RECOVERY OF LOWLAND TROPICAL FOREST OF EAST KALIMANTAN THIRTY YEARS AFTER MAJOR FIRE EPISODES

Subekti Rahayu, Meine van Noordwijk, Purwaningsih, Sonya Dewi and Degi Harja

52nd Annual Meeting ATBC 201512-16 July 2015

Honolulu, Hawaii

Symposium 2Integrated Landscape Management:

From Theory to Practice

BACKGROUND• Specific effects of disturbance remain visible in forest

structure and species composition long after a closed-canopy status is regained (Newbold et al. 2014; Winter 2012)

• Canopy structure and basal area recover in 56 years after selective logging (Priatna et al. 2004)

• Species richness recover in 150 years after clear felling (Riswan et al. 1984)

• Aboveground biomass recovery was estimated to take 80 years and biodiversity, assessed across plant and faunal groups may take 120-150 years (Martin et al. 2013)

• Published studies of long term forest recovery after fire is limited due to limited long term permanent plot observation established in tropical forest of Indonesia; most studies discussing impact of fire events on biodiversity (van Nieuwstadt 2001; Slik 2002; Eichhorn 2006)

• Established permanent plot of 10.5 hectares area in Samboja Research Forest, East Kalimantan by Indonesian Institute of Sciences (IIS) in 1979 – 1981 is good chance to learn the forest recovery process after fire

BACKGROUND

OBJECTIVES

• To understand the recovery process of species composition after repeated fire events

• To understand the biomass recovery due to repeated fire events

• To understand the implications on restoration

STUDY SITE• Samboja Research Forest (SRF)

established 1978 in 504 ha lowland dipterocarp forest and expanded with another 3000 ha of logged-over forest 3000 ha in 1991.

• Currently, 504 ha covered by forest (undisturbed and disturbed by fires) and 3000 ha of shrubs and grasslands.

• Since 2011, SRF is managed by BALITEK KSDA, FORDA, Ministry of Forestry, now Ministry of Environment and Forestry

Fire histories and Inventories• 10.5 hectares permanent plot in lowland mixed dipterocarp

of SRF established and observed by IIS 1979-1981 (row 1-70), published by Kartawinata et al. (2008)

• Row 1-63 affected by two major fire events in 1983 and 1998• Row 21-31 (1.65 ha) revisited by IIS, published by Simbolon

(2005)• Row 21-32 (1.8 ha) revisited in 2011 by IIS and ICRAF

Unburned forest

2011 observation

1979-1981 observation

1 70

Column: 700 m

Row: 150 m

21 32 63

METHODS• 2011 survey:– 10 m x 10 m sub-plot set up

in row 21-32– all tree above 10 cm DBH

were recorded– stem diameter and tree

height were measured – tree position was mapped at

sub-plot level– leave samples were collected

and identified

METHODS• Data analysis– Comparing tree population based on diameter

classes to 1979-1981 for row 21-32– Aboveground biomass estimation with four

allometry equations (Chave et al. 2005; Ketterings et al. 2001; Basuki et al. 2009; Brown 1987)

– Constructed species accumulation curve with bootstrapping techniques on the 10 m x 10 m sub-plot data

ALLOMETRY EQUATIONS• Chave et al. (2005)– B = exp(-1.499+2.148ln(D)+0.207(ln(D)2) -

0.0281(ln(D)3)• Ketterings et al. (2001)– B = 0.11D2.62

• Basuki et al. (2009)– ln(B) = -0.744+2.188ln(D)+0.832ln(D)

• Brown (1987)– B = 0.118D2.53

Where: B = biomass (kg tree-1); = wood density (g cm-3); D = diameter at breast height (1.3 m) (cm).

RESULTS

Recruited after second fire

Shade tolerant recruited after first fire; light demanding recruited after second fire

Recruited after first fire

Survived from first and second fire: Dipterocarpus cornutus, Eusideroxylon zwageri; Diospyros borneensis, Polidocarpus majadun

30%

Ketterings’ and Brown’s allometry equations are similar, except that Ketterings’ includes wood density, which results in 20% lower estimation than Brown’s

After fire, biomass of bigger tree >60 cm DBH dramatically decrease

• 95% of tree populations are of lower wood density species after fire compared those before fire

• 60% of them significantly lower up to 0.2 point different that contributed by light demanding “pioneer” species

• the remaining 5% are similar in wood density

Macaranga sp. is dominant canopy cover in naturally regeneration 30 years after initial fire

Performance of disturbed forest in SRF

1979-1981 survey: 552 species in 10.5 hectares (Kartawinata et al. 2008)Number of species increases dramatically up to 200 sites and at slower rate for larger number of sites

RESULTS for 1.8 ha inventories

1979-1981 survey: 254 species2011 survey: 192 species

Natural regeneration during 30 years after initial fire has not led to full species recovery regarding the species richness

CONCLUSIONS• Thirty years after a major fire:

– Tree population 100% recovered; proportion of tree 10-20 cm DBH increase about 10%, but the bigger tree decrease

– Aboveground biomass 60% recovered, but bigger tree above 60 cm DBH contribution still reach 30%

– Species richness 75% recovered at 1.8 ha, but for larger area can be lower

– Tree population dominated by light demanding species that contain lower wood density

• Implication to conservation measures:– Natural regeneration is possible after fire disturbance, but

recovery relies on seed availability from the surrounding areas– Ecological restoration is needed, especially when the landscape is

patchy such as the case in SRF

THANK YOU