Limnol. Oceanogr.

, 52(6), 2007, 2410–2417

E 2007, by the American Society of Limnology and Oceanography, Inc.

Biogeochemistry of the Dumai River estuary, Sumatra, Indonesia, a tropical black-

water river
M. Alkhatib1 and T. C. Jennerjahn2
Center for Tropical Marine Ecology, Fahrenheitstrasse 6, 28359 Bremen, Germany

Joko Samiaji
University of Riau, Jl. Simpang Panam Km 12.5, Pekanbaru, Riau, Indonesia

Abstract
The biogeochemistry of the Dumai River estuary in eastern Sumatra, Indonesia, was studied in order to obtain
information on the sources, transformation, and fate of organic matter. Between October and December 2003,
water, total suspended matter (TSM), and sediments were sampled along a salinity gradient during four
campaigns, and plants and soils were collected from the catchment. Water samples were analyzed for dissolved
inorganic nutrients and dissolved organic carbon (DOC). The concentrations of organic carbon (Corg) and total
nitrogen (N) and the stable carbon (d13Corg) and nitrogen (d15N) isotope distributions were determined in TSM,
sediments, plants, and soils. The pH as well as the concentrations of dissolved inorganic nutrients and TSM were
very low in the river and increased toward the sea. A maximum DOC concentration of 5,050 mmol L21 was
measured in the river, and concentrations decreased toward the sea. Low-gradient relief and a dense vegetation
cover, and hence little weathering and erosion, appear to be responsible for low river loads of dissolved nutrients
and TSM in this black-water river. Leaching from extensive peat soils in its catchment may account for the high
DOC content of the Dumai River. Peat swamps drained by numerous small rivers are estimated to cover 3.3 3
104 km2 in eastern Sumatra, suggesting that leaching of DOC may be a significant source of carbon to the
adjacent coastal seas. A comparison with ‘‘normal’’ rivers shows that black-water rivers can export similar
amounts of DOC from catchments that are orders of magnitude smaller. Thus, export from small black-water
rivers may be quantitatively more significant for the global DOC input into the ocean than previously thought.

Tropical rivers contribute about 60% of water, sediment, rivers are characterized by a low pH, low concentrations of
and organic carbon input into the ocean. Indonesia is dissolved inorganic nutrients and suspended sediments, and
among the regions with maximum weathering and erosion high amounts of dissolved organic carbon (DOC) (e.g.,
rates, and hence maximum dissolved and particulate river Vegas-Vilarrubia and Rull 1988). The almost exclusively
fluxes, on Earth (Ludwig et al. 1996; Gaillardet et al. 1999; organic nature of the soil and low mineral content are
Syvitski et al. 2005). However, as yet, little attention has responsible for the acidic character and low suspended
been paid to the leaching of dissolved organic matter from load. The South American Rio Negro and Caroni are the
lowland peat soils in the tropics. The Indonesian island of most prominent examples of this river type. They are
Sumatra has both high mountains and steep gradients with tributaries of two of the largest rivers in the world, the
high weathering and erosion rates along its western coast Amazon and Orinoco. Mostly, black-water rivers are small
and lowlands covered by extensive peat swamps with low streams that have been considered quantitatively insignif-
weathering and erosion rates on its eastern side. On the one icant for carbon input into the ocean. However, it has been
hand, tropical peatlands are one of the largest terrestrial suggested that the release of DOC from peatlands may
carbon stores, but on the other hand, they export more change dramatically during times of global warming and
organic carbon per unit area than any other significant changes in moisture distribution (Freeman et al. 2001a;
biogeographical land type in the world (Freeman et al. Tranvik and Jansson 2002). It is conceivable that black-
2001a; Page et al. 2002). In many cases, peatlands are water rivers are more important for carbon cycling than
drained by so-called black-water rivers. These dark-colored previously thought. With regard to this, the existing
knowledge on the biogeochemistry of black-water rivers is
astoundingly small.
1 Present address: Centre GEOTOP-UQAM-McGill, Université
Here, we present results of a biogeochemical study carried
du Québec à Montréal, CP 8888, Montréal, Québec H3C 3P8, out in the Dumai River estuary, Sumatra, Indonesia. It is
Canada
2 Corresponding author (phone +49-421-2380044, fax +49-421- a small black-water river that drains the vast lowland peat
2380030, e-mail [email protected]). soils of eastern Sumatra. The biogeochemical characteristics
of the Dumai River are compared to those of other black-
Acknowledgments water rivers, and the role of the latter for DOC export into
We thank students of the University of Riau Luffi Hedayat, the ocean is examined.
Lia Melintan, Yudi, Musni, and Eboo for help during fieldwork,
Dorothee Dasbach, Matthias Birkicht, and Ole Morisse for
laboratory work, Antje Baum for helpful comments and providing Materials and methods
a map, and two anonymous reviewers for thoughtful comments.
M. Alkhatib gratefully acknowledges financial support through Study area—The Dumai River is located in the province
a German Academic Exchange Service (DAAD) scholarship. of Riau on the Indonesian island of Sumatra (Fig. 1). It is
2410
 19395590, 2007, 6, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.2007.52.6.2410 by Nat Prov Indonesia, Wiley Online Library on [17/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dumai River estuary biogeochemistry 2411

