Available online at www.sciencedirect.com
Catena 73 (2008) 212 – 224
www.elsevier.com/locate/catena
Peat–water interrelationships in a tropical peatland
ecosystem in Southeast Asia
J.H.M. Wösten a,⁎, E. Clymans a , S.E. Page b , J.O. Rieley c , S.H. Limin d
a
Alterra, Wageningen University and Research Centre, P.O. Box 47, 6700 AA Wageningen, The Netherlands
b
Department of Geography, University of Leicester, Leicester LE1 7RH, United Kingdom
c
School of Geography, University of Nottingham, Nottingham NG7 2RD, United Kingdom
d
University of Palangka Raya, Central Kalimantan, Indonesia
Received 20 November 2006; received in revised form 31 May 2007; accepted 18 July 2007
Abstract
Interrelationships between peat and water were studied using a hydropedological modelling approach for adjacent relatively intact and
degraded peatland in Central Kalimantan, Indonesia. The easy to observe degree of peat humification provided good guidance for the assignment
of more difficult to measure saturated hydraulic conductivities to the acrotelm–catotelm hydrological system. Ideally, to prevent subsidence and
fire, groundwater levels should be maintained between 40 cm below and 100 cm above the peat surface. Calculated groundwater levels for
different years and for different months within a single year showed that these levels can drop deeper than the critical threshold of 40 cm below the
peat surface whilst flooding of more than 100 cm above the surface was also observed. In July 1997, a dry El Niño year, areas for which deep
groundwater levels were calculated coincided with areas that were on fire as detected from radar images. The relatively intact peatland showed
resilience towards disturbance of its hydrological integrity whereas the degraded peatland was susceptible to fire. Hydropedological modelling
identified areas with good restoration potential based on predicted flooding depth and duration. Groundwater level prediction maps can be used in
fire hazard warning systems as well as in land utilization and restoration planning. These maps are also attractive tools to move from the dominant
uni-sectoral approach in peatland resource management toward a much more promising multi-sectoral approach involving various forestry,
agriculture and environment agencies. It is demonstrated that the combination of hydrology and pedology is essential for wise use of valuable but
threatened tropical peatland ecosystems.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Hydropedological modelling; Histosols; Peat subsidence; Fire susceptibility; Peat humification; Peatland restoration
1. Introduction
In Southeast Asia, peatlands cover more than 26 million
hectares (69% of all tropical peatlands), at altitudes from sea level
to about 50 m above, mostly near the coasts of East Sumatra,
Kalimantan, West Papua, Papua New Guinea, Brunei, Peninsular
Malaya, Sabah, Sarawak and Southeast Thailand (Fig. 1: Page
et al., 2004). There are approximately 6 million hectares of
peatland in Kalimantan (RePPProT, 1990; Radjagukguk, 1992)
with a thickness varying from 0.3 m to 20 m (Anderson, 1983).
Natural lowland tropical peatlands are dominated by trees (peat
swamp forest) and are important reservoirs of biodiversity, carbon
⁎ Corresponding author. Tel.: +31 317 486478; fax: +31 317 419000.
E-mail address: [email protected] (J.H.M. Wösten).
0341-8162/$ - see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.catena.2007.07.010
and water. Tropical peat swamp forests in their natural state make
an important contribution to regional and global biodiversity
(Andriesse, 1988; Page and Rieley, 1998) and provide a vital, but
undervalued habitat, for rare and threatened species, especially
birds, fish, mammals and reptiles (Ismail, 1999). They are one of
the most important remaining habitats for the last remaining large
populations of orangutan (Meijaard, 1997). At the present time, in
the absence of human intervention, many tropical peatlands are
actively forming peat or are in a steady state, although climate and
land use are no longer conducive to continued accumulation at all
sites (Rieley et al., 1996).
According to FAO-UNESCO (1990) Histosols are soils with
a surface layer containing more than 30% organic matter in
40 cm of the upper 80 cm of the profile (i.e. peat). Owing to the
low bulk density of most of the peat, tropical peatlands have a
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
213
Fig. 1. Distribution of lowland peatland in Southeast Asia and location of the study area.
high porosity and, as a consequence, a high water-holding
capacity that provides them with an important water regulation
function with respect to downstream tropical lowlands. Under
natural conditions tropical peatlands serve as reservoirs of fresh
water, moderate water levels, reduce storm-flow and maintain
river flows, even in the dry season, and they buffer against
saltwater intrusion. The physical, chemical and hydrological
characteristics of tropical peat differ considerably from those of
mineral soils. Within Histosols, the state of peat decomposition,
or peat humification degree, determines to a large extent
properties such as hydraulic conductivity, water retention,
nutrient availability and load-bearing capacity. Linking the
relatively easy to observe degree of humification of tropical peat
to the relatively difficult to measure saturated hydraulic
conductivity of the peat is an option that is explored in this study.
Peat consists of dead, partially decomposed plant remains
that have accumulated on the land surface for millennia under
waterlogged conditions. Recent investigations (Page et al.,
2004) have revealed that initiation of contemporary peat
deposits in Southeast Asia occurred in the Late Pleistocene
(∼ 30,000–24,000 14C yrs BP). Accumulation was most rapid
in the early Holocene (∼ 9600–7000 14C yrs BP, ∼ 11,000–
8000 cal yrs BP) and continued at a reduced rate until the
present day (Neuzil, 1997). These peatlands occupy mostly low
altitude coastal and sub-coastal situations but may extend inland
for distances of more than 150 km along river valleys and across
catchments. The conditions under which the peat has accumulated are those of poor drainage, permanent waterlogging, high
rainfall and substrate acidification. Most of the peatlands of
Southeast Asia have a characteristically domed, convex surface,
their water and nutrient supply is derived entirely from rainfall
(ombrogenous) and the organic substrate on which plants grow
is nutrient poor (Andriesse, 1988).
