Sr dj an Keren Renzo Motta Dusan Rozenbergar Zoran Govedar Jurij Diaci

Comparison of structural attributes Comparison of structural attributes between between

old-growth and adjacent old-growth and adjacent managed managed forests in Bosnia and Herzegovinaforests in Bosnia and Herzegovina

Srdjan KerenRenzo Motta

Dusan RozenbergarZoran Govedar

Jurij Diaci

Birmensdorf, 2014

IntroductionIntroduction

Mixed forests composed of Mixed forests composed of Fagus Fagus sylvaticasylvatica L., L., Abies albaAbies alba Mill and Mill and Picea Picea abiesabies L.H. Karst represent ecologically L.H. Karst represent ecologically and economically most important forest and economically most important forest category in Bosnia and Herzegovina. category in Bosnia and Herzegovina.

Of special interest to scientific Of special interest to scientific community are also three wide-known community are also three wide-known mixed old-growth forests (OGF): mixed old-growth forests (OGF): PeruPeruccica, Lom and Janj. ica, Lom and Janj.

IntroductionIntroduction An important aspect of sustainability An important aspect of sustainability

verification is a comparison of managed verification is a comparison of managed forests with unmanaged references. forests with unmanaged references.

Specific objectives of this study were toSpecific objectives of this study were to test for differences in terms oftest for differences in terms of:: tree tree density, density, dbh distribution shapes, tree dbh distribution shapes, tree species composition, mean diameters, BA, species composition, mean diameters, BA, growing stock and CWDgrowing stock and CWD..

Materials and methodsMaterials and methods The research was carried out The research was carried out in in old-growth old-growth

forests Janj and Lom and ten adjacent forests Janj and Lom and ten adjacent managed stands (by five on both sites), managed stands (by five on both sites), which are located on mountainous massifs of which are located on mountainous massifs of central and western Bosnia and central and western Bosnia and Herzegovina.Herzegovina.

OGF Janj OGF Lom

Area 57.2 ha (core area) 55.8 ha (core area)

Geografic position 44°08' N, 17°17' E 44° 27' N, 16°28' E

Bedrock dolomite limestone

Altitude a.s.l. 1240 – 1400 m 1250 – 1522 m

Mean annual rainfall 1200 mm 1600 mm

Mean ann. temperature 5.0 °C 3.5 °C

Year of measurement 2011 2011

Plant association Piceo-Abieti-Fagetum Piceo-Abieti-Fagetum

MeasurementsMeasurements

Regular 100Regular 100x100x100 m grid m grid of of sampling sampling plots plots was superimposedwas superimposed (total 80 plots in OG (total 80 plots in OG forests and 120 plots in managed forests)forests and 120 plots in managed forests)..

FFollowing inventories were carried out:ollowing inventories were carried out:

in circular plotin circular plotss (r = 12 m) species dbh for (r = 12 m) species dbh for all living trees above 7,5 cm were all living trees above 7,5 cm were measured. Heights of living trees were measured. Heights of living trees were measured for a sample of 100 trees for measured for a sample of 100 trees for each species (spruce, fir, and beech). Ineach species (spruce, fir, and beech). In smaller inner smaller inner plotplotss (r = 5 m) species and (r = 5 m) species and height of regeneration individuals from 10 height of regeneration individuals from 10 cm height to 7,5 cm dbh were tallied.cm height to 7,5 cm dbh were tallied.

Figure 2. Outline of sample plot and subplots

Measurements Measurements

CWD measurements were carried out on two 50 m CWD measurements were carried out on two 50 m line intersect oriented northward (the first) and line intersect oriented northward (the first) and eastward (the second) from the centre of the eastward (the second) from the centre of the sampling pointsampling point..

Results

Tree densities (N/ha)

Statistics OGF Janj MF Janj OGF Lom MF Lom

Mean 516.4 681.5 489.4 595.6

Std. Error of Mean 24.5 27.5 18.9 23.2

Median 487.0 663.0 498.0 553.0

Std. Deviation 155.1 213.3 120.1 179.3

Range 752.0 774.0 669.0 730.0

Independent t-tests involving total means of number of trees indicated significant differences between pairs OGF Janj - MF Janj (p = 0.0000) and OGF Lom - MF Lom (p = 0.0010), but also between MF Janj and MF Lom (p = 0.0190)

Diameter distributions (sensu Janowiak et al., 2008)

The best-fitting multiple regression models for dbh frequency distributions

MF Janj (all five stands)

OGF Janj

Fir Spruce Beech Total Fir Spruce Beech Total

RMSE 0.03 0.04 0.01 0.03 0.18 0.14 0.07 0.11

Adj. R2 0.99 0.99 1.00 0.99 0.86 0.85 0.99 0.97

N (dbh class)

8 7 6 8 12 11 7 12

Shape RS IQ RS RS UNI UNI RS RS

MF Lom (all five stands)

OGF Lom

RMSE 0.18 0.18 0.15 0.16 0.33 0.22 0.04 0.21

Adj. R2 0.92 0.94 0.96 0.96 0.74 0.85 0.99 0.94

N (dbh class)

9 9 8 9 12 11 7 12

Shape NE UNI CO IQ IQ IQ RS NE

Mean diameters (cm)

Independent t-tests indicated significant differences in mean diameters at p ≤ 0.05 between all possible pairs: OGF Janj vs. MF Janj, OGF Lom vs. MF Lom,OGF Janj vs. OGF Lom, and MF Janj vs. MF Lom.