Fig. 1. Map of the investigated area in the province of Riau, Sumatra, Indonesia.

a lowland river with low steepness and has an approximate analysis of dissolved inorganic nutrients, 25 mL of filtered
length of 12 km and average depth of 6 m. Tropical peat water was preserved with HgCl2 (20 g L21) and kept frozen
swamp and lowland forest form the major vegetation in the in polyethylene vials. Surface sediment samples from the
river catchment. Small patches of land near the coast are estuary were collected during the first sampling campaign
used for agriculture, the major part of which is used for the with a simple self-made grab.
cultivation of rice and coconut palms. The estuary is
fringed by a mangrove forest. The major soil type in the Physicochemical and current velocity measurements—
catchment is tropical forest peat (Laumonier 1997). Conductivity, dissolved oxygen (DO), temperature, and
Consequently, the river’s water is dark brown to black. pH were measured using WTW LF196, WTW Oxi96, and
The Dumai River flows through the city of Dumai and Knick 913XPH meters, respectively. Water depth was
discharges into Rupat Strait, the passage between the measured with a handheld echosounder (Plastimo Echotest
islands of Sumatra and Rupat. It is strongly influenced by II). Current velocity was determined using an Ekman style
the tides, the amplitude of which varies between 0.5 m and 231WA300 GEMWARE current meter. The following
3 m between neap and spring tide (mesotidal range). formula was used to calculate current velocity: V 5
Climate of the region is governed by the monsoons. Due
to its equatorial position, seasonality is weak. The annual
average precipitation in Dumai is 2,500 mm, and the major
rainy periods occur between March and May and
September and November (Fig. 2).

Sampling—Four sampling campaigns were carried out in

the Dumai River and its estuary between October and
December 2003. Sampling stations were distributed along
a salinity gradient from zero salinity at spring high tide at
steps of ,5 units toward the sea.
The Dumai River estuary was sampled for water,
suspended matter, and sediments. Plants and soils of the
potential terrestrial organic matter sources were also
sampled, i.e., mangrove, rice fields, Nypa palms. Water
was filtered through precombusted (10 h, 450uC) Whatman
GF/F filters. Filters, sediments, plants, and soil samples
were dried at 50uC. For DOC analysis, 10 mL of water was Fig. 2. Average monthly precipitation at CALTEX Meteo-
filtered into precombusted (4 h, 450uC) glass ampoules, rological Station, Dumai airport, in 2003. The black bar denotes
acidified to pH 2 with 1 mol L21 HCl, and then sealed. For the sampling period from 13 October to 15 December 2003.
 19395590, 2007, 6, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.2007.52.6.2410 by Nat Prov Indonesia, Wiley Online Library on [17/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2412 Alkhatib et al.

