1CONSTRAINTS

2LOGGED

ON THE HARVEST OF LINE-PLANTED TIMBER TREES IN

AND ENRICHED DIPTEROCARP FOREST IN KALIMANTAN,

3INDONESIA
4

K Niti Putro1, F Yusro2, Ruslandi3* G Hardiansyah2 & FE Putz3

5R
6

71PT
8

Sari Bumi Kusuma, Seruyan and Katingan, Central Kalimantan, Indonesia

Faculty of Forestry, Tanjungpura University, Pontianak, West Kalimantan,

9Indonesia
103Department

of Biology, University of Florida, Gainesville, FL 32611 USA

11
12*

r.ruslandi@ufl.edu

13NITI

PUTRO R K, YUSRO F, RUSLANDI, HARDIANSYAH G & PUTZ FE.

142015.

Constraints on the harvest of line-planted timber trees in

15logged
16Where

and enriched dipterocarp forest in Kalimantan, Indonesia.

enrichment planting of timber trees results in high densities of

17planted

and naturally regenerated trees larger than a minimum cutting

18diameter
19residual

(e.g., 40 cm DBH), harvest of all those trees would result in major

stand damage. One proposed harvest strategy for the enriched

20dipterocarp

forest we studied is to cut only the planted trees and to fell them

21down

the 3 m wide planting lines so as to protect the intervening 17 m wide

22strips

of regenerating natural forest. Based on crown and bole characteristics

23of

1
2

line-planted dipterocarp trees, felling down the planting line should be

1

1possible
2trees
3at

for 90% of the trees. When felled, the other 10% of the planted

would cross at least one strip of natural forest. A bigger concern is that

the time of the planned harvest (25 years after planting), 31% of the

4planted
5stand

trees will not quite have reached harvestable size. To avoid excessive

damage and destruction of future crop trees, alternative harvest

6strategies

are needed. For example, enriched stands could be harvested in

7two

phases with a modified shelterwood system. Alternatives to chainsaws

8and

crawler tractors should also be considered such as feller bunchers or

9mobile

crane cable yarders.

10
11Keywords:

silviculture intensification, enrichment planting, selective logging

12INTRODUCTION
13Enrichment

planting along cleared lines is a frequently invoked silvicultural

14intervention
15forests
16Safa

for sustaining timber yields from selectively logged tropical

(e.g., Adjers et al. 1995, Montagnini et al. 1997, Paquette et al. 2009,

et al. 2004, Soekotjo 2009). Although many millions of dollars have

17been

wasted on planting seedlings that subsequently died (F E Putz, personal

18observation),
19Dawkins

in the 1960s are followed (Dawkins 1961). Basically, where healthy

20seedlings
21with

are carefully planted under canopy openings that provide them

sufficient light and those seedlings are liberated from vines and

22encroaching
23rates

1
2

enrichment planting works well if the rules spelled out by C.

understory vegetation for several years, survival and growth

can be substantial. This finding is supported by plentiful data from

2

1experimental
2silviculture
3the

plots (e.g., Adjers et al. 1995) and in areas where intensive

is heavily subsidized by external agencies (i.e., not paid for by

property owner or concessionaire). Recent evidence from a timber

4concession

in Indonesia indicates that under the appropriate economic and

5political

conditions, self-funded and large-scale enrichment planting along

6cleared

lines following Dawkins Rules can also be successful in terms of

7seedling
8we

growth and survival (e.g., Soekotjo 2009, Hardiansyah 2011). Here

consider some harvesting options and impacts when these successfully

9enriched

stands mature and are harvested.

Prior to the harvest of the planted stands, the biodiversity impacts of

10

11enrichment
12supported

planting along cleared lines are small. This claim is well-

by a study in stands established in Sabah, Malaysia by Innoprise

13Corporation
14of

the Netherlands. Ansell et al. (2011) found that stand rehabilitation

15through

enrichment planting was associated with little change in bird species

16composition
17harvests

1914-18

felling diameter? With heights of 30-40 m and crown diameters of

m, the harvest of >40 dipterocarp trees per hectare would result in

20substantial
21with

devastation, at least using currently accepted approaches to RIL

ground-based skidding.
Due to lack of markets for most trees in species-rich tropical forests,

23timber

1
2

relative to primary forest . But what will be the consequences of

of heavily enriched stands where harvests are constrained only by a

18minimum

22

with the financial support of the FACE (Free Air CO2 Enrichment)

harvests are typically selective (e.g., Bertault and Sist 1997).