Statistics OGF Janj MF Janj OGF Lom MF Lom

Mean 41.3 26.5 35.4 30.4

Std. Error of Mean 1.1 0.6 0.8 0.8

Median 41.0 25.9 35.1 29.5

Std. Deviation 7.1 4.5 5.3 5.8

Range 32.4 21.5 26.0 28.9

Statistics OGF Janj MF Janj OGF Lom MF Lom

Mean BA (m2/ha) 66.7 35.5 47.1 41.5

Mean GS (m3/ha) 1215.1 498.9 763.1 664.3

Basal area (BA) and growing stock (GS) in studied forests

Results of t-tests showed no significant difference in basal areas between MF Lom and OGF Lom, while between other pairs difference was statistically significant at p ≤ 0.05.

Tree species composition above 7.5 cm dbhTree species composition above 7.5 cm dbh

OGF Janj was similar to MF Janj regarding basal area composition, but very different regarding tree number composition.OGF Lom and MF Lom differred considerably upon both criteria.

Considering composition based on Considering composition based on tree numbertree number non-significant results non-significant results were yielded were yielded for pairs OGF for pairs OGF

Janj Janj vsvs. OGF Lom. OGF Lom and MF Janj and MF Janj vsvs. MF Lom, while . MF Lom, while significant difference at significant difference at pp ≤ 0.05 was indicated for ≤ 0.05 was indicated for two pairs: two pairs:

OGF Janj OGF Janj vsvs. MF Janj . MF Janj χχ22 = 26.77 = 26.77

OGF Lom OGF Lom vsvs. MF Lom χ. MF Lom χ22 = 16.81 = 16.81

Considering composition based on Considering composition based on basal areabasal area similar outcomes of testing were obtained as shown similar outcomes of testing were obtained as shown aboveabove

Replacement patterns and neighborhood effects

In all studied forests overall 14 transition matrices were calculated. One example from OGF Janj (poles to upper story) is shown here:

PolesBeech Fir Spruce Maple

Upperstory trees

Beech 0.74 0.15 0.10 0.00Fir 0.83 0.10 0.07 0.00Spruce 0.83 0.09 0.08 0.00

Transition probability matrices indicated domination of beech regeneration under mature trees in OGF Janj and both managed forests, whereas beech poles had potentially greatest chance to reach the upperstory in both old-growth forests. Silver fir poles were given rather high chances to replace mature trees in the upperstorey of managed forests. On the other hand, despite significant shares of Norway spruce in the upperstorey of all studied forests, regeneration and poles of this species were given low chances to replace mature trees in both old-growth and managed forests in the near future.

Coarse woody debris (mCoarse woody debris (m33/ha)/ha)

Independent t-tests indicated significant differences in CWD amounts between pairs: OGF Janj vs. MF Janj (p = 0.0000) and OGF Lom vs. MF Lom (p = 0.0000), whereas no significant difference was found between two OG forests nor between two managed forests.

Statistics OGF Janj MF Janj OGF Lom MF Lom

Mean 386.5 75.4 327.3 62.6

Std. Error of Mean 26.8 6.8 20.3 4.2

Median 389.0 69.3 316.2 56.2

Std. Deviation 169.7 52.4 128.2 32.9

Range 731 227 490 137

CONCLUSIONSCONCLUSIONS

1. Smallest dbh class contained more trees in managed forests, whereas in OG forests there was incomparably more large living trees and snags above 50 cm dbh. Nevertheless, dbh distributions in managed and OG forests Janj and Lom exhibited shapes considered to provide demographic equilibrium;

2. High growing stocks with dominant conifers in the upperstorey did not cause deviations from sustainable dbh distributions; however, the issue of regulating tree species composition came to surface.

3. Neither low-intensity disturbance in OG forests nor single tree selection system in managed forests led to sufficient recruitment of spruce in understorey and midllestorey despite significant presence of this species in the upperstorey.

4. Contrary to highly stocked OG forests, small canopy openings in managed forests made through single tree selection cutting provided small but constant share of maple.

5. If the share of spruce and maple is to be increased in these managed forests in the future, slight change in silvicultural approach may be necessary.

Thank you for your attention!