0.257N + 0.0039, where V is current velocity (m s21), and N salinity to a maximum of 8.7 at the seawater end. From
is the number of propeller revolutions per second. river to sea, nitrate increased from ,0.5 mmol L21 to ,4.5
mmol L21, phosphate decreased from a maximum of 3.2
Analyses—Dissolved inorganic nutrients (NO { {
2 , NO 3 , mmol L21 to 0.5 mmol L21, and silicate increased from 0.5
z 3{ {
NH 4 , PO 4 , and Si[OH] 4 ) were determined using mmol L21 to 25.8 mmol L21 (Fig. 3). Nitrite was not
a Skalar-SAN-plus continuous-flow autoanalyzer and detectable, and ammonium, with a maximum of 0.7 mmol
detected spectrophotometrically as a colored complex L21 in the river, decreased toward the sea. In a similar
(Grasshoff et al. 1999). To correct for the water color, manner, DO concentration and saturation increased from
absorption of the samples was measured without chemicals minimum values of 1.9 mg L21 and 23.9% to 7.0 mg L21
first, and the values were subtracted from those obtained and 98.0%, respectively, while DOC decreased from 5,050
with chemicals. Precision of the method was better than mmol L21 to 600 mmol L21 from river to sea (Fig. 4).
3.4%, and all nutrient concentrations were near the
detection limit. DOC was measured by high-temperature Suspended matter concentration and its biogeochemical
catalytic oxidation (680uC) with a Dohrmann Rosemount composition—The concentration of total suspended matter
DC-190 instrument equipped with a standard catalyst that (TSM) varied between 5.3 mg L21 and 6.8 mg L21 at zero
consisted of Al2O3 particles with 0.5% Pt (Skoog et al. salinity and did not display any trend with salinity
1997). Precision of the method was better than 1%. (Table 1). Maximum TSM concentration was observed in
Soils, sediments, and plant samples were crushed and the salinity range between 10 and 20 and, in particular,
homogenized in an agate mortar and pestle prior to during the first sampling campaign at spring tide, during
analysis. Total carbon, Corg, and total N analyses were which a maximum tidal amplitude of 2.5 m was recorded
carried out by high-temperature oxidation in a Carlo Erba between high and low tide. Maximum Corg and N ranged
NA 2100 elemental analyzer (Verardo et al. 1990). Corg was between 9.0% and 12.2% and 0.8% and 1.2% of TSM,
determined similarly after removal of carbonate by respectively, at zero salinity and decreased drastically
acidification with 1 mol L21 HCl and subsequent drying toward the sea. d13Corg varied between 228.7% and
at 40uC. Inorganic carbon was calculated by subtraction of 227.4% in the river and increased moderately toward the
organic carbon from total carbon. Precision of the method sea. d15N varied between 0.7% and 7.3% and did not
was better than 4.5% for carbon and better than 8% for display any trend with salinity except for a narrowing of the
nitrogen. d15N was determined in a Finnigan Delta Plus range from river to sea.
gas-isotope-ratio mass spectrometer after high-temperature
combustion in a Flash 1112 EA elemental analyzer. d15N is Biogeochemical composition of sediments, plants,
given as % deviation from the nitrogen isotope composi- and soils—In sediments, maximum concentrations of Corg
tion of atmospheric air. d13Corg was determined similarly and N of 11.2% and 0.3%, respectively, were measured at
after removal of carbonate by adding 1 mol L21 HCl and Sta. 1 in the river. Seaward, Corg varied between 1.2% and
subsequent drying at 40uC. d13Corg is given as % deviation 5.0%, and N varied between ,0.1% and 0.3% without
from the carbon isotope composition of the Peedee displaying any trend. d13Corg ranged between 229.6% and
belemnite (PDB) standard. The standard deviation of 227.2%, and d15N ranged between 2.4% and 4.7%, which
replicate measurements was 0.2%. is in the range of the plants and soils that form the major
part of the land cover/use (Table 1).
Discharge calculation—Cross-sectional area of the river
was calculated from measurements of its width and depth Discussion
with a handheld echosounder along the cross-sectional
river profile. River discharge (Q) was then calculated from The very low pH, low concentrations of dissolved
current velocity (CV ) and cross-sectional area (CSA) using nutrients and TSM, and the high DOC content qualify
the formula: Q (m3 s21) 5 CV (m s21) 3 CSA (m2). the Dumai River as a typical black-water river. The sources
Discharge measurements were not corrected for tidal and fate of the dissolved and particulate river loads are
currents. However, the tidal component should be small discussed below.
because discharge measurements were performed during
slack high tide. Sources and fate of suspended and sedimentary
organic matter—The SE Asian/western Pacific region is
Results generally characterized by high weathering and erosion
rates and hence high river loads of suspended sediments
The results of the physicochemical measurements and (e.g., Milliman and Meade 1983; Gaillardet et al. 1999;
biogeochemical analysis conducted along a salinity gradi- Milliman et al. 1999). When compared to those, however,
ent displayed little temporal variation over the four weathering and erosion are minimal in the Dumai River
sampling campaigns. catchment due to the low gradient, low river discharge, and
the large surface covered by peatlands, which results in the
Physicochemical data, dissolved nutrients, and DOC— observed low TSM concentration in the Dumai River.
Water temperature ranged between 27.9uC and 32.0uC and Despite a slight increase of d13Corg toward the sea, which
generally displayed a slight increase toward the sea. The pH points to a small contribution of marine organic matter
displayed a strong gradient, from a minimum of 4.0 at zero (OM), suspended particulate OM appears to originate
 19395590, 2007, 6, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.2007.52.6.2410 by Nat Prov Indonesia, Wiley Online Library on [17/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dumai River estuary biogeochemistry 2413