In common with other forest types in Indonesia, peat swamp
forests have been logged intensively through the official concession system. Most of this has now stopped because the
licenses have expired but illegal logging now threatens the
integrity and long-term stability of the peat swamp forest ecosystem (Böhm and Siegert, 2002). The narrow, shallow canals
that are dug into the surface of the peat to extract the felled
timber also promote rapid drainage of water from the peatland
landscape, leading to lowering of the groundwater level and
destruction of the hydrological integrity of the ecosystem and
increasing the risk of fire. Fires in these peatlands are mainly
human induced and are used in land clearance prior to conversion to agriculture or plantations or by local small farmers to
keep their land free of weeds.
Since the 1970s large areas of lowland tropical peatland in
Southeast Asia have been converted to agriculture, following
forest clearance and drainage. In Indonesia, for example, the
coastal swamps on shallow peat have been cultivated successfully
for many decades using traditional techniques that utilise daily
tidal movements to flush toxic organic substances out of the rice
fields (Notohadiprawiro, 1997). Agricultural development of
thicker tropical peats, beyond tidal influence has failed, largely
because planners considered peatlands to be simply another type
of land and did not take into account the special physical and
chemical properties of this peat. A glaring example is the Mega
Rice Project (1996–1999) in Central Kalimantan, which failed in
the attempt to convert about 1 million hectares of wetland (mostly
peatland) to rice fields (Muhamad and Rieley, 2002). Similar
problems of exploiting thick peat (N3 m) have been experienced
elsewhere in the Southeast Asia region (Page and Rieley, 1998).
Agricultural development and plantation forestry on tropical
peatlands both require drainage of the water saturated peat. As
soon as peat swamps are drained, however, the process of
subsidence starts, which can only be stopped by re-wetting the
peat again. On average 60% of peat subsidence is caused by
oxidation and 40% by irreversible drying or shrinkage of the
peat (Wösten et al., 1997). It has been estimated that subsidence
of every centimetre of a tropical peatland results in the emission
of 13 t CO2 per hectare per year (Wösten et al., 1997). Because
subsidence is positively related to drainage depth, different
types of land use, through their associated optimal water levels,
leads to substantial but varying carbon emissions (Wösten and
Ritzema, 2001).
Most peats shrink when dried but swell when re-wetted,
unless their water content falls below a threshold value at which
214
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
irreversible drying occurs. In this case, a layer of dry peat is
created on the surface, which is very susceptible to fire (Hooijer
et al., 2006). Oxidation of peat in combination with burning of
dry surface peat results in significant carbon outputs to the
atmosphere contributing to climate change processes (Siegert
et al., 2001; Page et al., 2002). Extensive, human induced fires
have already damaged severely the natural resource functions of
large tracts of tropical peatland, while repeat fires in the same
locations have caused irreversible damage. These fire events
have potentially devastating environmental consequences, both
regionally and globally. The Mega Rice Project disrupted the
peat swamp forest ecosystem over an area of at least one and a
half million hectares as it became fire prone because of
excessive deforestation and drainage (Page et al., 2002). In the
aftermath of the fires water absorption and retention properties
of the peatland were impaired. As a result, there has been
increased flooding of these vast landscapes during the rainy
season with impacts on downstream habitations, whilst during
subsequent dry seasons there has been an increased susceptibility to fire (Siegert et al., 2002).
The objective of this study was to assess the interrelationships
between peat and water in a large area of tropical peatland in
Central Kalimantan, Indonesia by focusing on fire susceptibility
as influenced by changes in groundwater levels modelled with
the SIMGRO hydropedological model. The influence of
relatively dry and wet years on fire susceptibility is demonstrated
as well as the influence of monthly fluctuations in rainfall. In
addition, the vegetative restoration potential of degraded
peatlands is assessed in relation to flooding depth and duration.
2. Site information
2.1. Location and history
The study area is the catchment of Sungai (= river) Sabangau
in Central Kalimantan, Indonesia, (Fig. 1), bounded by Sungai
Kahayan in the East, the Java Sea to the South, Sg. Katingan in
the West and the City of Palangka Raya to the North. Sg.
Sabangau is a blackwater river that originates in, and drains, the
last remaining, large continuous area (13,000 km2) of dense
peat swamp forest in Borneo (Fig. 2). The area between the Sg.
Sabangau and the Sg. Kahayan is ‘Block C’ of the former Mega
Rice Project area. Since the study location consists of both the
relatively intact Sabangau catchment and the degraded Block C
of the former Mega Rice Project, it is possible to compare two
contrasting peatland landscapes in terms of their fire susceptibility and restoration potential.
In 1996 the Indonesian Government initiated the One Million
Hectare Mega Rice Project, the principal objective of which was
to grow rice in the area between the Sg. Sabangau in the West,
Sg. Kahayan, Sg. Kapuas and Sg. Barito in the East, and the Java
Sea in the South. More than 4000 km of drainage and irrigation
channels were excavated and some 50,000 migrant workers
were transported to the area (Muhamad and Rieley, 2002). The
project failed and was closed down in 1999 because it focussed
solely on drainage while it ignored the fact that drained peatlands
start to subside and eventually become fire prone. The hard
lesson learned was that changing the hydrology of intact
peatlands has a marked effect on its pedological characteristics
Fig. 2. Satellite image of the study area and its surroundings as well as the boundary of the hydropedological modelling (Source: Google Earth, 2006).
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
to an extent that continued drainage causes irreversible damage
to the peatland ecosystem.