3

1Numerous
2soils

by selective logging can be substantially mitigated if reduced-impact

3logging
4has

studies have demonstrated that damage to residual stands and

(RIL) techniques are employed (reviewed by Putz et al. 2008). But RIL

its limits; based on the results of an experimental logging study in East

5Kalimantan,
6RIL

Sist et al. (1998, 2003) found that the environmental benefits of

are lost if >8 trees or 60 m3 are harvested per hectare. Based on simple

7linear

projections of the first 14-15 years of growth of planted S. leprosula

8trees

in our study site in Central Kalimantan, legally harvestable trees (i.e.,

9>40

cm DBH) will greatly exceed both those limits; harvestable volumes of

1050-60

trees are predicted to exceed 200 m3 per hectare (Kusuma et al.

112014).

One frequently-discussed option for harvesting enriched stands in

12

13Indonesia
14natural
15this

is to harvest only the trees along the planted lines and retain the

forest retained between the lines. Here we evaluate the viability of

option with data on enrichment planted Shorea leprosula trees in twice-

16logged
17likely

lowland dipterocarp forest in Kalimantan, Indonesia. To evaluate the

impacts of this approach, we scored the planted trees on the basis of

18whether
19line.

a trained feller with a chainsaw could fell them down the planted

We supplemented the expert opinion of an experienced forest

20concession

staff member whose duties include tree marking for directional

21felling

with measurements of crown eccentricities and stem lean. We discuss

22these

findings in light of the expected number of planted trees that will reach

23the

1
2

40 cm DBH threshold by the end of the 25-year rotation.

4

1
2METHODS
3Study
4The

site

study was conducted in the Sari Bumi Kusuma (SBK) concession in

5Central
6with

Kalimantan, Indonesia (0o56N, 111o68E) at elevations of 356-425 m

undulating topography. Precipitation for 2001 2012 averaged 3631

7mm/year

(3024 4762 mm/year), no months averaged < 200 mm, and

8temperatures
9(Suryatmojo
10internal

et al. 2013). Soils are deep red-yellow clay loams with good

drainage. During the 1999-2014 period, SBK applied the Indonesian

11selective
12Pilih

averaged 30 33oC at midday and 22-28oC at night

cutting and strip enrichment planting silvicultural system (Tebang

Tanam Jalur TPTJ) to 49,000 ha in its 147,600 ha concession. The

13minimum

cutting cycle under TPTJ is 25 years; the oldest plantations will

14therefore

be eligible for harvest of trees >40 cm DBH 10 years after we

15conducted
16year

this study. TPTJ guidelines continue to evolve, but in the 14-15

old plantations we sampled, nursery-grown seedlings of Shorea

17leprosula
18along

and four other congeneric species were planted at 5 m intervals

parallel 3 m-wide lines cleared at 20 m intervals through twice-logged

19forest.

We studied S. leprosula because it was the most commonly planted

20species.
21Field
22We
2314

1
2

data collection

measured the DBHs of 500 planted S . leprosula planted trees that were
15 years old; groups of 5 trees were selected for measurement in a
5

1stratified
2DBH

random fashion. From each group of five, the tree with the largest

was selected for more detailed assessment (N =100); sampled trees

3ranged

19 38 cm DBH. To estimate whether a trained feller could fell each

the selected trees within 5o of the planting line in both directions, the

4of

5range
6on

of possible felling directions for each of these 100 trees was assessed

the basis of the expert judgment of a forest worker whose regular job is to

7mark

trees for directional felling. Then, for each of the selected trees we

8measured
9line

5 m from the trunk, crown radii in four directions (perpendicular to and

10along
11the
12

the terrain slope perpendicular to and parallel with the planting

the planting line), and the degree of the trunk lean perpendicular to

planting line.

