Decreasing the fragmentation of old forests in landscapes involving multiple ownership in Finland: economic, social and ecological consequences

Decreasing the fragmentation of old forests in landscapesinvolving multiple ownership in Finland: economic,

social and ecological consequences

Mikko Kurttilaa,*, Janne Uutterab, Sakari Mykrac, Sami Kurkid, Timo Pukkalaa

aFaculty of Forestry, University of Joensuu, P.O. Box 111, FIN-80101 Joensuu, FinlandbForestry Development Centre Tapio, Soidinkuja 4, FIN-00700 Helsinki, Finland

cSection of Ecology, Department of Biology, University of Turku, FIN-20014 Turku, FinlanddUniversity of Helsinki, Seinajoki Institute for Rural Research and Training, Keskuskatu 34, FIN-60100 Seinajoki, Finland

Received 7 June 2000; received in revised form 2 March 2001; accepted 31 May 2001

Abstract

The management of biological diversity at the landscape level in Finland is complicated by the relatively small size of the

holdings and the fact that the habitats of species do not necessarily follow the holding borders. Therefore, there is a great need

to develop such forest-planning approaches that aim at solving the problems that some species have faced due to modern

forestry, including the fragmentation of old forests. One way to deal with fragmentation is to aggregate old forests into larger

patches, a goal that may be achieved by excluding some areas from forestry operations or by restricting the intensity of

treatments. In a planning area under multiple ownership, the sizes, shapes and relative locations of set-aside areas are

important, since they affect the social equity of the proposed forest plans. In this study, the effects of two conflicting goals, i.e.

the improvement of old-forest patterns, and the equal participation of forest owners were studied in a privately owned forest

area in Finland. This was made possible through the examination of the consequences of alternative strategies to delineate

areas set aside. The two main results of the study were, first, that it is difficult to reach these goals simultaneously, and, second,

the time interval relevant in the management planning of private forestry is quite short with respect to efforts to change

landscape structure significantly. However, in real planning situations, it is important to specify the locations of forest holdings

whose owners are willing to acknowledge biodiversity values above the level defined in the legislation. In addition,

ecologically important areas should be located and connected to the above information. According to the results of this study,

it seems that a compensation system would facilitate the restructuring of the forest landscapes towards an ecologically

enhanced structure. # 2002 Elsevier Science B.V. All rights reserved.

Keywords: Biodiversity; Forest planning; Private forest holdings; Spatial structure of forests

1. Introduction

In Finland, the stand compartment and forest holding

are the most often used operational units in the manage-

ment planning of private forestry. Typically, stand-level

treatment alternatives are produced and combined to

yield an optimal forest plan at the forest-holding level

Forest Ecology and Management 166 (2002) 69–84

* Corresponding author. Present address: Finnish Forest Research

Institute, P.O. Box 18, FIN-01301 Vantaa, Finland.

Tel.: þ358-9-857-05-804; fax: þ358-9-857-05-531.

E-mail address: [email protected] (M. Kurttila).

0378-1127/02/$ – see front matter # 2002 Elsevier Science B.V. All rights reserved.

PII: S 0 3 7 8 - 1 1 2 7 ( 0 1 ) 0 0 6 6 3 - 6

withrespect togoalssetbytheforestowner.Biodiversity

maintenance has emerged as management objective

in private forestry during the past decade in Finland

(Luonnonlaheinen, 1994; Metsalaki, 1997). In a parti-

cular managed forest area ‘‘biodiversity maintenance’’

meanscomprehensivemaintenanceofall speciesknown

or assumed to be members of the native community

of the area. The components of natural diversity do

continue to exist in Finnish forests, but forest manage-

ment has vastly altered their distribution and quantity.

As the present potential and constraints concerning

the maintenance task vary between areas because of

diverse human interventions, the objectives of biodi-

versity maintenance must be allocated spatially (Mykra

and Kurki, 1998).

Current planning at the forest-holding level sup-

ports some stand-level biodiversity maintenance

operations, including the preservation of the biotopes

listed in the current forest law (Metsalaki, 1997) and

the application of ecologically justified forest manage-

ment practices (Luonnonlaheinen, 1994). It does not,

however, support adequately biodiversity maintenance

efforts in larger areas. Forest holding borders are often

administratively delineated and may not coincide with

natural borders. Similarly, forest plans are formed

without considering the possible interdependencies

between forest holdings with respect to ecological

characteristics, and without the co-operation of forest

owners. This planning tradition and the mosaic-like

ownership pattern have resulted in fine-grained and

fragmented forest landscapes in Finland (Uuttera et al.,

1998). Compared to the natural forest landscape

structure the mosaic of successional stages is small in

scale, and variation in the size distribution of the patch

mosaic in particular has diminished (Mykra et al.,

2000). This causes adverse ecological effects in the

form of habitat loss and habitat fragmentation: the

case is that in managed forest landscapes the important

resources of a number of forest-dwelling species

(including animals, plants and fungi) occur in small

and isolated patches (e.g. Rolstad and Wegge, 1987;

Esseen et al., 1992; Raivio, 1992; Andren, 1994, 1997;

Haila, 1994; Helle and Linden, 1994; Siitonen and

Martikainen, 1994; Jokimaki and Huhta, 1996;

Monkkonen et al., 1997; Petterson, 1997; Wikars,

1997).