In the El Niño extended dry seasons of 1997, 1998 and 2002
over 60,000 fire hotspots were detected in Borneo by satellite
(Langner and Siegert, 2006), many of which were in the former
Mega Rice Project area. A similar situation occurred again in
2006. Fires ignited both the surface vegetation (primary and
secondary forest, brush and agricultural land) and the
underlying peat. As a result, the area of peat swamp forest
around Palangka Raya was reduced from 2,406,732 ha in 1991
to 1,112,151 ha in 2001 (Böhm and Siegert 2002).
2.2. Topography and peats
Peatlands in the Sabangau catchment and in Block C of the
former Mega Rice Project are dome shaped, with an elevation
that increases from sea level at the coast to only some 30 m above
sea level 200 km inland. These peatlands overlie marine clay and
mud near to the coast (potential acid sulphate soil) while inland
they are underlain by sand, gravel and clay deposits of fluvial
origin (Sieffermann et al., 1988). In the upper Sg. Sabangau
catchment there are extensive areas where the peat reaches a
thickness of 10 m and more towards the centre of the domes. Peat
thickness decreases towards the major rivers where it is absent
from the alluvial levees (Page et al., 1999). The peat consists
mainly of slightly or partially decomposed trunks, branches and
roots of trees within a matrix of almost structureless organic
215
material that also originates from rainforest plants, mostly trees
(Rieley et al., 1996). The decomposition status of the peat is
fibric (least humified) on the surface, hemic (moderately
humified) throughout most of its thickness and sapric (most
humified) near the bottom. Dry bulk densities of a peat core
sampled in the Sabangau catchment under low pole forest were
low and ranged from 0.02 to 0.21 g cm− 3. Samples are collected
as cores in a cylindrical corer of known volume. Bulk density is
obtained after drying samples to constant weight. The peat in the
study area is ombrotrophic and for most of the profile bulk
density is very low. Higher bulk densities occur (a) near to the
mineral layer underlying the peat owing to presence of mineral
material and (b) at the surface owing to oxidation and
compaction. Ash contents in the same core varied between
0.33% and 1%, while mean pH was 3.2 ± 0.4 (Weiss et al., 2002).
The water draining from this peatland is black owing to high
levels of humic and fulvic acids, polyphenols and other products
of organic matter decomposition.
2.3. Climate, hydrology and land use
The climate of the island of Borneo is characterised by a
rather constant temperature throughout the year, high humidity
and high rainfall. Annual precipitation for the period 1994–
2004, used in the hydropedological modelling fluctuates
considerably (Fig. 3) (Takahashi et al., 2004). Over this 11
year period, 1997 was the driest year with 1848 mm rainfall,
Fig. 3. Annual precipitation for the period 1994–2004 (top), and monthly precipitation for the dry year 1997 (bottom) at the test site (Source: Takahashi et al., 2004).
216
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
2003 was a ‘normal’ year with 2570 mm rainfall and 1999 was
the wettest year with 3788 mm rainfall. The average evaporation is fairly constant, varying between 3.5 mm d− 1 and
4.8 mm d− 1 with a total of around 1500 mm per year (Takahashi
et al., 2004).
In unsaturated tropical peatlands, almost all of the rainfall
infiltrates owing to the high hydraulic conductivity of the surface
peat layer (Takahashi and Yonetani, 1997). Once the peat is
saturated with water, subsequent rainfall flows over and through
the surface towards surrounding rivers or drainage canals. The
high hydraulic conductivity results from the physical characteristics of the top peat layer (fibric and hemic) with its open, large
pore structure (Rieley et al., 1996). As a consequence, only a
very small proportion (b3%) of the rainfall penetrates into the
deeper, (below 1 m) peat layers with their lower hydraulic
conductivities (Hooijer, 2005). Lateral water flow towards rivers
and drainage canals is thus large compared to downward flow
into deeper layers.
Land use in the study area varies according to the intensity of
human activities: 1) natural peat swamp forest, 2) destroyed,
abandoned and fire-damaged peatland, and 3) cultivation. The
Sabangau catchment falls largely into category 1 with villages
situated along the Sg. Sabangau and Sg. Katingan. The main
sources of income for people living in this area are (illegal)
logging, fishing and subsistence agriculture. Block C falls
mostly in categories 2 and 3 although there are some remnants
of peat swamp forest in the northern part. The canals that were
excavated in establishing the Mega Rice Project (Fig. 2) were
part of a water management system designed to maintain
optimal groundwater levels for rice crop production. While a
small part of Block C is in agricultural use, the majority is firedamaged and degraded wasteland (Limin et al., 2006).
2.4. Problem description
In recent decades the natural vegetation of peat swamp forest
covering much of the peatland area in Central Kalimantan has
been removed as a result of both legal (concession) and illegal
logging and conversion to agricultural use, the last of which
requires drainage and water management. Logging (Siegert
et al., 2001), fires and drainage (Page et al., 2002) destroy many
of the ecosystem functions of tropical peatland, in particular
those relating to their hydrology. Currently, much of the former
Mega Rice Project area floods in the rainy season while, in the
dry season, groundwater levels fall far below the surface
thereby decreasing moisture contents of the surface peat and the
vegetation it supports and creating huge amounts of dry fuel that
are a fire hazard. The surface peat is only moderately to little
humified (hemic to fibric) and it contains many large pores that
lose their water quickly when the groundwater level is only
slightly lowered.
The relationship between water content and groundwater
level is determined by the water retention characteristic of the
surface peat layer. When the groundwater level in tropical peat
drops below 40 cm from the surface, which under hydrostatic
equilibrium is equivalent to a pressure head of -4 kPa, the
moisture content of the little humified top layer decreases from
about 0.90 cm3 cm− 3 at saturation to about 0.50 cm3 cm− 3 at a
pressure head of − 4 kPa (Rieley and Page, 2005), thus making it
susceptible to fire (Takahashi et al., 2003; Usup et al., 2004).