Data analysis

13A

binomial code was used to describe whether the felling expert thought

14each
15was

tree could be felled within 5o of the planting line. Crown eccentricity

calculated as the ratio of longest crown radii perpendicular to and

16parallel
17As

with the planting line. The same method was used for the slope ratio.

the response variable was binomial, generalized linear model (GLM)

18analysis

using the binomial family was used to assess the effects of DBH,

19slope

ratio, crown ratio, and tree lean on the probability of the tree to be

20felled

down the planting line. Statistical analysis was conducted in R (R core

21Development
22
23the

1
2

Team 2013).

To estimate how many of the line-planted trees will be harvestable at

end of the planned 25 year rotation, we used the average rates of DBH
6

1increment
215

up to the age of the trees at the time of our measurement (14 or

years since planting) to extrapolate to 25 years, the minimum rotation

3duration

stipulated by governmental regulations (MoF 2009). This simple

4approach

is fraught with assumptions, but does reflect the typically strong

5temporal

autocorrelation in growth rates of moderately light-demanding

6tropical

trees like our study species (Brienen et al. 2006).

7RESULTS
8Based
9for

AND DISCUSSION

on the expert opinion of a forester whose normal job is to mark trees

direction felling as part of pre-harvest planning, at age 14-15 years and

10DBHs

of 19-38 cm, 90 of the 100 planted trees sampled could be

11directionally

felled down the planting line. The probability of a tree being

12classified

as fellable in either direction down the planting line decreased with

13increased

DBH (P=0.860), slope ratio (P=0.319), crown eccentricity ratio

14(P=0.666),
15expected
16ratio,

success of directional felling decreased with increasing DBH, slope

and crown ratio, tree lean was the only predictor that decreased the

17probability
18observed

20crown

values (Figure 2).

diameter perpendicular to the planting line, 26 displayed the opposite

21condition,

and the remainder had symmetrical crowns. Open spaces lateral

the planting lines that were mostly down slope, provided opportunities for

23crown

1
2

of felling down the planting line to <0.5 across the range of

Of the 100 planted trees sampled, the crowns of 58 were largest in

19

22to

and stem lean (P<0.001; Figure 1). Graphically, while the

expansion; along the planted lines, in contrast, the nearest neighbor

7

1was

only 5 m away. These findings also suggest that 5 m inter-tree spaces

2along
3the

planting lines did not allow full crown formation by more than half of

planted trees. When considering the crown space available to trees

4planted

at such close spacing, it is important to recognize that the actual

5available

space is further constrained because dipterocarp crowns do not

6interdigitate
7referred
8of

to as crown shyness gaps (Ng 1977). Although at the current age

14-15 years, DBH is not strongly related to crown eccentricity (Figure 3),

9given
10a

but instead are surrounded by bare areas of about 0.5 m

that the mean crown diameter of a 40 cm DBH Shorea leprosula tree in

dominant canopy position in natural forest is about 16 m (T. Inada,

11unpublished
1210

years before the planned harvest.

Although our data on 14-15 year old line-planted S. leprosula trees

13

14suggest
15to

data), it is likely that crown eccentricity will increase in the final

that the measured crown characteristics do not affect the capacity

directionally fell trees down planted lines, thinning might both promote

16growth

and provide space for the development of more circular crowns.

17Thinning
18starting
19our

of planted trees is listed as a required treatment in TPTJ stands

5-10 years after planting, but has not been carried out in SBK or, to

knowledge, in any other concession. Insofar as thinning would reduce

20crown

competition and might increase growth of the retained trees, it should

21at

least be tested experimentally. Alternatively, given the high survival rates

22of

the well-tended planted trees, perhaps trees should be planted at intervals

23>5

1
2

m.
8

Although at age 15 years and DBHs that averaged 24.2 cm 91% of

1
2the

planted trees were judged to be fellable down the planting line, it is not

3clear
4rate
5to

that this would be the optimal direction. In particular, if the average

of DBH increment of each planted tree over the last 14-15 years is used

predict its DBH at age 25, 31% of the planted trees will not have reached

6harvestable
7attain

a DBH of >40 cm would therefore almost unavoidably result in the

8damage
920-40
10lines.
11and
12in

diameter by harvest time (Figure 4). Felling all the trees that

or death of 24 future crop trees (i.e., trees of commercial species of

cm DBH) per hectare if the planted trees are felled down the planting
With an average cost of TPTJ over the first 3 years of about US$700/ha