Fig. 3. The (A) pH, (B) dissolved nitrate, (C) dissolved phosphate, and (D) dissolved silicate
concentrations vs. salinity in the Dumai River estuary.

mainly from the surrounding vegetation and soils, as the C : N ratio of sediments is higher by a factor of two over
indicated by a C : N ratio basically between 10 and 20 and that of TSM (Table 1). Preferential decomposition of labile
a d13Corg value between 229% and 227%. While d13Corg nitrogenous OM compounds, such as amino acids, for
values of sediments and TSM fall within the same range, example, may be responsible for the high C : N ratio (e.g.,

Fig. 4. Concentrations of (A) DO and (B) DOC vs. salinity in the Dumai River estuary.
 19395590, 2007, 6, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.2007.52.6.2410 by Nat Prov Indonesia, Wiley Online Library on [17/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2414 Alkhatib et al.

Table 1. Biogeochemical characteristics of TSM and Ocean-derived nitrate appears to be the source for the very
sediments from the Dumai River estuary and plants and soils low autochthonous estuarine OM production. Phosphate
from the river catchment. Data for TSM and sediments displayed concentration in the Dumai River is in the same range as in
here are from samples obtained during the first sampling the Brantas River and in other mangrove creeks and
campaign. Coordinates of stas. 1–7: 101u25.9099E, 01u40.2609N; estuaries (e.g., Alongi et al. 1992; Dittmar and Lara 2001;
101u25.7779E, 01u40.5049N; 101u25.8449E, 01u40.6099N; Jennerjahn et al. 2004). In contrast to nitrate, phosphate
101u25.8919E, 01u40.6409N; 101u26.1219E, 01u40.7989N; decreases from maximum values in the river with increasing
101u26.1309E, 01u40.8149N; 101u26.0749E, 01u40.9309N.
salinity toward the sea (Fig. 3). Sulfate reduction in the
TSM N Corg d13Corg d15N suboxic or anoxic peat swamps may have bound iron in
Sample Salinity (mg L21) (%) (%) C : N (%) (%) sulfides that originally were bound in iron-phosphorus
compounds. This mechanism has been observed in freshwa-
TSM ter wetlands (e.g., Lucassen et al. 2004). The maximum
Sta. 1 0.0 5.9 0.9 10.9 12.1 228.7 0.7
Sta. 2 6.7 74.2 0.3 3.8 12.7 228.0 2.5
phosphate and minimum DO content of river water at zero
Sta. 3 13.7 164.5 0.2 3.1 15.5 228.0 5.1 salinity indicates that phosphate may have been released
Sta. 4 15.8 209.2 0.2 2.9 14.5 228.0 6.2 from the peat swamps by this mechanism.
Sta. 5 18.9 136.7 0.1 2.4 24.0 228.0 5.1 The extremely high DOC concentration of .5,000 mmol
Sta. 6 21.3 81.8 0.1 2.1 21.0 227.7 4.1 L21 in the dark-colored water of the Dumai River is the
Sta. 7 26.7 88.0 0.1 2.7 27.0 227.8 5.2 highest measured in world rivers so far (e.g., Degens et al.
Sediments 1991; Table 2). Physicochemical characteristics of the high-
Sta. 1 0.0 0.3 11.2 37.3 229.5 3.0 DOC water are important factors for the production and
Sta. 2 6.7 0.2 4.6 27.9 227.8 2.5 decomposition of OM. Leaching from litter and peat soils is
Sta. 3 13.7 0.0 1.4 33.0 227.4 4.4 the major source of dissolved organic matter (DOM) in the
Sta. 4 15.8 0.1 4.2 32.1 228.2 2.4 Dumai River. In general, carbohydrates and amino acids are
Sta. 5 18.9 0.0 1.3 30.6 227.5 4.7 major labile compounds of DOM, while humic and fulvic
Sta. 6 21.3 0.1 5.0 39.2 229.6 3.1 substances make up the more refractory portion (Schlesinger
Sta. 7 26.7 0.1 1.2 24.4 227.2 4.5
1991). The humic and fulvic substances on average make up
Plants and soils 60–80% of riverine DOM, and the maximum portions are
Mangrove 1.3 44.3 34.1 227.5 2.9 found in black-water rivers and swamps (Moran and
leaf