Fires that take hold in dry surface biomass and dry peat spread
rapidly (Frandsen, 1997). Fire produces large gaps in the forest
leading to increased wind circulation which, when combined
with greater penetration of sunlight to the forest floor,
encourage rapid growth of secondary understorey vegetation.
Fig. 4. Digital Elevation Model for the study area (Source: BAKOSURTANAL, 1997).
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
As a result, temperature and humidity increase while soil
moisture decreases adding greatly to the susceptibility of the
landscape to subsequent fires. In this way, the initial fires
damage the remaining forest severely and increase the risk of
recurrent fires significantly (Siegert et al., 2001). This
destructive sequence of events can be stopped only by rewetting
the peatland and reinstating the hydrological integrity of the
peat swamp ecosystem (Wösten and Ritzema, 2001).
217
Table 1
Degrees of peat humification and associated saturated hydraulic conductivities
Fibre content
USDA
Corresponding Assigned saturated hydraulic
classification von Post class conductivity (m d− 1)
Over 66%
Fibric
33–66%
Hemic
Less than 33% Sapric
H1 to H3
H4 to H6
H7 to H10
30
30
0.5
In the absence of an accurate, detailed Digital Elevation
Model (DEM) for the study area the soil elevation map prepared
by the Indonesian National Coordinating Agency for Surveys
and Mapping (BAKOSURTANAL, 1997) was used to generate
a DEM. As shown in Fig. 4, in total 8 different elevation classes
have been distinguished while each class has a width of 2.5 m.
Despite the limited resolution of this DEM, using grid sizes of
1000 by 1000 m, it provides appropriate information on
elevation for the Sabangau catchment and Block C of the former
Mega Rice Project.
realized that the relatively few measurements available for
tropical peatlands show a considerable range. The two layer
system applied in this study is consistent with the ‘acrotelm’ or
surface layer and ‘catotelm’ or deeper layer concept as used to
stratify boreal and temperate peatlands (Ingram, 1978).
Differences in peat thickness and hydraulic conductivity result
in hydraulic transmissivities (cumulative thickness multiplied
with conductivity) ranging from 30 to 40 m2 d− 1. In addition, a
peat water storage coefficient is required as model input
parameter. This coefficient was not measured directly but
obtained in the model calibration process and set at 0.5 (Wösten
et al., 2006a). Next, modelling of groundwater levels was performed for combinations of peat thickness (Fig. 5) and the three
land use categories defined above.
3.2. Hydropedological modelling
3.3. Model calibration and validation
The physically-based SIMGRO (SIMulation of GROundwater flow and surface water levels) model was used to simulate
water flow in the saturated zone, unsaturated zone, river
channels and over the peat surface (Dik, 2004). The catchment
boundaries of Sg. Sabangau and Block C were chosen as the
model boundary (Fig. 2). Using the DEM and the watercourses
map, delineations of the project area were determined using the
hydrology extension in the GIS package Arcview. In the
modelling distances between the grid sizes was set to 1500 m,
resulting in a network of 5791 grids. Saturated groundwater
flow was modelled using the finite element method for which
the model area was subdivided into triangular segments. The top
of the mineral layer was set as aquifer bottom. Hydraulic
conductivity of the peat is an essential element of hydropedological modelling. In turn, the hydraulic conductivity and
also the moisture retention relationship of the peat is strongly
influenced by the degree of humification of the peat.
Humification of peat has been classified using the von Post
(1922) scale which ranges from H1 (least humified) to H10
(most humified). For simplification purposes the 10 von Post
classes are condensed based on their fibre content into 3 main
peat types using the US Department of Agriculture (USDA,
1993) terminology (Table 1). Based on hydraulic conductivity
measurements using the pumping test method as reported by
Ong and Yogeswaran (1992) and by Takahashi and Yonetani
(1997) the peat profile in this study is schematized in a two layer
system consisting of a fibric to hemic peat top layer (0–100 cm)
with an average saturated hydraulic conductivity of 30 m d− 1
and a deeper, sapric peat layer with an average saturated
hydraulic conductivity of 0.5 m d− 1 (Table 1). While using
these average saturated hydraulic conductivities it should be
Groundwater levels estimated using both the original
and calibrated model are shown in Fig. 6. The correlation
coefficient (R2 ), the root mean square error (RMSE) and the
mean square error (MSE) improved from 0.67, 27.23 and
60.70 for the original model to 0.74, 5.22 and 7.79 for the
calibrated model. After calibration of the peat water coefficient, the model represents groundwater levels measured in a
dip well at the test site (Fig. 2) with acceptable (within 10 cm)
accuracy.
During July in the El Niño year 1997, 40% of the study area
had calculated groundwater levels lower than the critical
threshold of 40 cm (Usup et al., 2004) below the peat surface
and flooding occurred in only a few locations in the southeast
(top of Fig. 7). Fire damaged areas derived from radar satellite
images (Siegert et al., 2002) is shown in the map at the bottom
of Fig. 7. GIS overlay of these two maps yields a
correspondence of 74% between areas with groundwater levels
more than the critical threshold of 40 cm and areas damaged by
fire. For 18% of the area groundwater levels were below the
critical 40 cm threshold and thus fire risk was modelled, but no
actual fire damage was detected from the radar satellite images.
For 8% of the area actual fire damage was detected while this
area was not modelled as a fire risk zone. Given the
uncertainties involved, the results show that in general modelled
groundwater levels are good indicators of fire susceptibility.
It is also important to note (Fig. 7) that in 1997 groundwater
in Block C dropped to significantly lower levels than in the
Sabangau catchment, with fire damage much greater in the
former. From comparison of measured and modelled groundwater levels (Fig. 6) and predicted and observed fire damage
(Fig. 7), it is concluded that the results of the hydropedological
3. Methods
3.1. Digital elevation model (DEM)
218
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
Fig. 5. Peat thickness map for the study area.
modelling are realistic and agree well with independently obtained, measured and observed data.