an average of 80 trees planted/ha (Hardiansyah 2011), that means that

each planted tree the concession invested about $10. If all 24 future crop

13trees

are destroyed during the harvest, that represents a loss of $240 per

14hectare

to which foregone future revenues from the growth and subsequent

15harvest

of these trees should be added. These economic estimates are only

16approximate,
17trees

but make clear that alternatives to clearcutting the line-planted

should be sought. At the very least, efforts should be made to avoid

18felling

or damaging trees that have not yet passed their peak mean annual

19increments.

To avoid excessive stand damage and economic losses, we

20recommend

that there be limits on the number of trees harvested per

21hectare

at any one entry. Experiments with shelterwood cuts would seem

22warranted
23planted

1
2

given that even at 15 years, we observed that some of the

trees were reproductive; natural regeneration of commercial species

9

1is

likely to be abundant when the stands are harvested. Also, alternatives to

2the

current practice of timber yarding with crawler tractors should also be

3considered

including long-line cable yarding with mobile towers. Most

4fundamentally,
5management
6regeneration
7mixtures
8wider

to avoid the dilemma of over-stocking, we suggest that

guidelines be modified to benefit more from natural

and to anticipate the constraints on harvesting by planting

of species that grow at different rates and provide timbers of a

range of values.

9ACKNOWLEDGEMENTS
10Financial

support for this research was provided by the Academy of

11Distinguished

Teaching Scholars at the University of Florida. We also

12acknowledge

PT Sari Bumi Kusuma for hosting the study team and providing

13logistical

support.

14REFERENCES
15ADJERS

G, HADENGGANAN S, KUUSIPALO J, NURYANTO K & VESA L. 1995. Enrichment

16

planting of dipterocarps in logged-over secondary forests: effect of

17

width, direction and maintenance method of planting line on selected

18

Shorea species. Forest Ecology and Management 73: 259-270.

19ANSELL

FA, EDWARDS DP & HAMER KC. 2011. Rehabilitation of logged

20

rainforests: avifaunal composition, habitat structure, and implications

21

for biodiversity-friendly REDD+. Biotropica 43:504-511.

22BERTAULT
23

1
2

JG & SIST P. 1997. An experimental comparison of different

harvesting intensities with reduced-impact and conventional logging in

10

East Kalimantan,Indonesia. Forest Ecology and Management 94: 209

218.

3BRIENEN

RJW, ZUIDEMA PA & DURING HJ. 2006. Autocorrelated growth of tropical

forest trees: unraveling patterns and quantifying consequences. Forest

Ecology and Management 237: 179-190.

6DAWKINS
7

HC. 1961. New methods of improving stand composition in tropical

forests. Caribbean Forester: 12-20.

8HARDIANSYAH
9

G. 2011. Potensi pemanfaatan sistem TPTII untuk mendukung

upaya penurunan emisi dari deforestasi dan degradasi hutan (Studi

10

Kasus Areal IUPHHK PT.Sari Bumi Kusuma di Kalimantan Tengah).

11

Disertasi program Doktor. Institut Pertanian Bogor, Indonesia.

12KUSUMA

AP, DAMAYATI A, AMIJAYA AM, HARDIANSYAH G, CHOZIN M, SUPARNA N &

13

SARIBANON N. 2014. Road Map: Implemantasi silvikultur intensif pada

14

hutan alam producksi Indonesia 2045. Fakultas Kehutanana Universitas

15

Tangungpura Pontianak Press, Pontianak, Indonesia.

16MINISTRY OF

FORESTRY (MOF). 2009. MoF Decree P.11/Menhut-II/2009 concerning

17

silviculture system to be implemented in forest concessions. Jakarta,

18

Indonesia.

19MONTAGNINI

F, EIBL B, GRANCE L, MAIOCCO D & NOZZI D. 1997. Enrichment

20

planting in overexploited subtropical forests of the Paranaense region

21

of Misiones, Argentina. Forest Ecology and Management 99:237246.

