PARTICLE DISTRIBUTION AND BIO-OPTICAL PROPERTIES IN THE
EASTERN CARIACO BASIN, VENEZUELA: INFLUENCE OF LOCAL
RIVERS
ABSTRACT
[you can’t study temporal variability with only one snap shot] Biogeochemical
oceanographic data of the Eastern Cariaco Basin, located along the southern margin of the
Caribbean Sea, during the wet season (September 2003) were collected and analyzed to
determine the influence of local rivers . The two main contributors of colored dissolved organic
matter (CDOM) and terrigenous particles to the study region were the Neverí and Unare rivers,
with no indication of the Orinoco or the Amazon River influence. The influence of local rivers
was mainly restricted in a narrow (30-40 km) band along the southern coast, where over 80% of
the total absorption coefficient at 443 nm was contributed by detrital (ad443: ?-?) and dissolved
matter (ag443: ? – 0.15 m-1) , which decreased rapidly further offshore [this is redundant,
deleted] Along a salinity gradient of 35.8 to 37.0 [check these numbers] from inshore to offshore
waters, ag443, ad443, and cp660 (beam attenuation coefficient, a measure of particle
concentration) varied between ?-?, ?-?, and ?-?, respectively. The sediment input of the Unare
and Neverí Rivers affected mostly the shelf nearby, and their low salinity plumes were carried
northwestward toward the CARIACO time series station. [now, this is extremely important and
also the major goal of your study: Did the CARIACO station experience any riverine influence,
from either particle or low-salinity plume?]
[You did two cruises. The results here are from one cruise only? Why?]
1. Introduction
Continental margins play an important role in the global cycles of carbon and other
biogeochemical elements (Muller-Karger et al., 2005). Land materials enter the sea mainly
through river runoff in particulate or dissolved form (Liu et al., 2000), and some fraction is
deposited on the adjacent continental shelf. The delivery of terrigenous materials via rivers varies
according to changes in weather patterns, such as seasonal rainfall and ocean circulation.
Ultimately, climate changes have impacts on all of these processes.
The Cariaco Basin is located on the continental shelf off Venezuela (Figure 1), and is
composed of two depressions of ~ 1400 m each and connected to the Caribbean Sea by a shallow
sill (~140 m maximum depth; Richards, 1975). [need to add arrows on the figure to show where
are the two depressions, and where is the sill] Because of its anoxic condition and high
sedimentation rate, the basin conserves one of the best sediment records available in the marine
environment (Hughen et al., 1996; Lin et al., 1997; Black et al., 1999; Peterson el al. 2000; Goñi
et al., 2003). In order to carry out an accurate interpretation of past climate variation based on
this record, it is necessary to understand the sources of particulate matter to the basin. Three
main local rivers discharge onto the southern Cariaco shelf known as the Unare Platform: the
Manzanares, Neverí and Unare Rivers (Figure 2), all originating in the nearby Coastal Mountain
Range (Cordillera de la Costa). These rivers show high discharge during the rainy season (July to
November) with maximum discharge in August and September. The Unare is the largest among
1
these rivers, with a drainage basin area of 22.3 Km2 and an approximate average discharge rate
of 56 m3/sec (Zinck, 1977). The Neverí and Manzanares are smaller (drainage basin of 3.9 and
1.0 km2, respectively), discharging approximately 35 and 22 m3/sec, respectively, into the
Cariaco Basin (Zinck, 1977). These discharge rates were measured during 1958-1967.
Unfortunately, data are not available to assess more recent discharges. Several of the rivers have
since been dammed, while others now receive contributions from sewer systems, industry and
urban areas.
[this is redundant]
The CARIACO (Carbon Retention in a Colored Ocean) oceanographic time series project
has collected over 9 years of data, including sediment traps, optics and hydrography in the
Cariaco Basin (Muller-Karger et al., 2005) at lat lon (Figs. 1 and 2). The lowest surface salinities
and highest terrigenous inputs at the time series station occur in September (Astor et al., 1998).
There has been a lingering question as to whether these changes are associated with local rivers
or the larger Orinoco plume that enters the Caribbean Sea (see Muller-Karger et al., 1989). In the
framework of the CARIACO project, we conducted extensive cruises in the rainy and dry
seasons (September 2003 and March 2004, respectively) to help understand river runoff into the
Cariaco Basin and whether small, local rivers are more influential than larger, more distant
rivers, such as the Orinoco.