4. Results and discussion
4.1. Yearly variations in fire risks
To assess other interrelationships between peat and water,
hydropedological modelling was used to calculate groundwater
levels for different years and for different months within a single
year. Results are interpreted in terms of fire susceptibility and
restoration potential of the relatively intact Sabangau catchment
and the degraded Block C of the former Mega Rice Project.
Groundwater levels calculated for July in the non El Niño
year 2003 (map at top of Fig. 8) reveal that 23% of the study
area was below the critical threshold of 40 cm and thus at high
fire risk. In July 1999, however, the area at risk was only 2%
(map at bottom of Fig. 8). In contrast, 40% of the area was prone
to fire in July 1997 (El Niño year) (Fig. 7) showing that there is
considerable variation in fire risk from year to year depending
Fig. 6. Measured and calculated groundwater level (m above mean sea level) at the test site versus time. Distinction is made between results obtained with the original
and calibrated model. The critical threshold groundwater level of 40 cm below surface is also indicated.
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
219
Fig. 7. Calculated groundwater levels (top) in the study area and a clip of fire damaged areas as derived from radar satellite images (bottom) for July in the El Niño year
1997.
upon annual rainfall, length of the dry season and consequent
depth of the groundwater level below the peat surface.
Comparison of the relatively intact peat swamp forest in the
Sabangau catchment with the degraded peatland in Block C
shows that in the 3 years, 1997, 1999 and 2003, groundwater
levels were lower in the latter and therefore the associated fire
risks were higher. This highlights the negative effect of the
drainage canals on peatland hydrology in Block C that are
leading to enhanced peat oxidation, subsidence and loss of
stored carbon (Page et al., 2002). Maps of groundwater levels
can thus be used in land utilisation and restoration planning, for
example, to indicate where hydrological restoration efforts
should be targeted, as well as in fire hazard warning systems.
Such systems become even more important as future climate
220
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
Fig. 8. Calculated groundwater levels in the study area for the months July in the normal year 2003 with 2570 mm rainfall (top) and in the wet year 1999 with 3788 mm
rainfall (bottom).
scenarios predict an increase in years with prolonged dry seasons and thus an increase in the number of days with high risk of
fire (Goldammer and Price, 1998).
4.2. Monthly variations in fire risks
The El Niño year 1997 was characterised by low rainfall and
the high incidence of forest and peat fires. These fires contributed greatly to the so-called haze (particulate-laden smog
and cocktail of hazardous chemicals) that drifted across
Southeast Asia. Because of the severity and scale of the fires,
1997 was chosen to study the monthly variation in groundwater
levels and associated fire risks. During the 5 months from June
to October, it hardly rained at all, making the dry season
exceptionally long (Fig. 3). The calculated groundwater levels
for the 15th day of every month (Fig. 9) show that from January
to May they seldom dropped below 40 cm. Flooding occurred in
February, April and May but from June onwards groundwater
levels dropped progressively and sharply, with the result that
water levels were below the 40 cm critical threshold in 91% of
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
221
Fig. 9. Calculated groundwater levels in the study area for the dry year 1997 at every 15th day of the month.
the area by October and therefore susceptible to fire. By the
same time, 33% of the area had groundwater levels deeper than
100 cm below the surface making these an extreme fire risk. By
November and December groundwater levels had started to rise
once more resulting in some flooding in December. Groundwater level rise occurs some time after the rain commences
222
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
owing to water retention within the dry surface peat as the water
reservoir recharges (Fig. 9). The ecological health of this
peatland area, indicated in Fig. 9, shows the resilience of the
ecosystem to disturbances of its hydrological integrity caused
by a variety of human activities, including canal construction,
(illegal) logging and conversion to agriculture (Wösten et al.,
2006b). From Fig. 9 it is also evident that the Sabangau catchment returned to its original hydrological status in December
1997, whereas Block C remained drought affected and thus
susceptible to fires for longer.
4.3. Restoration potential
Restoration of a vegetation cover on damaged peatlands
involves replanting with native peat swamp tree species.
Successful replanting requires rewetting of the peatland by
raising groundwater levels and maintaining them as close as
possible to the peat surface during the dry season. It is also
important, however, to prevent prolonged flooding of the surface
during the wet season. Construction of dams across drainage
canals will be required to elevate water levels and reinstate the
hydrological integrity of the peat swamp ecosystem (Wösten and
Ritzema, 2001). Water levels should be regulated to prevent
excessive surface flooding during the wet season.
To optimise this restoration process it is useful to apply
hydropedological modelling for the identification of areas with a
good restoration potential. Towards this aim, groundwater levels
were calculated for the wet month of November in the wet year
1999 and these are presented in five different classes in Fig. 10.
Based on a study of a comparable peatland in the Berbak National
Park in Sumatra, Wösten et al. (2006b) derived a relationship
between flooding depth and flooding duration and the vegetation
types that could be established. They concluded that for replanting
to be successful it is necessary to maintain groundwater levels
above the critical 40 cm level below the surface while flooding
more than 100 cm above the peat surface for a prolonged period of
time should be avoided. Applying this relational diagram to the
Sabangau catchment suggests that areas with flooding deeper than
100 cm for a prolonged period of time have no restoration
potential for peat swamp forest and will change into areas of
secondary vegetation dominated by screw palm (Pandanus sp.),
ferns, sedges and grasses with a characteristically low biodiversity. Areas with groundwater levels between 40 cm below and
100 cm above surface have a high potential for natural or assisted
regeneration of native tree species with a high diversity. In Berbak
National Park fern species such as Blechnum indicum dominated
while pioneer tree species such as Macaranga pruinosa and
Alstonia pneumatophora proliferated well under these hydrological
conditions (Wösten et al., 2006b). Because circumstances in
Central Kalimantan differ from those in Sumatra, other species
may be more suitable for restoration of the study area and this
requires further investigation. Successful restoration depends of
course not only on the groundwater levels in a wet month of a
wet year (Fig. 10), but to a large extent also on prevention of
deep groundwater levels (more than 40 cm below surface) in dry
months of dry years (Fig. 9). Under both wet and dry weather
conditions modelling of groundwater levels for larger spatial
areas and for larger time periods will yield information on the
spatial and temporal variations in the soil water regime which is
important for the assessment of the restoration potential of
peatlands.
