S

Es

Fa

b

a

ARRA

KDTSCF

1

udtsalSccl

h0

Ecological Engineering 71 (2014) 375–379

Contents lists available at ScienceDirect

Ecological Engineering

jou rn al hom ep age: www.elsev ier .com/ locate /eco leng

hort communication

ffects of human disturbances on Korean pine coverage and agetructure at a landscape scale in Northeast China

uqiang Zhaoa, Hongshi Hea,b, Limin Daia, Jian Yanga,∗

State Key Laboratory of Forest and Soil Ecology, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang 110164, PR ChinaSchool of Natural Resources, University of Missouri-Columbia, 203m Anheuser-Busch Natural Resources Building, Columbia, MO 65211-7270, USA

r t i c l e i n f o

rticle history:eceived 4 September 2013eceived in revised form 15 June 2014ccepted 20 July 2014

eywords:isturbanceimber harvesteed harvesthangbai Mountainorest management

a b s t r a c t

Korean pine (Pinus koraiensis) is an important species of forests in Northeast China and the RussianFar East. In recent years, human disturbances have increasingly threatened the long-term populationsustainability of Korean pine in Northeast China. Though the Chinese government has implementedvarious programs to restore and protect Korean pine, results have not been satisfactory. Economicincome often conflicts with ecosystem function, so figuring out ways to balance ecologic and eco-nomic values is important for Korean pine sustainability and management. The effects of humanson Korean pine forest are complex and interact in ways that are difficult to predict at the land-scape scale. In this paper, we used a simulation approach to examine multiple intensities of timberharvest and seed harvest on landscape-scale Korean pine coverage and age structure. We found thatthe influences of seed harvest became increasingly important over the simulation period. Seed harvestnoticeably influenced the percent coverage of young (<60 yr) Korean pine. Young age classes of Koreanpine disappeared when seed harvest intensity was high, even with a low timber harvest intensity. Exceptduring high timber harvest intensities, Korean pine maintained its coverage and age structure to a certainextent under a medium seed harvest level. When seed harvest was prohibited, Korean pine coverage

remained steady compared to initial conditions, regardless of the timber harvest intensity. Percent cov-erage of young and middle age (60–120 yr) Korean pine increased late in the simulation period. Our studyshowed that some level of seed harvest intensity combined with medium and low timber harvest inten-sities may be optimal for managers to maintain a viable economy together with ecosystem sustainability.

© 2014 Elsevier B.V. All rights reserved.

set

fCerig

. Introduction

Humans are an important component of both managed andnmanaged forests. With continued population growth, humanisturbance has intensified and is likely altering ecosystem struc-ure and function (DeFries et al., 2004; Yu et al., 2011). Activitiesuch as the over-harvesting of trees and cones can dramaticallylter interspecific interactions and feedback processes, which mayead to a decline in ecological communities (Finnegan et al., 2014;ummerville and Crist, 2002; Piao et al., 2011). Individual and

ombined effects of human disturbances on forest dynamics areomplex and interact in ways that are difficult to predict at theandscape scale (Gustafson et al., 2010). In designing a scientifically

∗ Corresponding author. Tel.: +86 24 80970331.E-mail address: yangjian@iae.ac.cn (J. Yang).

cfb2p

e

ttp://dx.doi.org/10.1016/j.ecoleng.2014.07.072925-8574/© 2014 Elsevier B.V. All rights reserved.

ound management plan, evaluating some of the more disruptiveffects of humans on forest ecosystems from a landscape perspec-ive is worthwhile.

Korean pine broadleaf mixed forest is an important systemor maintaining biodiversity and ecosystem function in Northeasthina and the Russian Far East. Major human disturbances of thiscosystem include seed harvesting and timber harvesting. In thisegion forests are harvested primarily by clear-cutting or highntensity selective cutting (Dai et al., 2009). Due to excessive log-ing, about 85% of the mature and old-growth forests have beenonverted into different types of secondary forests and plantedorests (Meng et al., 2008). The age structure of the forests hasecome younger and simpler (Shao et al., 2001; Zhao and Shao,

002). This poses an immediate need for ecological restoration torotect and improve forest health.