22NG

1
2

FSP. 1977. Shyness in trees. Nature Malaysiana 2: 34-37.

11

1PAQUETTE

A, HAWRYSHYN J, VYTA SENIKAS A & POTVIN C . 2009. Enrichment

planting in secondary forests: a promising clean development

mechanism to increase terrestrial carbon sinks. Ecology and Society

14: 31. [online] URL:

http://www.ecologyandsociety.org/vol14/iss1/art31/

6PUTZ

FE, SIST P, FREDERICKSEN TS & DYKSTRA D. 2008. Reduced-impact logging:

challenges and opportunities. Forest Ecology and Management 256:

1427-1433.

9R

DEVELOPMENT CORE TEAM .2013. R: A language and environment for

10

statistical computing. R Foundation for Statistical Computing, Vienna,

11

Austria.

12SAFA,

SAMAUN M, ZAHARI I & ARIFIN A. 2004. Potentialities of new line planting

13

technique of enrichment planting in Peninsular Malaysia: A review of

14

resource sustainability and economic feasibility. MPRA Paper No.

15

10889, posted 03 October, 2008. http://mpra.ub.uni-

16

muenchen.de/10889/

17SCHIELZETH,
18
19SIST

H. 2010. Simple means to improve the interpretability of

regression coefficients. Methods in Ecology and Evolution 1: 103-113.

P, NOLAN T, BERTAULT JG & DYKSTRA DP. 1998. Harvesting intensity versus

20

sustainability in Indonesia. Forest Ecology and Management 108: 251

21

260.

22SIST
23

1
2

P, FIMBEL R, NASI R, SHEIL D & CHEVALLIER MH . 2003. Towards sustainable

management of mixed dipterocarp forests of South East Asia: moving
12

beyond minimum diameter cutting limits. Environmental Conservation

30: 364-374.

3SOEKOTJO.

2009. Intensive Silviculture to Improve Productive Capacity of

Forests: Large Scale Enrichment Planting of Dipterocarps. XIII World

Forestry Congress, Buenos Aires, Argentina, 18-23 October 2009.

6SURYATMOJO

H, FUJIMOTO M, YAMAKAWA Y, KOSUGI K & TAKAHISA M. 2013. Water

balance changes in the tropical rainforest with intensive forest

management system. International Journal of Sustainable Future for

Human Security J-SustaiN 1: 5662.

10
11
12

1
2

13

1
2Figure 1
Expected success of directional felling planted trees down the
3planting line as a function of DBH and measured variables that contrast tree
4characteristics perpendicular to and parallel with the planting line (stem
5lean, crown ratio, and slope ratio). Variable represents predictor variables in
6the regression model and value represents the predictors relative strengths
7of influence on the response/predicted variables. The center dots indicate
8predicted values, the thick horizontal bands indicate the 68% confidence
9interval, and the thinner, wider bands indicate the 95% confidence interval.
10A predictor is significant if its confidence interval does not contain 0. Thus,
11for example, DBH is not a significant predictor, Crown is a marginally12significant predictor, and Lean is a significant predictor. All predictors were
13scaled to a standard deviation of 1, and centered around 0, to improve
14interpretability of the predictors relative strength of influence on the
15predicted variable compared to the other predictors (Schielzeth, 2010). A
16negative estimate indicates that higher values of that predictor correspond
17to lower likelihood of the tree being successfully felled down the planting line
18in either direction.
19
20
21
22

1
2

14

1
2Figure 2
Predicted (blue line) and observed (red dots) probabilities of
3planted trees being felled down the planting line as function of lean, DBH,
4crown ratio and slope ratio. That probability decreased to <0.5 only for stem
5lean angle perpendicular to the planting line
6
7

8
9

1
2

15

1Figure 3
Ratio of the longest crown radii perpendicular to and parallel
2the planting line as a function of tree DBH. Crown ratio = 1.1963715 30.0008323*DBH (P=0.9383, R2 = 0.0006).
4
5
6

with

7
8Figure 4
Distribution of projected planted tree DBHs (N =500) at the end
9of the planned rotation (25 years) as predicted by a linear extrapolation of
10growth of each tree during the first 15 years after planting
11

1
2

16