2. Methods
2.1. Data collection
Two cruises were conducted to the Cariaco Basin between September 16-20 2003 and
March 15-20, 2004, using the R/V Hermano Gines of the Fundacion La Salle de Ciencias
Naturales de Venezuela (FLASA/EDIMAR). In September 2003, 39 stations were sampled, with
19 in shallow waters (<200 m) and the other 20 in a grid-like pattern (Fig. 2). Similar sampling
scheme was used for the March 2004 cruise.
At each station, salinity and temperature profiles were measured using a Seabird SBE25
CTD. Oxygen data and chlorophyll fluorescence were obtained with a Seabird SBE 43 oxygen
sensor and a Chelsea fluorometer (Chelsea, Inc.), respectively, attached to the CTD. The
ensemble also had a C-Star transmissometer (WetLabs) that measured beam attenuation
coefficient at 660 nm (cp(660), m-1).
One absorption/attenuation meter (AC-9, WetLabs) was deployed as part of the
hydrocasts during September 2003. A pump was used to draw water through the absorption and
attenuation tubes. At 14 stations, double casts were performed, one with an unfiltered AC-9, and
the second using a 0.2 m filter (Propor PES filter capsules) attached to the inlet of the
absorption tube. The instrument was calibrated before, during, and after the cruise using distilled
water as a reference.
Water samples from some selected stations and depths were filtered with a 0.2 m poresize anotop filter to determine the absorption coefficients of CDOM.
2.2. Data processing
2
AC-9 data were corrected for temperature and salinity effects, following Pegau et al.
(1997). A scattering correction was also applied to the absorption data by subtracting a reference
value at 715 nm from all data at other wavelenths (hypothesis 3 in the AC-9 WetLabs User’s
Guide, 2003). The good agreement between AC-9 beam attenuation coefficient at 650 nm
(cp(650)) and C-Star transmissometer data (cp(660)) (r =?, N=?, slope=?, intercept =?) suggested
high data quality. [You didn’t mention the C-Star transmissometer in the data collection section]
The spectral shape () of the beam attenuation coefficient, as an measure of particle size
distribution (Diehl and Haardt, 1980) was calculated as follows (Boss and Zeneveld, 2003):
cp() = cp(o)*(o)-

CDOM Absorbance (A) spectra were measured between 200 and 800 nm with a dual
fiber optic spectrometer (Ocean Optics) equipped with 10-cm quartz cuvettes. The detection
limit of the spectrophotometer was ± 0.002, equivalent to absorption coefficients of 0.046 m-1.
[note that in Fig. 5 many values are below this limit – do you still trust the S data?] Distilled
water was used as a blank. CDOM absorption coefficients were calculated as:
ag() = ln 10 A()/r
(2)
where “r” is the pathlength (0.1 m). ag() has been reported to decrease exponentially with
increasing wavelength as:
ag() = ag(o)exp[–S(o)]
(3)
where is a reference wavelength and S (nm-1) is the spectral slope (Blough et al, 1993; Green
and Blough, 1994; Ferrari, 2000, Blough and Del Vecchio, 2002). A non-linear least square fit
method was used to estimate S.
Absorption coefficients of phytoplankton pigments (aph) were not measured directly, but
rather derived from in situ fluorometric measurements, using relationships obtained by the
CARIACO project (199? – 200?). aph for the CARIACO dataset was measured in situ
through filtration following Kishino et al. (1985). The  factor from Mitchell and Kiefer
(1988) was used for the correction of the optical path elongation due to filter pads. The
relationship between aph at 440 nm and chlorophyll fluorescence was obtained from the
CARIACO time-series data::
aph(440) = 0.0955 * FChl + 0.0101
(r2 = 0.93, n=? RMS=?)
Using the equation above and fluorometric data from the September 2003 cruise, aph(440) was
obtained. [Now – how do you know if FChl from CARIACO and FChl from sept 2003 were
measured consistently? Also, in the data collection section you didn’t mention anything about
chlorophyll fluorescence measurement]
3
4
Table 1: Station location, cast and bottom depth, SSS, SST and mixed layer depth for the September 16-20, 2003
cruise.