Hodson 1990; Leff and Meyer 1991; Spitzy and Leenheer
(green)
Mangrove 0.7 43.5 64.6 227.9 3.9 1991). High concentrations of phenolic compounds usually
leaf found in peatlands inhibit biodegradation by reducing
(fallen) activity of biodegradative hydrolase enzymes. This is of
Rice plant 2.0 38.9 19.9 229.0 7.9 particular importance under anoxic conditions. Activity of
Nypa fruit 1.4 44.3 31.2 226.2 4.1 the enzyme phenol oxidase, which requires bimolecular
Nypa leaf 1.2 44.8 38.6 226.7 0.7 oxygen, is severely constrained under anoxic conditions.
Rice soil 0.1 1.8 34.3 223.9 4.3 Therefore, the lack of oxygen prevents phenol oxidase from
Mangrove 0.1 3.5 19.4 227.8 3.4 eliminating phenolic compounds, which, in turn, reduces
soil biodegradation of OM substantially (Wetzel 1992; Vuorinen
and Saharinen 1996; Freeman et al. 2001b). These processes
in combination with the low river discharge may explain the
Jennerjahn and Ittekkot 1999). It indicates that sedimen- extremely high DOC concentration in the vicinity of this
tary OM has undergone severe degradation. d15N is almost anoxic peat swamp forest, which is inundated during most
similar in TSM and sediments and falls in the range of the times of the year.
potential terrestrial sources (Table 1). It is also similar to In the river, maximum DOC concentrations coincide
the d15N of TSM and sediments from the Brantas River, with minimum DO concentration and saturation of 2–3 mg
a river on the Indonesian island of Java that has a high load L21 and of 20–40%, respectively. Physical mixing of river
of suspended sediments mainly originating from agricul- and seawater appears to be a major factor for the decrease
tural soils (Jennerjahn et al. 2004). The relatively high TSM of DOC and increase of DO toward the sea. However,
concentration in the estuarine mixing zone of the Dumai a nonlinear relationship between these indicates that the
River appears to result almost exclusively from resuspen- decrease of DOC cannot be explained solely with conser-
sion. The low Corg content of TSM there is probably vative mixing. A statistically significant relationship (expo-
a mixture of recycled sedimentary OM with minor nential decay) suggests DOM degradation in the estuary
contributions of freshly produced autochthonous OM. (Fig. 5). The tidal input of bimolecular oxygen may have
activated phenol oxidase to eliminate phenolic compounds,
Sources and fate of dissolved nutrients and DOC— which, in turn, may have resulted in degradation of the
Extremely low nutrient concentrations as a result of low DOM leached from the anoxic peat swamps. Biodegrad-
weathering and erosion are a typical feature of black-water ability of vegetation- and soil-derived DOC has been
rivers. Salinity-related gradients of pH and nutrients indicate demonstrated in a tropical wet forest in Costa Rica. DOC
that conservative mixing is a major determinant of the concentrations of up to 13 mg L21 were measured in
observed distribution patterns in the Dumai River estuary. throughfall (rainwater falling through tree canopy), litter
 19395590, 2007, 6, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.2007.52.6.2410 by Nat Prov Indonesia, Wiley Online Library on [17/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dumai River estuary biogeochemistry 2415

Table 2. Biogeochemical characteristics of black-water rivers. References: *, this study; 1, Baum et al. (2007); 2, Valentine and Zepp
(1993); 3, Leff and Meyer (1991); 4, Vegas-Vilarrubia and Rull (1988); 5, Castillo et al. (2004); 6, Richey et al. (1990); 7, Hastenrath et al.
(1999). DIN, dissolved inorganic nitrogen; #, nitrate only.