The above-mentioned biodiversity maintenance

operations respond to the requirements of a large

number of forest-dwelling species. However, several

recent studies stress the importance of the landscape

aspect (e.g. Angelstam, 1992; Dunning et al., 1992;

Hansson, 1992; Wiens et al., 1993; Andren, 1994,

1997; Monkkonen et al., 1997; Kurki et al., 1998,

2000). At the landscape level, one important objective

relates to the existence and pattern of old forests on

appropriate spatial and temporal scales. The loss of

habitats with old-forest characteristics has deleterious

effects on many forest-dwelling species in different

taxa in Fennoscandia (Helle and Jarvinen, 1986;

Virkkala, 1987, 1991; Angelstam and Mikusinski,

1994; Siitonen and Martikainen, 1994; Edenius and

Elmberg, 1996; see also Esseen et al., 1997). In

privately owned forests old forests typically occur in

small and isolated patches. When the average patch

size becomes too small and the inter-patch distance

too great, the landscapes cease to be functional for

some species depending on old forests. Mykra et al.

(2000) examined the spacing of mature forest

(growing stock volume >100 m3/ha; 80–100 years)

in relation to the species-specific scales of several

resident bird and mammal species (flying squirrel,

Siberian tit, Siberian jay, three-toed woodpecker) in

managed forest environment in north-eastern Finland

and found that continuous mature forest areas were

practically non-existent for these species whose home-

range size is 20–50 ha.

The ecological reasons for increasing the old-forest

patch size are, therefore, connected to the dispropor-

tion between the required home-range size of indivi-

dual animals and the small and invariable mosaic

‘grain’ size. Basically, the apparent graininess of any

habitat mosaic cannot as such be defined as ‘fine’ or

‘coarse’ due to the different scale dependencies of the

species. As the scale and pattern of stand mosaic is

mainly of artificial origin in the managed forests in

Fennoscandia, it seems sound to refer to the ‘grain’

definition originally presented by Levins (1968); the

habitat mosaic is fine-grained for an individual if

the patches are smaller than the home-range size and

the animals encounter several habitat types in their

routine activities. It is worthwhile to increase the

patch size of old forests whenever the relevant species

experience the present habitat mosaic as excessively

fine-grained. Increasing the old-forest patch size

produces ecological benefits also at the population

level. The probability for the occurrence of viable

70 M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84

populations of many poorly-dispersing old-forest

specialists (e.g. species depending on dead or dying

trees) usually increases with the increasing habitat

patch size. Furthermore, in small habitat patches, the

proportion of edges is greater, increasing the effect of

surrounding areas on the patch interior (see Angelstam,

1992, for review of edges in boreal environments).

This study has sought to investigate the effects of

the two conflicting goals in landscape-level forest

planning. The first goal, to achieve positive changes

in the old-forest pattern, was aimed at by excluding

parts of the forest from management operations during

the planning period. The second goal was related to the

equity of forest owners in the planning area. While the

areas set aside were placed differently, their total area

was always the same. The selection of these areas was

based on spatial rules. Carlsson et al. (1998), who

examined differences in the abundance, size and shape

of ecologically-important habitats (e.g. key biotopes,

old broad-leaved successions, riparian zones and old

forests) and their occurrence in forest areas covering

several private forest holdings, concluded that e.g.

legislative treatment restrictions on these habitats

would cause varying economic impacts between

individual forest owners in the planning area. For

some of the above-mentioned habitat types, their size,

shape, abundance and location are largely predefined

and cannot be influenced (e.g. key biotopes with

specific hydrology, geology or topography), whereas

in the case of some other habitat types (e.g. old forests)

there exist more possibilities to influence these

characteristics. Our simulation study examines one

broadly defined habitat type, namely old forests.

Possibilities to affect their spatial pattern during the

planning period in connection to resulting economic

and social effects are examined.

2. Materials and methods

2.1. Study area

The planning area is located in north Karelia,

Finland. It includes 44 non-industrial private forest

holdings over 5 ha in size (the holdings range from 5.1

to 395.1 ha). The total forest area is 1913.6 ha. This

area was divided into 1521 stands (Fig. 1A) in the

forest inventory carried out by the local Forestry

Centre. In the beginning of the planning period, the

average mean volume of the growing stock was

153.1 m3/ha (varying between 65.9 and 250.8 m3/ha

in individual holdings). The proportions of pine (Pinus

sylvestris), spruce (Picea abies) and broad-leaved

trees were, respectively, 64.5, 23.1 and 12.4%. The

initial age distribution was as follows: younger than 20

years 19.1%; 20–39 years 25.2%; 40–59 years 16.1%;

60–79 years 16.6%, and more than 80 years 23.1%. The

current annual increment was estimated at 5.9 m3/ha.

The length of the planning period was 30 years,

which was divided into three 10-year sub-periods. One

to 12 alternative treatment schedules, differing from

each other mainly with respect to the timings and types

of fellings, were simulated for each compartment. The

total number of treatment schedules was 5644. For

most compartments, one of the simulated treatment

schedules was the ‘‘no treatments’’ option. The other

treatment schedules were simulated in accordance

with Finnish forest management recommendations

(Luonnonlaheinen, 1994).