4.4. Discussion
Tropical peatlands perform many vital natural resource
functions. Their value in this respect is enhanced because of
the vast areas of landscape they cover and the carbon storage,
biodiversity preservation and water regulatory roles they
Fig. 10. Vegetative restoration potential for November in the wet year 1999 as related to calculated groundwater levels.
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
perform. In addition, their influence operates on several scales,
locally, regionally and globally. Owing to the sensitivity of this
ecosystem and the length of time it has taken for co-evolution to
take place, tropical peatlands must be treated with caution and
respect before conversion to other land uses. The reason why
there is still a large area of peat swamp forest remaining in the
tropics is probably because of the inherent problems of
maintaining productive, economic agriculture on deforested,
drained, acid peat. Pressure to convert tropical peatland has been
mounting, however, over the last 20 years or so because almost
all lowland and middle altitude dry land forests have been
exploited for their timber and the land converted to arable
agriculture or plantation crops. Requirements for food and space
for burgeoning populations in the tropical zone mean that the
remaining peatlands are now under major threat of development.
Continuing logging and drainage of peatlands, especially for
cultivation of plantation crops such as oil palm, are major
threats. To reverse this trend it is essential that future land use of
tropical peatland takes fully into account the principles and
practices of 'wise use' that involve several elements, foremost
amongst which is the identification of the benefits and values
that tropical peatlands can provide in their natural state and the
adverse environmental and human consequences resulting from
their disturbance, both in situ and ex situ (Rieley and Page,
2005). In this way, the role that different peatlands play in
providing goods and services to society can be assessed
properly and the scale and consequences of various impacts
forecast. The objective of wise use is to assist those involved in
tropical peatland planning and management to identify the best
land use option for any peatland and to plan its design and
implementation. Whatever may be the success of this approach
in influencing those most concerned with management of
tropical peatlands their activities should be directed holistically,
stressing the importance of studying and evaluating entire
catchments of which a peatland may constitute all or a major
part. Integrated catchment management that takes into account
all of the eco-hydrological functions of a peatland is essential in
order to achieve environmentally sustainable management of
the tropical peatland resource. Most important of all is that
within development proposals there must be major provision for
the maintenance of biodiversity and protection of the important
natural resource functions of this critical ecosystem.
5. Conclusions
This study shows that water management is a key element in the
wise use of peatlands. In dry years groundwater levels drop below
the critical threshold of 40 cm. Deep groundwater levels mean an
increased subsidence of the peat by oxidation as well as an increase
in fire susceptibility. Both oxidation and fire transform peatlands
from carbon sinks under pristine conditions into carbon sources
with important local, regional and global consequences under
drained conditions. In wet years, flooding depth and flooding
duration have adverse consequences for the restoration potential of
peatlands. Ideally, groundwater levels should vary between 40 cm
below and 100 cm above the land surface. Comparison of the
adjacent relatively intact Sabangau catchment and the degraded
223
Block C area reveals that in their natural state, tropical peatlands
show sufficient resilience to disturbance provided their hydrological integrity is maintained or restored. Once they are in a degraded
state, however, tropical peatlands become an increasingly fragile
ecosystem that is likely to disappear.
Hydropedological modelling yields information on the
groundwater dynamics that, once calibrated and validated for a
specific point in time and space, can be extrapolated using
groundwater level prediction maps to larger areas and for different
time periods. These maps are helpful tools in fire hazard warning
systems, which until now in their simple form, have relied mainly
on indirect information such as cumulative rainfall and sequence
of dry days. In addition, these maps can also be used in land
utilization and restoration planning as a strong correlation exists
between groundwater level and vegetation type.
One of the biggest obstacles to the wise use of tropical
peatlands is the division of responsibility for the resources within
the peatland landscape to several separate agencies (e.g. forestry,
agriculture, environment) that operate virtually independently of
each other. These agencies often fail to recognise the consequences that their unisectoral involvement has for others and
this means that problems of tropical peatland resource management are not addressed in a coordinated and integrated manner. A
multiple wise use approach is more likely to safeguard the range
of functions, attributes and services that peat swamp forest
provides to a wide range of stakeholders (Silvius and Giesen,
1996). Presenting results of hydropedological modelling in the
form of informative maps showing groundwater dynamics in
space and time is an attractive tool to move from the unisectoral
approach towards the inter-agency discussion as groundwater
level dynamics is the key issue for wise use of tropical peatlands.
Acknowledgement
This study was supported by the European Union RESTORPEAT project (INCO-CT-2004-510931).
References
Anderson, J.A.R., 1983. The tropical peat swamps of western Malaysia. In:
Gore, A.J.P. (Ed.), Mires: swamp, bog, fen and moor. B. Regional studies.
Elsevier, Amsterdam, pp. 181–199.
Andriesse, J.P., 1988. Nature and Management of Tropical Peat Soils. FAO Soils
Bulletin 59 Rome.
BAKOSURTANAL (National Coordinating Agency for Surveys and Mapping),
1997. Jakarta, Indonesia. www.bakosurtanal.go.id.
Böhm, H.D.V., Siegert, F., 2002. Land use change and (il)-legal logging in
Central Kalimantan, Indonesia. In: Rieley, J.O., Page, S.E. (Eds.), Peatlands for
people: natural resource functions and sustainable management, Proceedings
of the International Symposium on tropical peatland, Jakarta, 22-23 August
2001. Agency for the Assessment and Application of Technology and
Indonesian Peat Association, Jakarta, Indonesia, pp. 132–140.