DOC Silicate (mmol Phosphate DIN

River pH (mmol L21) L21) (mmol L21) (mmol L21) Ref.
Dumai, Sumatra, Indonesia 4.3 4,983 0.7 1.7–3.2 1.0 *
(averages at Sta. 1)
Siak, Sumatra, Indonesia 5.2–7.8 554–2,594 1.6–89.1 0.2–36.7 7.9–67.9 1
Mandau (tributary of Siak) 4.4–4.9 1,939–3,568 2.9–49.9 0.6–15.4 5.8–28.1 1
Suwannee, Georgia, USA 5.5–6.5 3,920 2
Ogeechee, Georgia, USA 417–1,667 3
Orinoco tributaries
San Jaime 4.6 783 1.4 4
Cauo Buja 5.2 792 3.6 4
Caroni 5.1 442 6.1 4
Autana, Venezuela 4.7 658 ,0.1 2.9# 4, 5
Sipapo 4.9 542 4.1 0.1 2.6# 4, 5
Atabapo 4.0 450 3.4 4
Chola 3.8 625 5.3 4
Loro 3.7 633 5.9 4
Darigua 3.7 675 4.9 4
Uniabo 3.7 617 5.1 4
Negro 3.9 633 7.2 4
Guiania 3.9 600 5.2 4
Amazon tributary
Rio Negro 4.6 542–883 57.0 6, 7

leachate, and soil solution, of which 23–46% was biode- catchment, it contributes 32.3% of its DOC load. Similarly,
graded within 7 d. There, the soil solution had a range of the Caroni makes up 9.3% of the Orinoco’s catchment, but
41–46% and displayed the maximum biodegradable frac- contributes 17.0% of its DOC load (Vegas-Vilarrubia and
tion of DOC (Schwendenmann and Veldkamp 2005). In the Rull 1988; Richey et al. 1990; Hastenrath et al. 1999). On
Dumai River estuary, the DOC concentration at the a global scale, riverine DOC input into the ocean tends to
seaward end was still in the range of 600 mmol L21 despite increase with catchment size (Ludwig et al. 1996). The same
substantial degradation. This is much higher than in appears to hold true for black-water rivers. When compared
numerous rivers from all climatic zones around the world to ‘‘normal’’ rivers, however, black-water rivers export
(e.g., Ludwig et al. 1996). similar amounts of DOC from catchments that can be

Relevance of black-water rivers for DOC export to

the ocean—Information on the biogeochemistry of black-
water rivers is scarce, and so their possible quantitative
significance for the oceanic DOC budget is as yet un-
explored. For this purpose, we calculated DOC export for
the Dumai River and the few other black-water rivers for
which the respective information was available from the
literature. Our budget relied on simple short-term discharge
calculations without correction for the tidal current. Due to
these constraints, our calculated DOC export may be an
overestimate. We measured a discharge of 16 m3 s21, which
results in an annual total of 0.5 km3 yr21. Multiplied by the
mean concentration of 59.8 mg L21, DOC export from the
Dumai River is estimated to be 30.2 3 106 kg yr21. Even
though this may be an overestimate, its total export is very
low on a global scale, even when compared to other black-
water rivers.
Two examples from South America highlight the relative
significance of black-water rivers for fluvial DOC input into
the oceans. The Rio Negro and Caroni are black-water
tributaries of the Amazon and the Orinoco, respectively. Fig. 5. DOC vs. DO in the Dumai River estuary. DOC 5
While the Rio Negro makes up only 12.2% of the Amazon’s 11,016 3 e20.36DO, r 5 0.99, n 5 22.
 19395590, 2007, 6, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.2007.52.6.2410 by Nat Prov Indonesia, Wiley Online Library on [17/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2416 Alkhatib et al.