2.2. Planning model

The alternative forest plans for each forest holding

were composed on the basis of the simulated

schedules using the Monsu forest planning software

(Pukkala, 1998) and a goal programming (GP) model

adopted from Kangas and Pukkala (1992). The GP

model was as follows:

Min z ¼X3

t¼1

wtdNI�tNImax

t

(1)

subject to

Xn

j¼1

Xnj

i¼1

cijxij þ dNI�t � NIobjt ; t ¼ 1; 2; 3 (2)

Xn

j¼1

Xnj

i¼1

aijxij � SV (3)

Xn

j¼1

Xnj

i¼1

bijxij � RAt; t ¼ 1; 2; 3 (4)

Xnj

i¼1

xij ¼ ej; j ¼ 1; . . . ; n (5)

xij � 0; i ¼ 1; . . . ; nj; j ¼ 1; . . . ; n (6)

M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84 71

Fig. 1. (A) Forest stand borders and the initial locations of old-forest stands (age �80 years) in the study area; (B) five 50-ha circles,

Alternative 1; (C) five 50-ha rectangles (1500 m � 333 m), Alternative 2; (D) an example of the locations of 10-ha circles in Alternative 3; (E)

selected forest areas in Alternative 4, in which the areas were located relatively equally between the forest holdings; (F) an example of the

locations of 10-ha circles in Alternative 5 in which 10-ha circles were located more in and around the eight forest holdings whose owners were

assumed to have nature-oriented objectives.

72 M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84

The model included (1) the objective function,

where the weighed sum of relative negative deviations

from the aspired level of net income during three

sub-periods t was minimized; (2) a goal constraint

for the net income for each sub-period t; (3) a

production constraint that ensures that the stumpage

value at the end of the planning period is at least SV;

(4) resource constraints that regulate the regeneration

areas (RAt) during each sub-period t; (5) area con-

straints for each compartment; and (6) non-negativity

constraints. Here, dNI�t is the amount that the income

goal falls short during period t, wt the weight given to

Fig. 1. (Continued ).

M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84 73

each unit of deviation, NImaxt the maximum value of

net income during sub-period t, NIobjt the aspired net

income of period t, n the number of stands, nj the

number of treatment alternatives for stand j, ej the area

of stand j, xij the area of stand j treated according to

the schedule i, cij the net income per hectare that

the treatment schedule i of stand j produces, aij the

stumpage value per hectare that the treatment schedule

i of stand j gives at the end of the planning period, and

bij is equal to one if stand j is regenerated according to

schedule i during the period t (otherwise it is equal to

zero).

Fig. 1. (Continued ).

74 M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84

In the GP models the aspired level of each 10-year

sub-period’s net income (NIobjt ) was set to be equal to

the current annual value increment multiplied by 10. If

the achieved value of the objective function was zero,

i.e. the solution was dominated, the aspired level of

each sub-period’s net income was increased by steps

of 5% until the objective function value was greater

than zero. The maximum regeneration area during

each sub-period (RAt) was set as 15% of the forest

area of the holding. The stumpage value at the end of

the planning period (SV) was set to be at a level where

it was 20% higher than at the beginning of the period.

The scaling of the stumpage value to a higher level was

intended to imitate the forest owners general goal of

developing their property. Together with the regen-

eration area constraint, it was supposed to ensure that

all cutting possibilities are not utilized during the

planning period. The weights (wt) given to each unit of

the underachievement deviation were equal. Together

with the usually increasing maximum value of net

income from sub-periods one to three (NImaxt ), this

resulted in an objective function in which it was more

important to achieve the aspired net income during

the first sub-period rather than the other periods. The

overall goal was, however, a relatively constant flow

of income.

2.3. Alternative setting-aside strategies

We created five strategies to place the set-aside

areas. The purpose of all strategies was to increase the

patch size of old forest. A forest was defined as old if

it had reached the age of 80 years. In addition to

clustering old forests, the strategies also aimed at

equality in the impacts to forest owners.

In all alternatives the area excluded from any

forestry operations during the planning period was

10% of the forest area (191.4 ha). Using GIS routines,

the areas were delineated, according to the preset

parameters, by creating polygons of varying size and

shape around the current old forests. Then the created

polygon layer was overlaid with one containing the

borders of forest holdings and stands. As a result,

many stands were split up. When the polygons were

created, the 10% forest area limit was exceeded in all

alternatives. The final selection of the stands inside the

polygons removed from forestry operations was based

on the stand area and was carried out by selecting

larger stands first. The stochastic delineation processes

in alternatives 3 and 5 were repeated five times and the

averages are reported. The created alternatives were as

follows.

2.3.1. Alternative 1

The first alternative aimed at the most clustered

pattern of old forests. It was based on the prevailing

situation in the study area. Five visually distinguish-

able clusters of old-forest stands were selected and a

50-ha circle centered upon these clusters was

simulated (Fig. 1B). The use of a circular shape is

based on the fact that it produces the lowest perimeter/

area ratio and, thus, also the greatest proportion of

interior conditions free from the effects of the

surrounding different habitats.