Dik, P.E., 2004. SIMGRO 4.1.2 User's Guide, Alterra-Report 913.3 ISSN 15667197. Alterra, Green World Research, Wageningen.
FAO-UNESCO, 1990. Guidelines for soil description, 3rd ed. Food and Agriculture
Organisation, Rome.
Frandsen, W.H., 1997. Ignition probability of organic soils. Canadian Journal of
Forest Research, 27, 1471–1477.
Goldammer, J.G., Price, C., 1998. Potential impacts of climate change fire
regimes in the tropics based on MAGICC and a GISS GCM-derived
lightning model. Climatic Change 39 (2-3), 273–296.
224
J.H.M. Wösten et al. / Catena 73 (2008) 212–224
Hooijer, A., 2005. Hydrology of tropical wetland forests: recent research results
from Sarawak peat swamps. In: Bonell, M., Bruijnzeel, L.A. (Eds.), Forest,
water and people in the humid tropics. Cambridge University Press, United
Kingdom, pp. 447–461.
Hooijer, A., Silvius, M., Wösten, H., Page, S., 2006. PEAT–CO2, Assessment of
CO2 emissions from drained peatlands in SE Asia. Delft Hydraulics report
Q3943.
Ingram, H.A.P., 1978. Soil layers in mires: function and terminology. Journal of
Soil Science 29 (2), 224–227.
Ismail, Z., 1999. Survey of fish diversity in peat swamp forest In: Yuan, C.T.,
Havmoller, P. (Eds.), Sustainable Management of Peat swamp Forest in
Peninsular Malaysia. Forestry Department Peninsular Malaysia, Kuala
Lumpur, pp. 173–198.
Langner, A., Siegert, F., 2006. Fires in Kalimantan and Sumatra 2006. Remote
Sensing Solutions and GeoBio Center Ludwig-Maximillians-University,
Munich, Germany. www.restorpeat.alterra.wur.nl.
Limin, S.H., Rieley, J.O., Page, S.E., Yunsiska, E., 2006. Peat thickness, type of
minerals in the bottom peat layer and hydrology status should be taken into
account when utilizing tropical peatland for agricultural purposes. In: Wösten,
J.H.M., Radjagukguk, B. (Eds.), The role of tropical peatlands in global change
processes, Proceedings of the Open Science Meeting 2005, Yogyakarta, 27–29
September 2005, pp. 155–165. www.restorpeat.alterra.wur.nl.
Meijaard, E., 1997. The importance of swamp forest for the conservation of the
orang utan (Pongo pygmaeus) in Kalimantan, Indonesia. In: Rieley, J.O.,
Page, S.E. (Eds.), Biodiversity and Sustainability of Tropical Peatlands.
Samara Publications, Cardigan, United Kingdom, pp. 243–254.
Muhamad, N.Z., Rieley, J.O., 2002. Management of tropical peatlands in
Indonesia: mega reclamation project in Central Kalimantan. In: Rieley, J.O.,
Page, S.E. (Eds.), Peatlands for people: natural resource functions and
sustainable management, Proceedings of the International Symposium on
tropical peatland, Jakarta, 22–23 August 2001. Agency for the Assessment
and Application of Technology and Indonesian Peat Association, Jakarta,
Indonesia, pp. 155–162.
Neuzil, S.G., 1997. Onset and rate of peat and carbon accumulation in four
domed ombrogenous peat deposits, Indonesia. In: Rieley, J.O., Page, S.E.
(Eds.), Biodiversity and Sustainability of Tropical Peatlands. Samara
Publications, Cardigan, United Kingdom, pp. 55–72.
Notohadiprawiro, T., 1997. Twenty-five years experience in peatland development for agriculture in Indonesia. In: Rieley, J.O., Page, S.E. (Eds.),
Biodiversity and Sustainability of Tropical Peatlands. Samara Publications,
Cardigan, United Kingdom, pp. 301–310.
Ong, B.Y., Yogeswaran, M., 1992. Peatland as a resource for water supply in
Sarawak. In: Aminuddin, B.Y., Tan, S.L., Aziz, B., Samy, J., Salmah, Z.,
Siti, Petimah, Choo, S.T. (Eds.), Proceedings of the International
Symposium on Tropical Peatland. Kuching, Sarawak, May. 1991. Malaysia:
Ministry of Agriculture, MARDI, pp. 255–268.
Page, S.E., Rieley, J.O., 1998. Tropical peatlands: a review of their natural
resource functions, with particular reference to Southeast Asia. International
Peat Journal 8, 95–106.
Page, S.E., Rieley, J.O., Shotyk, W., Weiss, D., 1999. Interdependence of peat
and vegetation in a tropical peat swamp forest. Proceedings of the Royal
Society of London. Series B 354, 1885–1897.
Page, S.E., Siegert, F., Rieley, J.O., Böhm, H-D.V., Jaya, A., Limin, S., 2002.
The amount of carbon released from peat and forest fires in Indonesia in
1997. Nature 420, 61–65.
Page, S.E., Wüst, R.A.J., Weiss, D., Rieley, J.O., Shotyk, W., Limin, S., 2004. A
record of Late Pleistocene and Holocene carbon accumulation and climate
change from an equatorial peat bog (Kalimantan, Indonesia): implications
for past, present and future carbon dynamics. Journal of Quaternary Science,
19, 625–635.
Post, L. von, 1922. Sveriges geologiska undersöknings torvinventering och
några av dess hittils vunna resultat. Sveriges Mosskulturföreningens, vol. 1.
Tidskrift, pp. 1–27.