Fig. 6. DOC export vs. catchment size for several black-water rivers and some world rivers.
Data are from Vegas-Vilarrubia and Rull (1988), Richey et al. (1990), Leff and Meyer (1991),
Valentine and Zepp (1993), Ludwig et al. (1996), Hastenrath et al. (1999), McCallum and Hickey
(2001), World Resources Institute (2003), Castillo et al. (2004), and Baum et al. (2007). No
official data are available on the catchment size of the Dumai River. The estimate used was
derived from a map and therefore has a high uncertainty.

orders of magnitude smaller, particularly in the case of small DITTMAR, T., AND R. J. LARA. 2001. Driving forces behind nutrient
rivers that drain peatlands (Fig. 6). This indicates that the and organic matter dynamics in a mangrove tidal creek in
numerous small black-water rivers that drain the 3.3 3 North Brazil. Estuar. Coast. Shelf Sci. 52: 249–259.
104 km2 of peatlands in eastern Sumatra (Thia-Eng et al. FREEMAN, C., C. D. EVANS, D. T. MONTELTH, B. REYNOLDS, AND
N. FENNER. 2001a. Export of organic carbon from peat soils.
2000) are very efficient DOC sources for the adjacent coastal
Nature 412: 785.
sea. On a global scale, it is conceivable that DOC export ———, N. OSTLE, AND H. KANG. 2001b. An enzymic ‘latch’ on
from black-water rivers is quantitatively more significant a global carbon store. Nature 409: 149.
than previously thought. This is corroborated by findings of GAILLARDET, J., P. LOUVAT, B. DUPRE, AND C. ALLEGRE. 1999.
Baum et al. (2007), who estimated the DOC export of Global silicate weathering and CO2 consumption rates
Indonesian rivers draining peatlands to be on the order of 17 deduced from the chemistry of large rivers. Chem. Geol.
3 109 kg yr21, which is about 8% of the global total DOC 159: 3–30.
input into the ocean. GRASSHOFF, K., M. EHRHARDT AND K. KREMLING [EDS.]. 1999.
Methods of seawater analysis. Wiley-VCH.
References HASTENRATH, S., L. GREISCHAR, E. COLON, AND A. GIL. 1999.
Forecasting the anomalous discharge of the Caroni River,
ALONGI, D. M., K. G. BOTO, AND A. I. ROBERTSON. 1992. Nitrogen Venezuela. J. Climate 12: 2673–2678.
and phosphorous cycles, p. 251–292. In A. I. Robertson and JENNERJAHN, T. C., AND V. ITTEKKOT. 1999. Changes in organic
D. M. Alongi [eds.], Tropical mangrove ecosystems: Coastal matter from surface waters to continental slope sediments off
and estuarine studies 41. American Geophysical Union. the São Francisco River, eastern Brazil. Mar. Geol. 161:
BAUM, A., T. RIXEN, AND J. SAMIAJI. 2007. Relevance of peat 129–140.
draining rivers in central Sumatra for riverine input of ———, ———, S. KLÖPPER, SENO ADI, SUTOPO PURWO NUGROHO,
dissolved organic carbon into the ocean. Estuar. Coast. Shelf NANA SUDIANA, ANYUTA YUSMAL, PRIHARTANTO, AND B.
Sci. 73: 563–570. GAYE-HAAKE. 2004. Biogeochemistry of a tropical river
CASTILLO, M., D. ALLAN, AND G. KLING. 2004. Seasonal and affected by human activities in its catchment: Brantas River
interannual variation of bacterial production in lowland rivers estuary and coastal waters of Madura Strait, Java, Indonesia.
of the Orinoco Basin. Freshwater Biol. 49: 1400–1414. Estuar. Coast. Shelf Sci. 60: 503–514.
DEGENS, E. T., S. KEMPE AND J. E. RICHEY [EDS.]. 1991. Biogeo- LAUMONIER, Y. 1997. The vegetation and physiography of
chemistry of major world rivers. SCOPE 42. Wiley. Sumatra. Kluwer.
 19395590, 2007, 6, Downloaded from https://aslopubs.onlinelibrary.wiley.com/doi/10.4319/lo.2007.52.6.2410 by Nat Prov Indonesia, Wiley Online Library on [17/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Dumai River estuary biogeochemistry 2417