2.3.2. Alternative 2

Alternative 2 differs from Alternative 1 in the

respect that instead of using circles, five rectangles

(1500 m � 333 m) were simulated. They were located

within the same old-forest clusters as in Alternative 1

(Fig. 1C). The direction of the longer side of the

rectangles was set subjectively so that the rectangles

cover as much old forest as possible and point to

some other nearby rectangle. Therefore, in addition

to clustering the stands, there was an attempt to

enhance the connectivity of the landscape. The width

of the rectangles was, however, kept sufficient to

reduce the edge effects inside the rectangles.

2.3.3. Alternative 3

In Alternative 3, all center points of the old-forest

stands were mapped. Then 40 of them were selected

randomly and a 10-ha circle simulated around each

one of them (Fig. 1D). The overlaps of two or more

circles, as well as the crossings of study area borders,

were not prevented in this or any other alternative.

2.3.4. Alternative 4

The aim of Alternative 4 was to distribute the areas

excluded from forestry operations relatively equally

among forest holdings. A 10-ha circle was simulated

around every old-forest stand’s center point. Inside

these circles, a maximum of 10% of a holding’s forest

area was selected for protection. Again, larger stands

were selected first. This yielded 186.2 ha of forest

land. To complete the 10% forest area objective, the

M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84 75

rest of the area (5.2 ha) was selected randomly

between the remaining stands located inside the

circles (Fig. 1E). The randomly selected area in this

alternative was considered so small that no repetitions

were made.

2.3.5. Alternative 5

In Alternative 5, the assumption utilized was that in

addition to timber production, some forest owners

may want biodiversity maintenance. In such a case, it

is reasonable to aggregate the areas where no forestry

operations are allowed in these forest holdings. Eight

forest holdings (18% of the forest owners) were

randomly selected (Pesonen et al., 1995; Karppinen,

1998). All center points of the old-forest stands inside

these eight holdings were selected and a 10-ha circle

was simulated around them (Fig. 1F). As in Alter-

native 3, the number of 10-ha circles was set to 40. If

this number was unattainable inside the selected eight

holdings, the rest of the 10-ha circles were placed

randomly around the center points of other old-forest

stands. It was assumed that the protection of areas

inside the selected eight forest holdings would not lead

to any financial costs.

2.4. Evaluation of the effects of the setting-aside

strategies

The consequences of the setting-aside strategies

were examined from economic, social and ecological

viewpoints. The calculations of the economic and

social effects of the alternatives are based on the

forest-holding-level differences between the cutting

income from the basic forest plan, developed for each

holding without excluding any areas from forestry

operations, and from a forest plan prepared in each

alternative for each forest holding. When the latter

forest plans were produced, the GP model (Eqs. (1)–

(6)) was adjusted according to the available produc-

tion-forest area. For example, the stumpage value of

the growing stock located inside the excluded areas

was not taken into account in constraint 3 of the GP

model. The sums of the differences between the basic

plan and Alternatives 1–5 reveal the effects over the

entire planning area. In this study, the social effects

refer to the equity of forest owners. They are described

by between-holding variation in the relative differ-

ences of loss of income from cutting.

The ecological effects of excluding certain areas

from forestry operations on the pattern of old forests

were measured at the end of the 30-year planning

period. The compartments of the planning area were

classified into two classes, old forests and other areas.

In optimization, some compartments were split due to

the constraints. When one part of a compartment was

old forest at the end of the planning period and the

other part was not, the classification was carried out in

accordance with the larger part. The total area of old

forest, mean old-forest patch size, and its standard

deviation (S.D.), the proportion of over 10-ha old-

forest patches, and the mean distance to the nearest

neighbor (MNND) were computed for each alter-

native by using GIS routines and the FRAGSTATS

software (McGarigal and Marks, 1995). The compu-

tations were carried out at the landscape level. In

addition, the computations were made by using only

old forests inside the simulated polygons in order to

assess the effects of alternative strategies on smaller

scale.

3. Results

In Alternative 1, in 32% of the forest holdings some

forest areas were suggested to be set aside (Table 1).

The set-aside areas in Alternative 2 were more evenly

distributed than in Alternative 1 because of the use of

rectangles that influence a larger proportion of forest

holdings than circles do. In Alternative 4, where the

aim was to divide the set-aside areas equally among

forest holdings, the area of forests decreased in all but

one of the forest holdings. Here the mean relative

decrease in forest area was 10%, whereas it was more

than three times higher in Alternative 1. In Alternative

4, the set-aside areas were distributed evenly accord-

ing to the standard deviation in the relative decrease

of forest area. The higher standard deviation of

Alternative 5, when compared to the otherwise similar

Alternative 3, can be explained by pointing to the fact

that in Alternative 5 eight forest holdings whose

owners were assumed to have objectives related to

biodiversity maintenance lost on average a larger

proportion of their forest area; the mean relative

decrease calculated without the eight selected forest

holdings was 13.0% and its standard deviation was

10.2%.