Radjagukguk, B., 1992. Utilisation and management of peatlands in Indonesia
for agriculture and forestry. In: Aminuddin, B.Y., Tan, S.L., Aziz, B., Samy,
J., Salmah, Z., Siti, Petimah, Choo, S.T. (Eds.), Proceedings of the
International Symposium on Tropical Peatland. Kuching, Sarawak, May.
1991. Malaysia: Ministry of Agriculture, MARDI, pp. 21–27.
RePPProt, 1990. A national overview from the regional physical planning
programme for transmigration. UK Overseas Development Administration
and Directorate BINA Programme. Jakarta: Ministry of Transmigration.
Rieley, J.O., Ahmad-Shah, A.A., Brady, M., 1996. The extent and nature of
tropical peat swamps. In: Maltby, E., Immirzi, C.P., Stafford, R.J. (Eds.),
Tropical lowland peatlands of Southeast Asia. IUCN, Gland, Switzerland,
pp. 17–53.
Rieley, J.O., Page, S.E., 2005. Wise Use of Tropical Peatlands: focus on
Southeast Asia. Alterra, Wageningen University and the EU INCO
STRAPEAT and RESTORPEAT Partnerships, Wageningen, the Netherlands. ISBN: 90327 0347 1, 266 pp www.restorpeat.alterra.wur.nl.
Sieffermann, R.G., Fournier, M., Truitomo, S., Sadelman, M.T., Semah, A.M.,
1988. Velocity of tropical peat accumulation in central Kalimantan, Borneo.
Proceedings of the Eighth International Peat Congress, Leningrad, USSR,
vol. 1. International Peat Society, pp. 90–98.
Siegert, F., Rücker, G., Hindrichs, A., Hoffman, A.A., 2001. Increased damage
from fires in logged forests during droughts caused by El Niño. Nature 414,
437–440.
Siegert, F., Böhm, H.D.V., Rieley, J.O., Page, S.E., Jauhiainen, J., Vasander, H.,
Jaya, A., 2002. Peat fires in Central Kalimantan, Indonesia: fire impacts and
carbon release. In: Rieley, J.O., Page, S.E. (Eds.), Peatlands for people: natural
resource functions and sustainable management, Proceedings of the International Symposium on tropical peatland, Jakarta, 22-23 August 2001. Agency
for the Assessment and Application of Technology and Indonesian Peat
Association, Jakarta, Indonesia, pp. 142–154.
Silvius, M.J., Giesen, W., 1996. Towards integrated management of Sumatran
peat swamp forests. In: Maltby, E., Immirzi, C.P., Stafford, R.J. (Eds.),
Tropical lowland peatlands of Southeast Asia. IUCN, Gland, Switzerland,
pp. 247–267.
Takahashi, H., Yonetani, Y., 1997. Studies on microclimate and hydrology of
peat swamp forest in Central Kalimantan, Indonesia. In: Rieley, J.O., Page,
S.E. (Eds.), Biodiversity and Sustainability of Tropical Peatlands. Samara
Publications, Cardigan, United Kingdom, pp. 179–187.
Takahashi, H., Usup, A., Hayasaka, H., Limin, S.H., 2003. Estimation of ground
water levels in a peat swamp forest as an index of peat/forest fire. Proceedings
of the International Symposium on Land Management and Biodiversity in
Southeast Asia. Bali, Indonesia, 17-20 September 2002. Hokkaido University
Japan and Indonesian Institute of Sciences, Bogor, Indonesia, pp. 311–314.
Takahashi, H., Usup, A., Hayasaka, H., Kamiya, M., Limin, S.H., 2004. The
importance of ground water level and soil moisture of subsurface layer on
peat/forest fire in a tropical peat swamp forest. In: Päivänen, J. (Ed.), Wise
Use of Peatlands. Volume 1. Proceedings of the 12th International Peat
Congress, Tampere, Finland, 6–11 June 2004. International Peat Society,
Jyväskylä, Finland, p. 760.
USDA, 1993. Soil Survey Manual, Soil Survey Division Staff. United States
Department of Agriculture.
Usup, A., Hashimoto, Y., Takahashi, H., Hayasaka, H., 2004. Combustion and
thermal characteristics of peat fire in tropical peatland in Central
Kalimantan, Indonesia. Tropics 14 (1), 1–19.
Weiss, D., Shotyk, W., Rieley, J., Page, S., Gloor, M., Reese, S., MartinezCortizas, A., 2002. The geochemistry of major and selected trace elements in
a forested peat bog, Kalimantan, SE Asia, and its implications for past
atmospheric dust deposition. Geochimica et Cosmochimica Acta 66 (13),
2307–2323.
Wösten, J.H.M., Ismail, A.B., Van Wijk, A.L.M., 1997. Peat subsidence and its
practical implications: a case study in Malaysia. Geoderma 78, 25–36.
Wösten, J.H.M., Ritzema, H.P., 2001. Land and water management options for
peatland development in Sarawak, Malaysia. International Peat Journal 11,
59–66.
Wösten, J.H.M., Hooijer, A., Siderius, C., Satriadi Rais, D., Aswandi, Idris,
Rieley, J.O., 2006a. Tropical peatland water management modelling of the
Air Hitam Laut catchment in Indonesia. International Journal of River Basin
Management 4 (4), 233–244.
Wösten, J.H.M., Van den Berg, J., Van Eijk, P., Gevers, G.J.M., Giesen, W.B.J.T.,
Hooijer, A., Aswandi, Idris, Leenman, P.H., Satriadi Rais, D., Siderius, C.,
Silvius, M.J., Suryadiputra, N., Tricahyo Wibisono, I., 2006b. Interrelationships
between hydrology and ecology in fire degraded tropical peat swamp forests
International Journal of Water Resources Development 22 (1), 157–174.