Since 2000, the Chinese government has pursued a new for-st management policy, where various harvest intensities have

3 ngine

roveasthfaenf

hsdiitufsdf

2

2

o(1mp2iksMmattata

2

vlihtctpsdhd

LC

tdo2adqwyt

bltcpsb6

2

tuahasnwaa(

3

3K

tuipoeToec

Khbsh

76 F. Zhao et al. / Ecological E

eplaced a monoculture-oriented high intensity harvest regime. Inrder to compensate for a decreasing economic value of timber har-est, Korean pine seed was intensively harvested due to its highconomic value. Korean pine (Pinus koraiensis) is an ecologicallynd economically important tree species for this system. Becauseeed source is one of the most important factors that can poten-ially affect successional dynamics (He and Mladenoff, 1999), thearvesting of Korean pine cones is the most significant obstacle

or restoration of this species (Liu et al., 2004, 2005). To restorend protect Korean pine, evaluating the tradeoffs between differ-nt levels of seed harvest intensity and timber harvest intensity areeeded to help determine the most sustainable management plan

or this ecosystem.In this study we examined the effects of timber and seed

arvesting on Korean pine coverage and age structure from a land-cape perspective. Specifically, we examined: (1) effects of theseisturbances on maintaining Korean pine coverage; (2) relative

nfluences of varying levels of timber and seed harvest, and theirnteraction, on Korean pine coverage; and (3) effects of these dis-urbances on Korean pine age structure. We conducted our studysing a simulation approach which considered two experimentalactors, timber harvest intensity (3 levels) and seed harvest inten-ity (3 levels). Percent cover of Korean pine, and area occupied byifferent age cohorts of Korean pine, served as response variablesor evaluating structural changes in the Korean pine forest.

. Materials and methods

.1. Study area

Our study site was located on the northwest-facing slopef the Changbai Mountains in Lushui River Forestry Bureau42◦20′–42◦40′N, 127◦29′–128◦02′E) and covered about30,000 ha. The forest is characterized by a continental cli-ate, with long cold winters and short hot summers. Mean annual

recipitation and temperature are approximately 894 mm and.9 ◦C, respectively. Major coniferous tree species in this area

nclude Korean pine (P. koraiensis) and Korean spruce (Piceaoraiensis). Broadleaf species include basswood (Tilia amuren-is), maple (Acer mono), Manchurian ash (Fraxinus mandshurica),anchurian walnut (Juglans mandshurica), Mongolian oak (Quercusongolica), white birch (Betula platyphylla), elm (Ulmus pumila),

nd aspen (Populus davidiana). Korean pine is the climax species inhis ecosystem. The forest land in our study area was divided intowo major management areas: a commercial harvest area (80%)nd a conservation area (20%). In the commercial harvest area,imber and seed harvesting were allowed. In the conservationrea, only seed harvesting was allowed.

.2. Experimental design

We designed a factorial experiment that included three har-est intensities and three seed harvest levels. Tree seed harvestevels were based on a local government rule, issued in 2003, defin-ng three conditions under which Korean pine seed can legally bearvested: (a) high seed harvest intensity—unlimited seed collec-ion up to 100%; (b) low seed harvest intensity—about 30% seedollection; and (c) no seed harvest—no seed collection permit-ed. Timber harvest levels were based on the current managementlan of the Lushui River Bureau. The three intensities used in our

tudy included: (a) low harvest intensity—cutting limit set at 3% perecade; (b) medium harvest intensity—set at 5% per decade; and (c)igh harvest intensity—set at 10% per decade. This work was con-ucted based on Forestry Standards “Observation Methodology for

iwht

ering 71 (2014) 375–379

ong-term Forest Ecosystem Research” of the People’s Republic ofhina.

We used LANDIS, a spatially explicit forest landscape model, ashe simulation tool for our study. Detailed information on modelesign, testing and model parameters of our study area may bebtained from other sources (He and Mladenoff, 1999; Zhao et al.,011). Based on our experimental design, nine scenarios were cre-ted using LANDIS to simulate forest succession under varyingisturbance regimes. We determined 100 years to be an ade-uate timeline for forest management because policy can changeith the changing of forest structure. Thus, the model ran for 100

ears, beginning in 2003. Each scenario was replicated five timeso account for stochasticity in the LANDIS model.

While recognizing that ecological effects of seed harvest, tim-er harvest, and their interaction are complex and interrelated, we

imited the focus of our analysis to species coverage and age struc-ure, which are key parameters of the forest landscape. Speciesoverage was expressed as the proportion of pixels indicating theresence of a species relative to that of the entire landscape. Agetructure was expressed as the proportion of landscape dominatedy different age classes of Korean pine (young: ≤60; middle age:1–120; mature: >120).