76 M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84

The effects of Alternatives 1–4 on total net income

were relatively equal (Table 2) and the relative

decreases in total net income were close to decreases

in forest area. Minor differences were caused by

variation in the quantity of forests that would have

yielded cutting income during the planning period but

were now excluded from wood production. This was

not taken into account in the selection of the set-aside

areas, however, since the selection principles con-

cerned only stand size and its location relative to initial

old-forest stands. In Alternative 5, the decrease in total

net income was one-third of that of other alternatives.

The maximum decrease in relative net income was

largest in Alternative 1. The standard deviation in the

proportional decrease of total net income was smallest

in Alternative 4, but almost four times higher than

the standard deviation in the proportional decrease

of forest area (Table 1). Consequently, some smaller

forest holdings lost those compartments that would

have been cut during the 30-year planning period.

Thus, the selected areas represent a substantial

proportion of possible cutting income from certain

forest holdings in Alternative 4 as well. In Alternative

5, when compared to Alternative 3, the aggregation

of areas in and around such forest holdings whose

owners were assumed to have objectives related to the

maintenance of biological diversity resulted in a more

even and acceptable distribution of economic and

social effects.

Different alternatives resulted in rather equal

proportions of old forests in the whole planning area

at the end of the planning period (Table 3). In the basic

plan, the proportion of old forests was 2 to 3% lower

than in Alternatives 1–5. The 30-year planning period

was long enough to make it visually evident that some

of alternatives produced a more clustered pattern of old

forests (Fig. 2B and C); yet the effects of selecting

only 10% of the forest area to be set aside are not

visually very clear in other alternatives (Fig. 2D–F). In

all alternatives, and particularly in Alternatives 1 and

2, the mean patch size and patch size S.D. were bigger

than in the basic plan. Also, the largest proportion

of old-forest patches over 10 ha in size were created

in these alternatives. Alternative 5 had a slightly more

clustered old-forest pattern than Alternative 3, although

the differences were relatively minor.

The differences in the spatial pattern of old forests

between the alternatives were more discernible when

Table 1

Proportion of forest holdings affected by set-aside areas in the alternative strategies, mean (and maximum) relative decreases in the forest area

and its standard deviation (S.D.)a

Alternative Proportion of forest

holdings affected (%)

Mean relative decrease in forest

area (and maximum decrease) (%)

S.D. of the relative

decrease in forest area (%)

1 31.8 33.9 (68.9) 23.1

2 40.9 23.5 (54.8) 19.6

3 50.0 17.7 (45.4) 11.8

4 97.7 10.1 (21.2b) 2.5

5 43.2 28.3 (85.3) 25.8

a Forest holdings not affected are excluded.b The 21.2% maximum decrease was in a 5.1-ha forest holding in which a 0.5-ha stand, in addition to the 10% area goal, was selected.

Table 2

Effects of alternative strategies on the net income during the planning perioda

Alternative Decrease in the total net income,

FIM 1000 (and relative decrease, %)

Mean relative decrease in net

income (and maximum decrease) (%)

S.D. of the relative

decrease in net income (%)

1 4616 (11.7) 30.1 (65.5) 22.2

2 4063 (10.3) 22.5 (54.8) 18.8

3 3944 (10.0) 18.9 (57.8) 14.6

4 4600 (11.7) 13.8 (40.7) 9.5

5 1342 (3.4) 8.0 (47.0) 13.5

a S.D.: standard deviation. Forest holdings not affected are excluded.

M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84 77

they were computed inside the simulated polygons

only. What was shared by all alternatives was that at

the end of the planning period more than 60% of the

protected areas had reached the old-forest threshold

age as set in the present study (Table 4). Alternative 1

resulted in the most densely clustered structure of

old forests. The mean distance to the nearest neighbor

was larger in Alternative 1 when compared to other

alternatives. Thus, only one large patch was formed

inside some of the 50-ha circles in Alternative 1. In

Alternative 5, the aggregation of areas in and around

forest holdings whose owners had different objectives

was beneficial in the light of the variables describing

the spatial structure of the old forests inside the

simulated circles.

4. Discussion

This simulation study has illustrated the effects of

alternative strategies concerning the locations and

sizes of forest areas excluded from forestry operations

when the goal is to decrease the fragmentation of old

forests. Some improvements in the pattern of old

forests were found out. The strategies resulted in a

very uneven distribution of effects on individual forest

owners. The achieved results are always partly specific

to the ownership pattern, age-class distribution and

initial pattern of forests in the planning area. However,

they are in line with the hypotheses presented by

Carlsson et al. (1998), hypotheses that concern the

distribution of habitats with respect to their abun-

dance, size and shape. It is, however, difficult to

connect the improvements achieved directly to the

population viability of the forest-dwelling species. As

the mosaic pattern of the landscape and the scale of its

utilization are species-specific, they should also be

assessed from that point of view.

The differences in the landscape structure between

the alternatives were logical but rather minor. The

main reason for the small differences was that forest

management actions were proposed to be restricted

Table 3

The pattern of old forests initially, in the basic plan and in the five alternatives after the 30-year planning period over the whole planning areaa

Proportion of

old forests (%)

Mean patch

size (ha)

Patch

size S.D. (ha)

MNND (m) Proportion of old forests

in >10-ha patches (%)

Initial 23.1 3.3 4.7 90.9 37.9

Basic plan 21.2 2.7 4.4 87.1 39.3

Alternative 1 23.2 3.6 7.6 93.2 57.4

Alternative 2 23.3 3.4 7.5 92.0 50.9

Alternative 3 24.0 3.2 5.2 83.5 45.5

Alternative 4 24.1 3.0 4.8 77.2 38.1

Alternative 5 24.0 3.2 5.5 91.7 48.5

a S.D.: standard deviation, MNND: mean distance to nearest neighbor.