LEFF, L. G., AND J. L. MEYER. 1991. Biological availability of SKOOG, A., D. THOMAS, R. J. LARA, AND K. RICHTER. 1997.
dissolved organic carbon along the Ogeechee River. Limnol. Methodological investigations on DOC determinations by
Oceanogr. 36: 315–323. HTCO method. Mar. Chem. 56: 39–44.
LUCASSEN, E. C. H. E. T., A. J. P. SMOLDERS, J. VAN DE SPITZY, A., AND J. LEENHEER. 1991. Dissolved organic carbon in
CROMMENACKER, AND J. G. M. ROELOFS. 2004. Effects of rivers, p. 213–232. In E. T. Degens, S. Kempe and J. E. Richey
stagnating sulphate-rich groundwater on the mobility of [eds.], Biogeochemistry of major world rivers. Wiley and Sons.
phosphate in freshwater wetlands: A field experiment. Arch. SYVITSKI, J. P. M., C. J. VÖRÖSMARTY, A. J. KETTNER, AND P.
Hydrobiol. 160: 117–131. GREEN. 2005. Impact of humans on the flux of terrestrial
LUDWIG, W., J. L. PROBST, AND S. KEMPE. 1996. Predicting the sediment to the global coastal ocean. Science 308: 376–380.
oceanic input of organic carbon by continental erosion. Glob. THIA-ENG, C., I. R. L. GORRE, S. A. ROSS, S. R. BERNAD, B.
Biogeochem. Cycles 10: 23–41. GERVACIO, AND M. C. EBARVIA. 2000. The Malacca Strait.
MCCALLUM, B. E., AND A. C. HICKEY. 2001. Water resources data Mar. Poll. Bull. 41: 160–178.
Georgia, 2000.Volume 1: Continuous water-level, stream- TRANVIK, L. J., AND M. JANSSON. 2002. Terrestrial export of
flow, water-quality data, and periodic water-quality data, organic carbon. Nature 415: 861–862.
water year 2000. US Geological Survey Water-Data Report VALENTINE, R. L., AND R. G. ZEPP. 1993. Formation of carbon
GA-00-1. monoxide from the photodegradation of terrestrial dissolved
MILLIMAN, J. D., K. L. FARNSWORTH, AND C. S. ALBERTIN. 1999. organic carbon in natural waters. Environ. Sci. Technol. 27:
Flux and fate of fluvial sediments leaving large islands in the 409–412.
East Indies. J. Sea Res. 41: 97–107. VEGAS-VILARRUBIA, T., AND V. RULL. 1988. A multivariate
statistical classification of Venezuelan blackwater rivers. Mitt.
———, AND R. H. MEADE. 1983. World-wide delivery of river
Geol.-Paläont. Inst. Univ. Hamburg, SCOPE/UNEP Sonder-
sediment to the oceans. J. Geol. 91: 1–21.
band Heft 66: 51–61.
MORAN, M. A., AND R. E. HODSON. 1990. Bacterial production on
VERARDO, D. J., P. N. FROELICH, AND A. MCINTYRE. 1990.
humic and nonhumic components of dissolved organic Determination of organic carbon and nitrogen in marine
carbon. Limnol. Oceanogr. 35: 1744–1756. sediments using the Carlo Erba NA-1500 Analyzer. Deep-Sea
PAGE, S. E., F. SLEGERT, J. O. RIELEY, H.-D. V. BOEHM, ADI JAYA, Res. I 37: 157–165.
AND SUWIDO LIMIN. 2002. The amount of carbon released
VUORINEN, A. H., AND M. H. SAHARINEN. 1996. Effects of soil
from peat and forest fires in Indonesia during 1997. Nature organic matter extracted from soil on acid phosphomonoes-
420: 61–65. terase. Soil Biol. Biochem. 28: 1477–1481.
RICHEY, J. E., J. I. HEDGES, A. H. DEVOL, P. D. QUAY, R. WETZEL, R. G. 1992. Gradient-dominated ecosystems—sources
VICTORIA, L. MARTINELLI, AND B. R. FORSBERG. 1990. and regulatory functions of dissolved organic matter in
Biogeochemistry of carbon in the Amazon River. Limnol. freshwater ecosystems. Hydrobiologia 229: 181–198.
Oceanogr. 35: 352–371. WORLD RESOURCES INSTITUTE. 2003. Water Resources eAtlas
SCHLESINGER, H. W. 1991. Biogeochemistry: An analysis of global [Internet]. World Resources Institute, Washington D.C.
change. Academic Press. Available from: www.iucn.org/themes/wani/eatlas/.
SCHWENDENMANN, L., AND E. VELDKAMP. 2005. The role of
dissolved organic carbon, dissolved organic nitrogen, and Received: 22 February 2007
dissolved inorganic nitrogen in a tropical wet forest ecosys- Accepted: 23 June 2007
tem. Ecosystems 8: 339–351. Amended: 9 July 2007