.3. Data analysis

Simulation results were analyzed using univariate analyses inhe General Linear Model (SPSS 16.0) for early, middle and late sim-lation periods. Type III sums of squares derived from univariatenalyses were used to quantify the unique contributions of timberarvest, seed harvest, and their interaction on coverage of youngge Korean pine at a landscape scale. The actual type III sums ofquare values were comparable within one statistical model, butot necessarily between two or more statistical models. Therefore,e used proportions of actual type III values to compare differences

mong the relative contribution of timber harvest, seed harvestnd their interaction for early (0–30 yr), middle (40–70 yr) and late80–100 yr) periods of a given simulation.

. Results

.1. Effects of multiple intensities of timber and seed harvest onorean pine coverage

Our results showed that timber harvest was more importanthan seed harvest for coverage of Korean pine in the early sim-lation period (Fig. 1). As the simulation periods progressed, the

nfluences of seed harvest increased. By the middle simulationeriod, the relative importance of seed harvest and timber harvestn Korean pine coverage was similar. By the late simulation period,ffects of seed harvest were far greater than those of timber harvest.hese results indicate that seed harvest has a long-term influencen Korean pine coverage that requires multiple decades to becomevident, but can exceed the influence of timber harvest within oneentury.

Our results also revealed differences in percent coverage oforean pine under the varying levels of seed harvest and timberarvest intensity. Coverage was highest with low and medium tim-er harvest and no seed harvest (Fig. 2). Coverage was lowest wheneed harvest intensity was high, especially when combined withigh timber harvest intensity (Fig. 2a). Under low seed harvest

ntensity, Korean pine maintained its coverage to a certain extent,hen timber harvest was low and medium (Fig. 2b). With no seedarvest, Korean pine coverage remained steady compared to ini-ial conditions, regardless of timber harvest intensity, and even

F. Zhao et al. / Ecological Engineering 71 (2014) 375–379 377

Ft

i(

3s

tKao

lAmawo

aKtittllhs

4

4

NdCttob

Ft

otHaltoiowi

ig. 1. Relative influences of timber harvest, seed harvest, and their interaction onhe percent coverage of Korean pine as measured by type III sums of squares.

ncreased a bit with low and medium timber harvest intensitiesFig. 2c).

.2. Effects of seed harvest and timber harvest on Korean pine agetructure

Because a simulation period of 100 years would cover lesshan half the longevity of the Korean pine, simulated regeneratingorean pines did not have the opportunity to achieve the maturege class in our model. Therefore, we did not analyze the coveragef mature age Korean pine.

For young Korean pine, a high timber harvest intensity generallyed to a high percent coverage which peaked at 40 years (Fig. 3).t the high seed harvest intensity, though, percent coverage plum-eted to near zero after 60 simulation years, suggesting that young

ge Korean pine may completely disappear after 60 years. Trendsere similar, but not as extreme, when seed harvest was prohibited

r at low intensity.For middle age Korean pine, a sharp shift in percent coverage

ppeared after 40 years (Fig. 4), reflecting movement of youngorean pine into the middle age group. After 40 simulation years,

he variation of middle age Korean pine percent coverage was sim-lar to that for the young age classes. Percent coverage continuedo decrease under high seed harvest intensity, especially with highimber harvest intensity. When seed harvest was prohibited or atow intensity, percent coverage steadily increased after 40 simu-ation years, regardless of timber harvest intensity. Coverage wasighest during the late simulation period under scenarios whereeed harvest was prohibited.

. Discussion

.1. Management implications

Korean pine plays a critical role in biodiversity in forests ofortheast China and the Russian Far East. In recent years, humanisturbances have increasingly altered forest structure in Northeasthina (Dai et al., 2004). Opportunities exist for optimizing ecosys-

em for Korean pine broadleaf mixed forest ecosystem restorationhrough application of ecology engineering, since the goals of ecol-gy engineering include the restoration of ecosystems that haveeen substantially disturbed by human activities (Mitsch, 2012). In

cath

ig. 2. Percent coverage of Korean pine for a 100 yr period for combinations of threeimber harvest intensities under varying seed harvest intensities.