Table 4

The pattern of old forests in the five alternatives after the 30-year planning period inside simulated circles and rectanglesa,b

Alternative Proportion of old forests

in set-aside areas (%)

Mean patch size (ha) Patch size S.D. (ha) MNND (m)

1 65.7 14.6 11.1 689.0

2 67.5 9.4 10.6 227.8

3 66.9 3.2 2.8 179.3

4 60.0 2.8 2.7 146.7

5 62.9 4.2 4.7 192.4

a S.D.: standard deviation, MNND: mean distance to nearest neighbor.b In Alternative 4, only those circles which overlapped with the selected set-aside areas were used in the calculations. The area of simulated

circles and rectangles exceeded 191.4 ha area in all the alternatives and some stands that are used for wood production are included in these

figures.

78 M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84

from a small proportion of the forest area. In

addition, the planning horizon was rather short,

especially when compared to the rotation lengths

(exceeding 80 years) applied in Finnish forestry. All

protected stands did not reach the threshold age set

for old forests within the time covered in this

simulation study. Were the planning period extended

and all the areas set aside allowed to reach the

threshold old forest age, the differences between the

alternatives might be clearer. However, changes in

Fig. 2. The locations of old-forest patches after a 30-year planning period in: (A) basic plan; (B) Alternative 1; (C) Alternative 2; (D)

Alternative 3; (E) Alternative 4; and (F) Alternative 5.

M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84 79

ownership (the average duration of ownership in

Finland is 30 years), forest owners objectives,

roundwood markets and forest legislation often make

long-term forest plans obsolete much before the end

of planning horizon. Yet voluntary commitment to

forest certification may require that forest plans are

fairly strictly followed.

In a number of landscape ecological forest plans

prepared in Finland and Sweden, some areas have

been excluded from forestry operations or their

treatments have been considerably restricted. For

example, in the landscape ecological plans prepared

by the Finnish Forest and Park Service, treatments are

prohibited, totally or at least partly, in 5–10% of the

Fig. 2. (Continued ).

80 M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84

planning area (Itkonen, 1998). These areas include

ecological corridors, ‘stepping stones’ (i.e. fragmen-

tary corridors), habitats of endangered species, and

other valuable natural entities. In a landscape

ecological plan prepared according to the Swedish

stream model, 41 forest owners protected 3.4% of their

2440 ha of forest area (Fries et al., 1998). The central

component of the model is the idea of stream corridors

that influence a large proportion of forest estates,

connecting other protected areas. However, the degree

to which the corridors may assist in the dispersal of

species, and how they are utilized as habitats or

Fig. 2. (Continued ).

M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84 81

sources of food in boreal conditions, is not entirely

clear (e.g. Monkkonen and Reunanen, 1998).

The fragmentation of old forests can also be

examined by applying spatial rules that aggregate or

disperse the cuttings (e.g. Franklin and Forman, 1987;

Baskent, 1997). At the landscape or forest-holding-

level, spatial cutting strategies such as the ‘‘nuclei

progressive harvesting pattern’’ and the ‘‘scatter

harvesting pattern’’ (Baskent, 1997) could also be

applied. These approaches would probably show

clearer effects on landscape structure than the

approach used in the present study, since in them

the whole planning area is under control. This will,

however, remain the subject of a separate future study.

If we wish to achieve results in this case where

the goals are contradictory, the best way to decrease

the fragmentation of old forests might be to include

appropriate spatial objectives in the optimization

calculations of forest planning (Kurttila, 2001). The

acceptability of the plan can be increased by introduc-

ing owner-specific constraints. This approach would

place the protected areas optimally over forest holding

borders. Therefore, the development of practical

planning and decision-supporting tools that in an

illustrative manner, enable participation of several

forest owners and the use of spatial objectives seems to

be desirable in the near future.

The definition of the old forest was based on the age

of the stand. This simplistic criterion was applied

because our aim was mainly to illustrate character-

istics embedded in planning approaches involving

multiple ownership and spatial goals. The threshold

age of old forest was set at 80 years, because it is

one of the criteria for economic maturity applied

in Finnish forestry. The well-being of a species is,

however, only rarely directly connected with the age

of trees. In addition, a clear distinction must be drawn

between natural old-growth forests, which are quite

rare, and old forests in a managed landscape, the latter

being the subject of the present study. True old-growth

specialists are tightly adapted to variable resource

combinations, structural elements, micro habitats and

community structures which materialize in natural

climax forests (Mykra et al., 2000); these character-

istics were not examined in this study. More detailed

criteria may also be used if the viability of some

species is examined and the home-range requirements

and typical daily movement distances for this species

are known. For species with relatively limited move-

ment ability, home-range requirements must be fulfilled

within a single habitat patch, i.e. the individuals

perceive the environment in a coarse-grained manner.