Latitude
Longitude
Station
10.83
10.66
10.50
10.33
10.49
10.66
10.83
10.83
10.66
10.49
10.33
10.33
10.23
10.14
10.19
10.28
10.49
10.66
10.83
10.66
10.49
10.41
10.33
10.23
10.23
10.33
10.41
10.33
10.41
10.49
10.83
11.00
11.00
11.00
10.99
10.44
10.14
10.16
10.41
-64.37
-64.36
-64.36
-64.55
-64.55
-64.55
-64.55
-64.70
-64.71
-64.66
-64.71
-64.88
-64.88
-65.03
-64.79
-64.8
-64.88
-64.88
-64.88
-65.05
-65.05
-65.05
-65.05
-65.05
-65.21
-65.21
-65.21
-65.38
-65.38
-65.21
-65.04
-65.04
-64.88
-64.71
-64.54
-64.26
-65.11
-65.21
-64.40
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
48
49
50
51
Maximum
profiling
depth (m)
225
157
400
120
400
400
220
225
400
400
40
56
37
15
26
45
400
400
224
400
400
63
59
25
24
56
75
55
77
113
220
80
130
174
273
70
15
10
180
Bottom
depth (m)
235
167
1058
128
1328
422
230
237
520
1220
51
66
47
19
36
55
1100
815
234
875
547
73
69
35
34
66
84
65
87
178
229
150
135
184
283
95
15
12
190
*Und. = Unable to determine
[this table should be put in a separate file]
5
Sea
Surface
Salinity
36.79
36.76
36.74
36.56
36.73
36.76
36.76
36.71
36.73
36.56
36.58
36.58
36.55
36.63
36.01
36.55
36.61
36.60
36.66
36.64
36.63
36.67
36.63
36.54
36.59
36.64
36.65
36.62
36.66
36.66
36.65
36.69
36.68
36.70
36.86
36.76
35.30
36.07
36.62
Sea Surface
Temperature
(°C)
28.49
28.26
27.90
28.01
28.14
28.34
28.50
28.20
28.18
28.35
27.73
27.96
27.73
28.11
27.58
27.68
27.66
27.66
27.84
27.68
27.61
27.41
27.86
28.45
27.62
27.39
27.32
27.66
27.31
27.42
27.73
27.63
27.70
27.96
28.33
28.03
27.64
27.98
28.15
Mixed
layer
depth (m)
15
14
14
9
17
21
23
21
18
10
11
12
12
7
6
10
18
23
28
29
20
20
7
8
8
14
24
12
20
20
17
19
21
27
22
12
7
Und.
9
3. Results
[Without mentioning salinity, how do you know they are river plumes, not upwelled
bottom CDOM?] Highest ag(440) values were found during September 2003 at Stations 14
(0.134 m-1) and 13 (0.148 m-1) (See Figure 2 for reference), accompanied by low salinity (? How
much? Fig.6 shows station 49 instead of 13 has the lowest salinity, why?]. Also, the CDOM
absorption spectral slope, S (nm-1), was lower at these stations (give range here) than at other
offshore stations (give range here). Clearly, these high-CDOM low-salinity waters are
characteristics of river plumes. Indeed, these stations are located near the river mouths of the
Neverí and Unare. There was no low salinity or high CDOM signal to the north of the Basin
(Figure 6) [where? Give a latitude threshold or water depth threshold]. ag(440) was relatively low
(~0.04 m-1) and the water was blue at Station 48 (Figure 3), close to the ManzanaresRive.
However, traces of leaves, plastic and other types of debris were seen at this site. [Now, do you
see any influence on the CARIACO location? This should be the most important thing in the
paper]
At stations near the coast (22, 21 and 11, See Fig. 2), total absorption coefficients (water
excluded) at depths of 1, 5, 10, and 25 m (Figure 4) [if this figure is mentioned in the text later
than Fig. 6, the order should be rearranged] showed small absorption peaks at 676 nm, indicating
the presence of phytoplankton. However, the typical absorption shape of phytoplankton in the
blue part of the spectrum was masked by CDOM, where ag443 contributed to >70% of the total
absorption coefficient at 443 nm. Further [this is irrelevant to your objective] Station 21 showed
some of the strongest absorption at 10 and 25 m, which corresponded to higher chlorophyll
concentrations. [so what?] [I think Fig. 4a is enough to say what you want to say, because the
rest doesn’t show big differences. The more concise the better].
[Fig. 5 is not necessary]
Figure 6 shows the distribution of sea surface salinity, temperature, and cp(660) (m-1) in
the Eastern Cariaco Basin for September 2003 and March 2004. During the September 2003
cruise, the Unare River plume, as indicated by the low salinity (~35.4), was associated with high
cp(660) values (~ 1 m-1). Further offshore, cp(660) decreased while salinity decreased,
suggesting dispassion and/or sinking of suspended particles. Most particles associated with the
rivers precipitated from surface waters within 10 km from their mouths. [I don’t understand this.