ur study area, current forest management is focused on protec-ing existing Korean pine by decreasing timber harvest intensity.owever, timber production drives the economy in this region,nd reducing it has led to a significant reduction of income forocal residents. Seed harvesting was permitted to compensate forhe reduced incomes. Unfortunately, Korean pine seeds have beenverharvested because of their high economic value. Our resultsndicate that seed harvesting strongly influences percent coveragef young age classes of Korean pine, which may disappear entirelyhen seed harvest intensity is high, even with low timber harvest

ntensity. This could further influence subsequent successional pro-esses of Korean pine. Total and middle age Korean pine coverage

lso showed this tendency. Korean pine has long been the dominantree species in the Korean pine broadleaf mixed forest, providingabitat for a variety of plant and animal species. Decreased Korean

378 F. Zhao et al. / Ecological Engineering 71 (2014) 375–379

Fv

pe

prosdptpl

iehcw

Fs

chiwtmbcne

Ktihs

ig. 3. Percent coverage of young age classes of Korean pine for a 100 yr period witharying seed harvest intensities for three timber harvest intensities.

ine coverage may alter its role in the ecosystem and threatencosystem sustainability in the future.

Forest managers can maintain Korean pine coverage throughlanting after seed and timber harvesting occurs. Research in thisegion has shown that planting can increase Korean pine coveragever short time periods, especially with high timber harvest inten-ity (Zhao et al., 2013). However, research also shows that plantingoes not necessarily lead to a sustainable forest for long timeeriods. Furthermore, the cost of planting is high, and yields some-imes do not meet expectations. Preserving Korean pine forests vialanting may not prove sustainable, because economic rewards will

ikely be offset by the high cost of planting.Our simulation results suggest that Korean pine can maintain

ts coverage and age structure under a low seed harvest intensity,

xcept where timber intensity is high. When seed harvest was pro-ibited, Korean pine coverage was steady compared with initialonditions during all simulation periods, even when timber harvestas high. Percent coverage of young and middle age Korean pine

rbKK

ig. 4. Percent coverage of middle age Korean pine for a 100 yr period with varyingeed harvest intensities for three timber harvest intensities.

overage increased during the late simulation period. Althoughigh seed harvest intensity combined with high timber harvest

ntensity would likely increase economic income, this combinationould likely threaten ecosystem sustainability. There appear to be

wo best-case scenarios for providing income at the same time asaintaining ecosystem function: a prohibited seed harvest com-

ined with a high timber harvest; and a low intensity seed harvestombined with medium or low timber harvests. Forest managerseed to be provided the flexibility to balance economic income andcosystem service function.

Our results also found that the influences of seed harvest onorean pine coverage become more important later in the simula-

ion period. These results are consistent with the research of othersn this region. He et al. (2005) found that impacts of Korean pinearvesting may produce long-term alterations at the landscapecale, which persist for decades to centuries. Although economicewards may decline when seed harvest intensity decreases, it may

e offset by an increased timber harvest intensity. In recent years,orean pine seed has become increasingly expensive. Becauseorean pine seed can be harvested within a short rotation period,

ngine

teohmrat

4

BlhlsTtWuuhaorOvwAragt

A

Ww

CCP

R

D

D

D

F

G

H

H

L

L

M

MP

S

S

Y

Z

F. Zhao et al. / Ecological E

he economic value of moderate seed harvest intensity may soonxceed that of wood production. This indicates that a certain degreef seed harvest intensity combined with medium to low timberarvest intensity may be an appropriate management practice foraintaining economic income and ecosystem sustainability. These

esults provide a science-based guideline for effective forest man-gement that has implications for similar ecosystems throughouthe world.

.2. Model limitations

Our simulation results may be affected by limitations of LANDIS.ecause LANDIS does not track amounts of seed (and seed harvest

evels are characterized by strength of seed source in this paper), wead to set the three seed harvest levels (no, low, and high harvest

evels). Also, we had to translate the effects of source strength oneed harvest into seed dispersal distance parameters for LANDIS.he probability of Korean pine seeds reaching a distant destina-ion site generally increases with increased seed source strength.

e assumed a maximum dispersal distance for Korean pine seednder the no seed harvest scenario, an effective dispersal distancender the low seed harvest level, and no seed dispersal under theigh seed harvest level. As LANDIS only tracks the presence orbsence of tree species within a pixel. Our analysis focused onlyn the coverage of Korean pine, the translation method we choseeflects the effects of seed harvest level on coverage of Korean pine.ur research also did not consider climate change. However, theariation tendency between levels of timber and seed harvestingould likely not have changed significantly with climate change.lthough the methods in this study are somewhat theoretical, ouresults provide pertinent evidence of the effects of human disturb-nce on Korean pine regeneration at a landscape scale, as well asuidance on how to conduct science-based forest management inhis ecosystem.