These home-range requirements can be used in

determining the size of set-aside areas. For species

with a relatively good ability to move across forest

landscape, the pattern of habitats is less important and

the home-range requirement can be sampled among

several patches within a typical movement scale.

Remarkable differences between economic impacts

on individual forest holdings were generated in all

alternatives. Indeed, few forest owners are actually

ready to agree to such distribution of protected areas

and resulting economic losses. However, some kind of

utilization of existing clusters of old forests (or other

resources) is needed for the maintenance of biodi-

versity because the relevant time period in private

forestry is short, impeding the formation of resource

clusters in entirely new locations. A network of

smaller protection areas may be able to replenish

these existing resource clusters. As a result, the results

will appear to be more acceptable to a greater number

of participating forest owners. Forest owners who

give more weight to biodiversity maintenance are

the key actors when ecologically important resources

are aggregated. This makes three points important.

Firstly, forest owners goals and willingness to

contribute to biodiversity maintenance need to be

clarified exhaustively during the planning process.

It is important to specify the locations of those

forest holdings whose owners are willing to acknowl-

edge biodiversity values above the level defined in the

legislation. Secondly, current and potential future

locations of ecologically important entities (e.g.

critical sites or resources for declined or endangered

species) must be clarified and this information

connected to the findings obtained in the previous

stage. Finally, a compensation system would facilitate

the planning process and achievement of ecological

goals. In fact, the need for forest protection in

southern Finland and Ostrobotnia is currently exam-

ined (Forest protection, 2000), and some of the

resulting operations are expected to affect private

forests. In the ongoing process, the quality and

amount of protected areas are important. However,

this study also emphasizes that the objectives of forest

owners of these areas should be considered, because it

82 M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84

could increase the acceptability of operations and

decrease the economic costs of protecting forests.

Acknowledgements

The authors wish to thank Mr. Janne Loikkanen for

his technical assistance and the two journal referees

for their helpful comments on an earlier version of the

manuscript.

References

Andren, H., 1994. Effects of habitat fragmentation on birds and

mammals in landscapes with different proportions of suitable

habitat: a review. Oikos 71, 355–366.

Andren, H., 1997. Habitat fragmentation and changes in biodi-

versity, boreal ecosystems and landscapes: structures. Ecol.

Bull. 46, 171–181.

Angelstam, P., 1992. Conservation of communities—the impor-

tance of edges, surroundings and landscape mosaic structure.

In: Hansson, L. (Ed.). Ecological Principles of Nature

Conservation. Elsevier, London, pp. 9–70.

Angelstam, P., Mikusinski, G., 1994. Woodpecker assemblages in

natural and managed boreal and hemiboreal forest—a review.

Ann. Zool. Fenn. 31, 157–172.

Baskent, E.Z., 1997. Assessment of structural dynamics in forest

landscape management. Can. J. For. Res. 27, 1675–1684.

Carlsson, M., Andersson, M., Dahlin, B., Sallnas, O., 1998. Spatial

patterns of habitat protection in areas with non-industrial

private forestry—hypotheses and implications. For. Ecol.

Manage. 107, 203–211.

Dunning, J.B., Danielson, B.J., Pulliam, H.R., 1992. Ecological

processes that affect populations in complex landscapes. Oikos

65, 169–175.

Edenius, L., Elmberg, J., 1996. Landscape level effects of modern

forestry on bird communities in north Swedish boreal forests.

Landsc. Ecol. 11, 325–338.

Esseen, P.-A., Ehnstrom, B., Ericson, L., Sjoberg, K., 1992. Boreal

forests – the focal habitats of Fennoscandia. In: Hansson, L.

(Ed.), Ecological Principles of Nature Conservation. Elsevier,

London, pp. 252–325.

Esseen, P.-A., Ehnstrom, B., Ericson, L., Sjoberg, K., 1997. Boreal

forests, boreal ecosystems and landscapes: structures, processes

and conservation of biodiversity. Ecol. Bull. 46 16-47.

Forest protection in southern Finland and Ostrobotnia. The Finnish

Environment 437. In Finnish with English summary, 284 p.

Franklin, J.F., Forman, R.T.T., 1987. Creating landscape patterns

by forest cutting: ecological consequences and principles.

Landsc. Ecol. 1, 5–18.

Fries, C., Linden, G., Nillius, E., 1998. The stream model for

ecological landscape planning in non-industrial private forestry.

Scand. J. For. Res. 13, 370–378.

Haila, Y., 1994. Preserving ecological diversity in boreal forests:

ecological background, research, and management. Ann. Zool.

Fenn. 31, 203–217.

Hansson, L., 1992. Landscape ecology of boreal forests. Trends

Ecol. E 7, 299–302.

Helle, P., Jarvinen, O., 1986. Population trends of north Finnish

land birds in relation to their habitat selection and changes in

forest structure. Oikos 46, 107–115.

Helle, T., Linden, H., 1994. Capercaillie (Tetrao urogallus) lekking

sites in fragmented Finnish forest landscape. Scand. J. For. Res.

9, 386–396.