At least up to 10.3 degree N you can still see the particle signature and this is at least 20 km
away from the river mouth] During September 2003, waters in the Basin were stratified and the
average mixed layer depth near the coast was 8.5 m. In the deeper parts of the basin (> 100 m),
the average mixed layer depth was approximately 18.6 m. (Table I). The stratified waters near
the coast enabled the river plumes to be also visible in the sea surface temperature contour of the
Basin (Figure 6). The Unare River was characterized by a plume of warmer water (~28.7°C),
possibly due to the strong light absorption and thin fresher water lens.The Neverí and
Manzanares river plumes were of colder water (~27.6°C and ~28°C, respectively) [colder? These
may be the background SST for that region].
6
During March 2004, cp(660) in the entire study area, especially in the coastal waters, was
higher, compared to September 2003, due to the enhanced primary productivity stimulated by the
seasonal upwelling [ref?]. The cold, salty, nutrient rich upwelled water was visible also in the sea
surface temperature and salinity data (Figure 6).
Figure 7 shows the distribution of cp(660) at the maximum sampling depth of each
station in September 2003 (see Table 1 for reference). High attenuation was observed throughout
the Unare Platform [where is this platform?], with higher values (~0.6 - 1 m-1) near the bottom of
stations 49 and 13 (close to the Unare River mouth). [to exclude the possibility that these are due
to phytoplankton bloom – what are the chl values at these two stations at the same bottom
depths?] This higher beam attenuation near the bottom was interpreted as bottom nepheloid
layers (BNL) (Pak et al., 1984; Boss et al., 2001; Xu et al., 2002; McPhee-Shaw et al., in press).
These BNL’s extended through station 12, to the east of the Unare River’s mouth (0.5-0.6 m-1).
Just like the Unare’s BNL, the Neveri’s was skewed to the right, towards the east, and it was less
attenuating than the Unare (0.55-0.6 m-1). The thickness of the BNL’s varied. Near the Neverí
River, the BNL ranged from 6 to 15 m, while near the Unare it was <5m. Suspended, or
intermediate nepheloid layers (INL’s), were seen at stations 15 and 11 at depths of 35 and 45 m,
respectively. [do these extend to the CARIACO station?]
Intermediate nepheloid layers were also observed in September 2003 near the
Manzanares River at station 2 (10o29’N 64o21’W), located over a submarine canyon (~0.09 m-1).
[I don’t see this from Fig. 7] During March 2004, INL’s were also observed at the same location,
though thinner and more attenuating (~ 0.15 m-1). [all sentences before this point are perhaps not
necessary] Figure 8 compares some sample profiles at Station 2 (10o29’N 64o21’W) [I told you
to be consistent in lat lon – in the graphs you used decimal degrees], where the INL’s were
observed in both September 2003 and March 2004. During September 2003, the INL was located
at ~220 m depth and had an approximate thickness of 50 m. (upper right box in figure 8). The
INL’s did not show a corresponding variation in absorption, nor were associated with the strong
pycnocline of the Cariaco Basin. The maxima observed in the beam attenuation profile of station
9 (bottom right box in figure 8) are all associated with either a maximum of fluorescence or the
Cariaco Basin’s pycnocline. During March 2004, the INL’s were located at a shallower depth
than in September 2003 (~ 70 m and 180 m; upper right box in Figure 8). Once more, these
layers were not associated with any density changes or phytoplankton maxima. During March
2004, waters in the Basin were well mixed, and typical beam attenuation profiles were similar to
that of station 9, depicted in the lower right box in Figure 8: high attenuation at the surface, due
to the increased productivity, but low attenuation below 50 m (below the pycnocline). [I feel
dizzy when reading this paragraph – what do you want to say?]
For September 2003,  was calculated for the INL’s.  is related to particle size
distribution, and as particles become larger,  decreases according mainly to the size of the
particles (Boss et al., 2001).  at the INL’s decreased, indicating the INL’s were probably
composed of larger particles than their surroundings, perhaps aggregates. [is this important?
