal
Plain. J Environ Eng Sci 2 (Suppl. 1): 23-34.
Wrbka T, Schindler S, Pollheimer M et al. (2008) Impact of the Austrian Agri-
Environmental Scheme on diversity of landscape, plants and birds. Community
Ecol 9: 217-227.
Zografou K, Sfenthourakis S, Pullin A et al. (2009) On the surrogate value of red-
listed butterflies for butterflies and grasshoppers: A case study in Grammos site
of Natura 2000, Greece. J Insect Conserv 13:505-514.
Appendix
Selected species used for the habitat suitability models for amphibians, small
birds and raptors, and user-defined weights (adding up to the value of 1 per
group). SPEC values for avian “Species of European Conservation Concern”
(BirdLife International 2004): 2- “concentrated in Europe and with an un-
favorable conservation status”; 3- “not concentrated in Europe, but with an
unfavorable conservation status”; 4- “concentrated in Europe, but with a fa-
vorable conservation status”.
Appendix 399
For the list of the 351 plant species, used for this analysis see Korakis et al. (2006),
available by the authors.
Species - SPEC Weight factor
Amphibians ---
Fire Salamander Salamandra salamandra -0.2
Yellow-bellied Toad Bombina variegata -0.15
Common Toad Bufo bufo -0.1
European Green Toad Bufo viridis -0.1
Common Spadefoot Pelobates fuscus -0.1
Smooth Newt Triturus vulgaris -0.1
European Tree Frog Hyla arborea -0.1
Marsh Frog Rana ridibunda -0.05
Balkan Stream Frog Ra na g raeca -0.05
Agile Frog Rana dalmatina -0.05
Small birds ---
Woodchat Shrike Lanius senator 20.1
Ortolan Bunting Emberiza hortulana 20.1
Black-headed Bunting Ember iz a me lanocephala 20.1
Woodlark Lullula arborea 20.1
Corn Bunting Milandra calandra 20.1
Bonelli’s Warbler Phylloscopus bonelli 20.1
Green Woodpecker Picus viridis 20.1
Olivaceous Warbler Hippolais pallida 30.05
European Bee-eater Merops apiaster 30.05
Orphean Warbler Sylvia hortensis 30.05
Red-backed Shrike Lanius col lurio 30.05
Middle Spotted Woodpecker Den drocopos medi us 40.05
Syrian Woodpecker Dendrocopos syriacus 40.05
Raptors ---
Eurasian Black Vulture Aegypius monachus 1 Special category
Egyptian Vulture Neophron percnopterus 3 Special category
Golden Eagle Aquila chrysaetos 30.3
Lesser Spotted Eagle Aqu il a pom ar in a 20.2
Booted Eagle Hierraetus pennatus 30.2
Black Stork Ciconia nigra 20.1
Short-toed Eagle Circaetus gallicus 30.1
Goshawk Accipiter gentilis -0.05
Honey Buzzard Pernis apivorus -0.03
Common Buzzard Buteo buteo -0.01
Sparrowhawk Accip it er ni su s -0.01