Itkonen, P., 1998. Metsahallituksen alue-ekologinen suunnittelu. In:

Jokimaki, J., Kangas, J., Varmola, M., Virtanen, E. (Eds.),

Alue-ekologista tietoa metsasuunnitteluun—Metsantutkimus-

paiva Rovaniemella 15.10. 1997. Finn. For. Res. Inst. Res. Pap.

685, 85–90.

Jokimaki, J., Huhta, E., 1996. Effects of landscape matrix and

habitat structure on a bird community in northern Finland: a

multi-scale approach. Ornis Fenn. 73, 97–113.

Kangas, J., Pukkala, T., 1992. A decision theoretic approach

applied to goal programming of forest management. Silva Fenn.

26 (3), 169–176.

Karppinen, H., 1998. Objectives of non-industrial private forest

owners: differences and future trends in southern and northern

Finland. J. For. Econ. 4 (2), 147–173.

Kurki, S., Nikula, A., Helle, P., Linden, H., 1998. Abundances of

red fox and pine marten in relation to the composition of boreal

forest landscapes. J. Anim. Ecol. 67, 874–886.

Kurki, S., Nikula, A., Helle, P., Linden, H., 2000. Effects of forest

fragmentation and composition on breeding success of grouse.

Ecology 81, 1985–1997.

Kurttila, M., 2001. The spatial structure of forests in the

optimization calculations of forest planning—a landscape

ecological perspective. For. Ecol. Manage. 142, 127–140.

Levins, R., 1968. Evolution in Changing Environments. Princeton

University Press, Princeton, New Jersey.

Luonnonlaheinen metsanhoito, 1994. Metsanhoitosuositukset.

Metsakeskus Tapion julkaisuja 6, 74 p.

McGarigal, K., Marks, B.J., 1995. FRAGSTATS: spatial pattern

analysis program for quantifying landscape structure. USDA

Forest Service, PNW, General Technical Report, 351.

Metsalaki perusteluineen, 1997. Lakikokoelma. Oy Edita Ab,

Helsinki.

Monkkonen, M., Reunanen, P., 1998. Ekologiset kaytavat alue-

ekologisen metsasuunnittelun valineena. In: Jokimaki, J.,

Kangas, J., Varmola, M., Virtanen, E. (Eds.), Alue-ekologista

tietoa metsasuunnitteluun—Metsantutkimuspaiva Rovanie-

mella 15.10. 1997. Finn. For. Res. Inst. Res. Pap. 685, 39–

48.

Monkkonen, M., Reunanen, P., Nikula, A., Inkerioinen, J.,

Forsman, J., 1997. Landscape characteristics associated with

the occurrence of the flying squirrel Pteromys volans in

old-growth forests of northern Finland. Ecography 20, 634–

642.

Mykra, S., Kurki, S., 1998. ESC—strategy for rational operatio-

nalization of forest biodiversity maintenance in Finland. Silva

Fenn. 32 (4), 389–399.

M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84 83

Mykra, S., Kurki, S., Nikula, A., 2000. The spacing of mature

forest habitat in relation to species-specific scales in managed

boreal forests in NE Finland. Ann. Zool. Fenn. 37, 79–91.

Pesonen, M., Malinen, J., Kettunen, A., Yksityismetsanomistajien

puuntuotantostrategiat ja potentiaaliset hakkuumahdollisuudet.

Finn. For. Res. Inst. Res. Pap. 583, 61 p.

Petterson, R.B., 1997. Lichens, invertebrates and birds in spruce

canopies: impacts of forestry. Doctoral thesis, University of

Umea, Sweden.

Pukkala, T., 1998. MONSU metsasuunnitteluohjelma. Versio 1.2.

Ohjelmiston toiminta ja kaytto. Mimeograph.

Raivio, S., 1992. Bird communities in fragmented coniferous

forests: the importance quantitative data and adequate scaling.

Doctoral thesis. University of Helsinki, Finland.

Rolstad, J., Wegge, P., 1987. Distribution and size of capercaillie leks

in relation to old forest fragmentation. Oecologia 72, 389–394.

Siitonen, J., Martikainen, P., 1994. Occurrence of rare and

threatened insects living on decaying Populus tremula: a

comparison between Finnish and Russian Karelia. Scand. J.

For. Res. 9, 185–191.

Uuttera, J., Maltamo, M., Kurki, S., Mykra, S., 1998. Differences in

forest structure and landscape patterns between ownership

groups in central Finland. Boreal Environ. Res. 3, 191–

200.

Wiens, J.A., Stenseth, N. Chr., Van Horne, B., Ims, R.A., 1993.

Ecological mechanisms and landscape ecology, Oikos 66, 369–

380.

Wikars, L.-O., 1997. Effects of forest fire and the ecology of fire-

adapted insects. Doctoral thesis. University of Uppsala.

Sweden.

Virkkala, R., 1987. Effects of forest management on birds breeding

in northern Finland. Ann. Zool. Fenn. 24, 281–294.

Virkkala, R., 1991. Population trends of forest birds in Finnish

Lapland in a landscape of large habitat blocks: consequences of

stochastic environmental variation or regional habitat altera-

tion? Biol. Conserv. 56, 223–240.

84 M. Kurttila et al. / Forest Ecology and Management 166 (2002) 69–84