Delete if not]
[you don’t need to cite others only to prove your values make sense]
4. Discussion
7
The two main contributors of CDOM, fresh water, and particles to the eastern Cariaco
Basin during September 2003 were the Neverí and Unare rivers. Their signatures were reflected
in all parameters measured, including salinity, CDOM absorption coefficient, and beam
attenuation coefficient (a measure of particle concentration). In contrast, the plume of the
Manzanares River was less apparent, and no evidence was observed to show the intrusion from
the much larger rivers (Orinoco and Amazon Rivers) from the east.
[In “discussion,” you don’t need to summarize what you found (that should be in the
conclusion). Rather, you need to discuss the implications or importance of your findings, and
perhaps argue with yourself why these results are important]
[if this is already documented, what is the purpose of this paper?] Thunell et al. (2000),
Goni et al. (2003) and Goñi et al. (2004) have reported terrigenous material in sediment traps
located near the CARIACO times-series station [give lat lon], likely from local rivers. However,
sediments discharged by the local rivers in September 2003 did not reach that far. The material
observed in the traps was probably transported at intermediate depths, in layers breaking off the
bottom of the shelf [what does this mean]. These layers at intermediate depths (150-300 m) could
constitute an important pathway of coastal sediment transport to the central Cariaco Basin. Such
layers were observed at 10o29’N 64o21’W over a submarine canyon during both the September
2003 and March 2004 surveys. Hickey et al. (1986) determined that the dominant mode of
sediment transport off the shelf in Quinault Canyon was by episodic formation of turbidity layers
at intermediate depths caused by horizontal advection. Puig et al. (2003) also observed such a
persistent layer, which contributed to the off-shelf sediment transport in Eel Canyon. Puig and
Palanques (1998) suggested that in canyons located on continental margins, such as the Foix
Canyon, sediment transport could be dominated by intermediate nepheloid layer detachments
and internal waves.
[these are pure speculations]
[you have nothing to support this]
[this appears very abrupt]
[Now, my question is, is the year of 2003 a “typical” year? i.e., can the wet season of 2003 represent a
“typical” case? To answer this you need to examine rainfall data. Without answering this it is hard to
draw any conclusions]
5. Conclusions
During the wet season (September 2003), the local rivers that most significantly affected
the bio-optical properties and particle distributions in the Eastern Cariaco Basin are the Neverí
and Unare rivers. No evidence of larger South American river (e.g., Amazon or Orinoco)
intrusion was observed in September 2003 or March 2004. The influence of local rivers was
mainly restricted within a narrow (30-40 km) band along the southern coast of the Cariaco Basin,
where all bio-optical parameters showed some anomaly. Salinity was between ?-? versus ? in
offshore waters; beam attenuation coefficient at 660 nm (cp660, a measure of particle
concentration) was between ?-? versus ? in offshore waters; CDOM absorption coefficient at 440
nm (ag(440), m-1) contributed to 70% of the total absorption coefficient at 440 nm and the
spectral slope is lower [give range] than that from offshore [give range].
During September 2003, the Unare River discharged more suspended particulate matter
than did the Neverí River. cp660 decreased rapidly from 0.6 m-1 at a station 10 km from the river
mouth to 0.4 m-1 10 km further offshore, indicating particle sinking to the bottom Indeed, the
8
bottom nepheloid layers, visible throughout the Unare Platform [where is this platform?], were
strongest near river mouths. The radius of the influence of the Unare’s BNL was approximately
40 [check this number] km from the mouth of the river, while the Neverí’s was close to 23 km.
The coast waters off the Manzanares River did not show any anomaly in the surface salinity,
CDOM, or suspended matter. However, internal [internal?] nepheloid layers were observed near
the Manzanares River, over a submarine canyon, both in September 2003 and March 2004. This
may be an important mechanism of sediment transport off the shelves and into the Eastern Deep
[what is this eastern deep?].
[this is nearly nonsense. Any river can discharge materials into adjacent ocean – you
don’t need to conduct two cruises to prove this. You also said in the conclusion of their
influence, so no need to repeat in a qualitative tone here]. [this is also nearly nonsense]
Clearly, the eastern Caribbean basin is under influence of local rivers. Although their
surface signatures in salinity and CDOM are restricted to…, the bottom nepheloid layers can
extend further offshore, up to 50 km from the coast. The CARIACO time-series station, where
monthly biogeochemical data have been collected since 1996, was (or was not) under the
influence of these local rivers. [if yes, so what?] [if no, may add the following: However, the
data were collected from two cruises only. More oceanographic surveys are needed to further
understand the connections between local rivers and the CARIACO station].
9