cknowledgements

The authors are thankful to Dr. Bernard Lewis and Mrs. Jennyeisberg for editing a preliminary version of this paper. This paperas supported by the National Key Technologies R&D Program of

Z

Z

ering 71 (2014) 375–379 379

hina “2012BAD22B04”, National Natural Science Foundation ofhina (31300526), and CFERN & GENE Award Funds on Ecologicalaper.

eferences

ai, L.M., Gu, H.Y., Shao, G.F., Wang, Q.L., 2004. The Korean pine broadleaf mixedforest on Changbai Mountain of China. Science and Technology Press, Liaoning.

ai, L.M., Zhao, F.Q., Shao, G.F., Zhou, L., Tang, L.N., 2009. China’s classification-basedforest management: procedures, problems, and prospects. J. Environ. Manage.43 (6), 1162–1173.

eFries, R.S., Foley, J.A., Asner, G.P., 2004. Land-use choices: balancing human needsand ecosystem function. Front. Ecol. Environ. 2 (5), 249–257.

innegan, J., Regan, J.T., O’Connor, M., Wilson, P., Healy, M.G., 2014. Implications ofapplied best management practice for peatland forest harvesting. Ecol. Eng. 63,12–26.

ustafson, E.J., Shvidenko, A.Z., Sturtevant, B.R., Scheller, R.M., 2010. Predictingglobal change effects on forest biomass and composition in south-centralSiberia. Ecol. Appl. 20 (3), 700–715.

e, H.S., Hao, Z.Q., Mladenoff, D.J., Shao, G.F., Hu, Y.M., Chang, Y., 2005. Simulatingforest ecosystem response to climate warming incorporating spatial effects innorth-eastern China. J. Biogeogr. 32, 2043–2056.

e, H.S., Mladenoff, D.J., 1999. The effects of seed dispersal on the simulation oflong-term forest landscape change. Ecosystems 2 (4), 308–319.

iu, Z.G., Ji, L.Z., Hao, Z.Q., Zhu, J.J., Kang, H.Z., 2004. Effects of cone-picking on naturalregeneration of Korean pine in Changbai Mountain nature reserve. Chin. J. Appl.Ecol. 15 (6), 958–962.

iu, Z.G., Ji, L.Z., Zhu, J.J., 2005. Impact of cone-picking on seed banks and animals. J.Grad. Sch. Chin. Acad. Sci. 22 (5), 596–603.

eng, T., Wang, M.C., Liang, L.H., He, Y., 2008. Dynamic analysis on ecological land-scape pattern of Changbai Mountain. Global Geol. 27 (3), 338–344.

itsch, W.J., 2012. What is ecological engineering? Ecol. Eng. 45, 5–12.iao, Z.J., Tang, L.N., Swihart, R.K., Wang, S.X., 2011. Human–wildlife competition for

Korean pine seeds: vertebrate responses and implications for mixed forests onChangbai Mountain, China. Ann. For. Sci. 68, 911–919.

hao, G.F., Zhang, P.C., Bai, G.X., Zhao, G., Wang, Z., 2001. Ecological classificationsystem for China’s natural forests: protection and management. Acta Ecol. Sin.21, 1564–1568.

ummerville, K.S., Crist, T.O., 2002. Effects of timber harvest on forest lepidoptera:community, guild, and species responses. Ecol. Appl. 12, 820–835.

u, D.Y., Shi, P.J., Han, G.Y., Zhu, W.Q., Du, S.Q., Xun, B., 2011. Forest ecosystemrestoration due to a national conservation plan in China. Ecol. Eng. 37 (9),1387–1397.

hao, F.Q., Yang, J., He, H.S., Dai, L.M., 2013. Effects of natural and human-assistedregeneration on landscape dynamics in a Korean pine forest in Northeast China.PLoS One 8 (12), e82414.

hao, F.Q., Yang, J., Liu, Z.H., Dai, L.M., He, H.S., 2011. Balancing multiple objectivesusing a classification-based forest management system in Changbai Mountains,China. Environ. Manage. 48, 1136–1147.

hao, G., Shao, G.F., 2002. Logging restrictions in China: a turning point for forestsustainability. J. For. 100 (4), 34–37.